Вы находитесь на странице: 1из 347

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/258834538

Different ladder, different story? Dissecting the talent management paradox


within the framework of the postmodern career

Thesis · January 2009

CITATIONS READS

3 249

1 author:

Nicky Dries
KU Leuven
100 PUBLICATIONS   1,572 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

5C (Cross-Cultural Collaboration on Contemporary Careers) View project

The Future of Work and Organizational Psychology View project

All content following this page was uploaded by Nicky Dries on 27 May 2014.

The user has requested enhancement of the downloaded file.


Young people today are increasingly for instance, people tend to look up
focused on themselves, on their to high-potential co-workers see

Different ladder, different story?


work-life balance. They are less
prepared to make sacrifices even Different them as role models rather than
be jealous of them. It is standard
so, people are still very sensitive procedure to start discriminating

Different ladder, different story?


Dissecting the talent management paradox within the
framework of the postmodern career
about being promoted or not In this
l a d d e r , between employees some years after
kind of organization, when you’re entry, and everyone is aware of this

Putting the literature on talent management alongside postmodern


not promoted or you don’t get a pay
rise, it means they probably think
different There’s really no use being secretive
about who’s on the list and who isn’t.
careers literature, a twofold ‘talent management paradox’ seems to arise. you’re not that good at what you s t o r y ? People will talk to each other anyway
First of all, talent management is simultaneously depicted as utterly do lateral moves can definitely be they’ll figure it out themselves. A
outdated, and as more pivotal than ever for the competitive advantage of challenging. It takes up a lot of time decade ago, about 10 to 15 percent
D i s s e c t i n g
organizations. Second, talent management practices seem to be aimed at and effort to take on this whole new of our employees were considered
retaining exactly those employees who are most likely to leave. role in the organization and I’m sure t h e high potentials. There were more
it’s gratifying when you succeed in opportunities to climb the ladder
The current dissertation presents four chapters covering eight papers,
a transition like that But you don’t t a l e n t back then and we figured, if some of
each approaching the topic of talent management within the context
really feel successful unless you get them didn’t meet expectations we
of the postmodern career from a different vantage point. It aims to
a real promotion or a rise and then
management could always take them off the list.
contribute to the field in a threefold manner: first, by integrating the
careers literature with the talent management literature, thus creating
there’s the fact that a lot of people
paradox But today we’re much stricter about
are involved in project work all the who gets on the list in the first place
a comprehensive framework; second, by building on empirical data time, so every time you get started it’s about 2 percent worldwide. For
collected from different talent management stakeholders (i.e. general on a new project, that’s already a w i t h i n t h e instance, we’ll probably only pay
managers, HRM professionals, and individual employees identified as lateral move So it doesn’t really mean for an expensive training program
high potentials); and third, by examining cross-level effects, in that anything unless it’s an upward move.
framework when a high potential is asking.

Nicky Dries
organization-level variables (i.e. the classification of employees into It’s dangerous when the system is all o f t h e Talented individuals can get stuck
different ‘talent categories’) are related to individual-level variables (i.e. about status. Then people start to
postmodern in the leadership pipeline when
career antecedents and outcomes). think that you have to be on the list to organizations do not succeed

career
be somebody The problem with that in developing them properly.
kind of mindset is that high potentials Organizations usually have a two
Nicky Dries

usually make up only a really to three-year vision, but then their


www.vubpress.be small portion of an organization’s development activities reach a
population, so that would imply plateau Young people today are
that everyone else is frustrated. All increasingly focused on themselves,
N i c k y D r i e s
employees want to see extra effort on their work-life balance. They are
translated into extra rewards and Advisor: less prepared to make sacrifices even
9 789054 876700 recognition but when communication Prof. Dr. Roland so, people are still very sensitive

ISBN 978 90 5487 670 0 [about workforce segmentation] is Pepermans about being promoted or not In this
too explicit, resentment between kind of organization, when you’re
colleagues will probably arise. This not promoted or you don’t get a pay
Dissertation submitted to obtain the degree of Doctor in Psychological Sciences
applies especially to the Belgian rise, it means they probably think
Financial support: Research Foundation – Flanders (FWO TM490)
culture. I get the impression that you’re not that good at what you
Vrije Universiteit Brussel, 2009
in America for instance, people do lateral moves can definitely be
tend to look up to high-potential challenging. It takes up a lot of time
 
 
 
 
Different ladder, different story?  
Dissecting the talent management paradox 
within the framework of the postmodern 
career 
 
Nicky Dries 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Print: DCL Print & Sign, Zelzate 
 
© 2009 Nicky Dries 
 
2009 Uitgeverij VUBPRESS Brussels University Press 
VUBPRESS is an imprint of ASP nv  
(Academic and Scientific Publishers nv) 
Ravensteingalerij 28 
B‐1000 Brussels 
Tel. ++32 (0)2 289 26 50 
Fax ++32 (0)2 289 26 59 
E‐mail: info@vubpress.be 
www.vubpress.be 
 
ISBN 978 90 5487 680 9 
Legal Deposit D/2009/11.161/135 
 
All rights reserved. No parts of this book may be reproduced or transmitted in any form 
or by any means, electronic, mechanical, photocopying, recording, or otherwise, without 
the prior written permission of the editor. 
 
Dissertation submitted in order to obtain the degree of Doctor in Psychological Sciences 
Advisor: Prof. Dr. Roland Pepermans 
Financial support: Research Foundation – Flanders (FWO TM490) 
Vrije Universiteit Brussel, 2009 
 
 
 
 
Dissertation committee 
 
Supervisor: 
 
Prof. Dr. Roland Pepermans  
Vrije Universiteit Brussel 
Belgium 
 
Exam committee (in alphabetical order): 
 
Prof. Dr. John Arnold 
Loughborough University 
United Kingdom 
 
Prof. Dr. Geert Devos 
Vrije Universiteit Brussel 
Belgium 
 
Prof. Dr. Marie‐Anne Guerry 
Vrije Universiteit Brussel 
Belgium 
 
Prof. Dr. Luc Sels 
Katholieke Universiteit Leuven 
Belgium 
 
Prof. Dr. Raoul Van Esbroeck 
Vrije Universiteit Brussel 
Belgium 
 
Prof. Dr. Claartje Vinkenburg 
Vrije Universiteit Amsterdam 
The Netherlands 
 
 
 
 
 
 
 
Thank you! 
 
Over  the  course  of  the  last  four  years,  many  fabulous  people  have 
crossed my path, each of them contributing to my quality of life and the 
dissertation  process  I  was  going  though  (two  distinct,  but  related 
constructs) in a different way. To all of these people, I would like to say: 
Thank you! You’re the best! I couldn’t have done it without you!  
 
For  taking  me  on  as  a  PhD  student  and  supporting  and  guiding  me  all 
the way through the dissertation process: Thank you, Roland! 
 
For  all  of  the  social  support,  good  times,  and  the  great  atmosphere  at 
work  (not  to  mention  the  many  conversations  about  research  and 
methodology): Thank you to all of my colleagues and former colleagues, 
Brigitte,  Ellen,  Frederik,  Jan,  Jemima,  Jeroen,  Joeri,  Olivier,  Peter,  Rein, 
Sara, and Tim!  
 
For all of the social support outside of the workplace: Thank you to my 
dear  friends  and  family!  (And  a  special  thank  you  to  Jeroen  for  all  the 
Luv.)   
 
And finally, thank you to all members of the academic community who 
provided me with great input, feedback and ideas; and thank you to all 
organizations who agreed to participate in the research! 
 
 
 
 
 
 
 
 
Table of contents 
 
 
Introduction  p. 1 
   
Chapter I. Exploring ‘real’ high potential careers  p. 37 
Paper 1. ‘Real’ high potential careers: An empirical study into the perspectives  p. 39 
of organizations and high potentials 
   
Chapter II. High potential identification  p. 71 
Paper 2. Identification of leadership potential: Is there consensus about ‘the’  p. 73 
criteria? 
Paper 3. Using emotional intelligence to identify high potentials: A  p. 107 
metacompetency perspective 
Paper 4. High potential identification: Examining the developmental  p. 139 
perspective 
   
Chapter III. Career outcomes in high potential careers  p. 163 
Paper 5. The role of employability and firm‐specific capital in shaping the  p. 165 
psychological contract: Do high potentials and long‐tenured employees get the 
better deal? 
Paper 6. Effects of the high potential label on performance, career success and  p. 199 
commitment: A matter of communication? 
   
Chapter IV. Constructing ‘career’  p. 237 
Paper 7. Exploring four generations’ beliefs about career: Is ‘satisfied’ the new  p. 239 
‘successful’? 
Paper 8. Career success: Constructing a multidimensional model p. 267 
   
Discussion  p. 299 
   
 
 
 
 
1
 
 
 
 
Introduction 
 
 
“It’s the talent, stupid!”. With this remarkable headline Buckingham and 
Vosburgh  (2001)  point  out  talent  management  as  the  number  one 
concern  for  early  21st  century  human  resource  management 
professionals around the world.  
 
Despite  the  ever‐increasing  popularity  of  talent  management  in  the 
management  practitioner  literature,  there  has  been  strikingly  little 
empirical  research  on  the  topic  (Collings  &  Mellahi,  2009).  On  the  one 
hand,  we  find  publications  adopting  a  strongly  normative  approach. 
Such  articles  typically  list  a  set  of  best  practices  prescribed  by 
management experts, or even, ‘gurus’ (Vinkenburg & Pepermans, 2005). 
On  the  other  hand,  we  find  literature  about  single‐organization  cases 
from  practice  that  are  presented  as  success  stories  (e.g.  Remdisch  & 
Dionisius,  1998).  Either  way,  neither  one  of  these  approaches  is 
particularly  well  suited  to  capture  the  complex  realities  faced  by 
organizations (Boudreau & Ramstad, 2005). 
 
There  are,  nonetheless,  some  streams  of  academic  literature  that  may 
provide  useful  input  for  scholars  interested  in  the  topic  of  talent 
management. The careers literature, in particular, covers a large volume 
of  interesting  research  on  careers  taking  place  both  inside  and  outside 
of  organizational  career  management  structures  (e.g.  Hall,  2002; 
Rosenbaum,  1979).  However,  a  paradox  becomes  apparent  when 
comparing  the  management  literature  to  the  careers  literature  as 
concerns  their  views  on  talent  management.  Where  the  management 
literature  identifies  talent  management  as  “a  strategic  imperative” 
(Ashton  &  Morton,  2005,  p.  28),  the  careers  literature,  simultaneously, 
refers to talent management as “at best an anachronism, and at worst a 
false  promise  used  to  keep  valuable  employees  in  organizations” 
(Baruch  &  Peiperl,  1997,  p.  356).  So  which  statement  is  closer  to  the 
truth?  Should  talent  management  be  buried  alongside  the  traditional‐
organizational  career,  which,  according  to  some  voices  in  the  careers 
literature  is  ‘dead’  (Hall,  1996)?  Or  can  talent  management  (still)  offer 
2
added value to organizations and individual career actors alike, even in 
today’s ‘postmodern’ career context?   
 
In  this  dissertation,  we  attempt  to  bridge  the  gap  between  these  two 
perspectives  and  integrate  the  talent  management  literature  with  the 
postmodern  careers  literature.  Eight  different  papers  are  presented 
consecutively.  Six  of  them  report  empirical  studies  directly  concerned 
with  the  topic  of  talent  management,  set  within  in  the  postmodern 
career context. The last two papers focus on the meanings ‘career’ and 
‘career  success’  take  on  in  the  perceptions  of  individual  career  actors. 
Although only indirectly linked to talent management, these two papers 
are valuable contributions to this dissertation in that they explicitly deal 
with the postmodern turn in the careers literature: 
 
Postmodernism, also referred to as post‐positivist or constructivist thought, 
emphasizes plurality of perspectives, contextual impacts, social constructions 
of  reality,  and  the  importance  of  the  meaning  individuals  give  to  their 
experiences (Hayes & Oppenheim, 1997, p. 24). 
 
The  first  six  papers  integrate  the  postmodern  careers  literature  as 
context, that is, as the framework within which the study results must be 
interpreted.  The  topicality  of  the  papers  varies  in  that  some  of  them 
focus more on the organizational level of talent management (e.g. “based 
on  which  criteria  are  talented  employees  typically  identified  within 
organizations?”),  whereas  others  focus  more  on  the  interactions  of 
organizational‐level  variables  with  individual  outcomes  (e.g.  “what  are 
the  effects  of  receiving  the  ‘high  potential’  label  on  organizational  and 
career  commitment?”).  Each  of  the  eight  studies  included  in  this 
dissertation links in to the overall research framework presented at the 
end of this Introduction in Figure I.   
 
This Introduction starts off with an overview of the historical evolutions 
that have taken place in the ‘world of work’ and, more specifically, how 
these  evolutions  have  affected  career  theory  and  the  definition  of 
‘career’.  We  then  move  on  to  include  the  talent  management  literature 
by  discussing  the  apparent  ‘talent  management  paradox’  between  the 
careers  literature  and  the  management  literature.  We  continue  by 
describing some of the issues faced by talent management literature and 
practice  in  defining  what  ‘talent  management’  means  and  how  ‘high 
potentials’ should be identified and retained. We conclude by identifying 
a  research  gap  from  all  of  the  above  and  outlining  the  structure  of  the 
current dissertation. At the end of this Introduction, we inserted a table 
(Table  I)  summarizing  the  research  questions  and  methodologies  for 
each of the eight papers presented in this dissertation.   
3
Setting the framework for the ‘postmodern’ career 
 
Historical evolutions 
 
The  historical  evolution  of  the  global  economy,  from  being  centered 
mostly  around  agriculture  to  the  postmodern  information  era,  has 
strongly  shaped  the  framework  and  the  boundaries  within  which 
individual careers can be enacted today.   
 
Around the onset of the 19th  century, the  industrial revolution marked 
the  end  of  the  agricultural  economy,  in  which  the  dominant  social 
institution  was  the  family  and  young  people  simply  inherited  their 
parents’  occupations    (Savickas,  2000).  The  dawning  of  the  industrial 
economy was characterized by the appearance of large, bureaucratically 
structured  organizations  providing  careers  for  life.  Job  security  was  all 
but  guaranteed  to  employees,  who  reciprocated  by  offering  their 
employers their loyalty and dedication. Since the typical organizational 
structure  was  hierarchical,  ‘career’  implied  vertical  movement,  and 
career  success  was  defined  by  upward  advancement  on  the  corporate 
ladder (Savickas, 2000; Savickas et al., 2009; Van Esbroeck, 2008).  Even 
today (and problematically so), the notion of hierarchical advancement 
within an organization remains associated with career success, although 
the  organizational  structures  at  the  origin  of  this  association  have 
changed  considerably  (Arnold  &  Cohen,  2007;  Miles  &  Snow,  1996; 
Sullivan, 1999).  
 
In the second half of the 20th century, society was transformed through 
globalization,  and  many  organizations  grew  into  multinational 
corporations.  Scientific  and  technical  evolutions  brought  societies 
worldwide  into  the  information  era.  The  postindustrial  economy, 
characterized by the declining importance  of manufacturing  relative to 
information  technology  and  knowledge  management,  was  a  fact  (Van 
Esbroeck,  2008).  As  a  result,  organizational  and  societal  structures 
changed  dramatically.  Economic  globalization  and  the  restructuring  of 
organizations  (e.g.  through  downsizing,  delayering,  outsourcing  and 
offshoring) have fundamentally altered the structure and nature of jobs 
and careers (Maranda & Comeau, 2000). 
 
As  many  organizations  have  been  ‘flattening’  their  hierarchical 
structures, the traditional premises upon which careers relied appear to 
be  fading.  Organizations  can  no  longer  promise  a  career  for  the  long 
term,  as  they  could  before  when  the  economy  was  more  stable  and 
predictable (Savickas, 2000). Careers in today’s postmodern society are 
4
no  longer  “logical,  stable,  depictable  and  predictable”  (Van  Esbroeck, 
Tibos  &  Zaman,  2005,  p.  6).  Careers  have  become  a  more  or  less 
unpredictable  series  of  small  steps  made  by  individuals  who  are 
continuously negotiating work and non‐work aspects of life throughout 
their  lifespan.  As  careers  are  no  longer  ‘owned’  by  organizations,  the 
responsibility  for  career  management  is  now  placed  primarily  in  the 
hands of the individual employee, who must develop transferable skills 
and  adaptive  strengths  to  cope  in  an  environment  without  definite 
securities  (Savickas,  2000).    Instead  of  being  depicted  as  a  ladder  (the 
typical  metaphor  for  steady  upward  movement),  career  can  now  be 
described  as  a  ‘lattice’  (among  other  metaphors,  see  Inkson,  2004), 
enabling  multiple  career  paths  and  possibilities  for  lateral  job 
enrichment, rather than  upward movement alone (Iles, 1997).  
 
Definitional evolutions 
 
In  sync  with  the  historical  evolutions  (re)shaping  the  nature  and 
structure of careers, the definition of ‘career’ has evolved, as well. As a 
result  of  widespread  organizational  restructuring  and  economic 
uncertainties since the late 1980s, many of the traditional assumptions 
about  careers  no  longer  seem  valid.  As  a  result,  it  has  been  said  that 
there is no longer a clear and consensual understanding of what career 
means,  both  for  individuals  and  organizations  (Adamson,  Doherty  & 
Viney, 1998).  
 
Below, we list some of the more established definitions of career found 
in the literature, organized from older to more recent definitions. Note 
that  Hughes  (1937)  offers  two  definition,  the  first  referring  to  the 
objective  career,  the  second  to  the  subjective  career.  Overall,  the 
definition  by  Arthur,  Hall  and  Lawrence  (1989)  appears  to  be  the 
definition of career most frequently cited in the careers literature today 
(Arnold & Cohen, 2007).  
 
The  moving  perspective  in  which  persons  orient  themselves  with  reference 
to the social order, and of the  typical sequences and concatenation  of  office 
(Hughes, 1937, p. 409).  
 
The  moving  perspective  in  which  the  person  sees  his  life  as  a  whole  and 
interprets  the  meaning  of  his various  attributes,  actions  and  the  things  that 
happen to him (Hughes, 1937, p. 413).  
 
A  succession  of  related  jobs,  arranged  in  a  hierarchy  of  prestige,  through 
which  persons  move  in  an  ordered  (more‐or‐less  predictable)  sequence 
(Wilensky, 1961, p. 523).  
 
5
The combination and sequence of roles played by a person during the course 
of a lifetime (Super, 1980, p. 282).  
 
The pattern of work‐related experiences that span the course of a person’s 
life (Greenhaus, 1987, p. 9). 
 
The  evolving  sequence  of  a  person’s  work  experiences  over  time  (Arthur et 
al., 1989, p. 8). 
 
A process of development of the employee along a path of experiences and 
jobs in one or more organizations (Baruch & Rosenstein, 1992, p. 478). 
 
The individual’s development in learning and work throughout life (Collin & 
Watts, 1996, p. 393). 
 
Several  communalities  and  differences  can  be  identified  among  these 
definitions.  First,  all  of  these  definitions  refer  to  evolution  or 
advancement in one way or another. This fact is hardly surprising when 
we  consider  the  etymology  of  the  term  ‘career’.  The  word  stems  from 
the French ‘carrière’ which, historically, refers to ‘road’ or ‘race course’. 
Therefore,  inherently,  career  implies  following  a  route  which  has  both 
direction  and  purpose  (Dalton,  1989).  Consequently,  Adamson  et  al. 
(1998)  argue  that  “without  some  logical  sequence  in  the  ordering  of 
work  experiences  over  time,  the  career  journey  would  cease  to  have 
meaning” (p. 253).  
 
Second, while some definitions emphasize that careers take place in the 
work domain, others expand the career concept by relating it to the life 
domain as a whole. Viewing the concept of career as much broader than 
exclusively  work‐related  is  typically  advocated  by  sociologists 
(Adamson  et  al.,  1998).  The  Chicago  School  of  Sociology  for  instance, 
around  the  1920s,  studied  the  life  histories  of  their  local  communities. 
The  School  was  interested  in  social  ecology,  demography,  urbanization 
and social deviance, and their goal was to study how people constructed 
their  lives  (e.g.  Barley,  1989).  When  the  Chicago  sociologists  talked 
about  career,  then,  they  were  referring  to  a  concept  much  broader  in 
application  than  it  is  in  everyday  conversation  (Adamson  et  al.,  1998). 
One could assert, however, that by defining career so broadly it becomes 
almost  indistinguishable  from  the  concept  of  adult  development 
(Adamson et al., 1998). 
 
In  general,  the  meanings  attributed  to  career  differ  somewhat  across 
academic  disciplines.  Reviews  of  career  theory  seem  to  imply  that  the 
field  is  composed  mainly  out  of  psychological  and  sociological  views 
(Arthur  et  al.,  1989).  Psychological  theory  focuses  mostly  on  static 
dispositional differences and their implications on career outcomes, and 
6
on  career  stage  dynamics  influencing  occupational  choice  and  career 
development.  In  addition,  it  focuses  on  how  careers  can  contribute  to 
personal  growth,  and  how  personal  growth  can  in  turn  contribute  to 
organizations and societies (Arthur et al., 1989). Sociological theory, on 
the other hand, is typically concerned with social class determinants of 
career.  It  studies  the  tension  between  voluntarism  (i.e.  choices 
determined by individual autonomy and free will) versus structure (i.e. 
choices  determined  by  constraints,  norms,  and  sanctions)  and 
adaptation (i.e. changing people so that they can integrate into society) 
versus  transformation  (i.e.  changing  society  in  order  to  integrate 
people)  (Maranda  &  Comeau,  2000).  Arthur  et  al.  (1989,  p.  10)  outline 
some more viewpoints on career, grounded in yet other disciplines (e.g. 
economics,  political  science,  anthropology).  We  will  not  delve  further 
into  these  other  disciplines,  however,  as  this  would  lead  us  too  far. 
Either way, an important aspect of the careers literature is that it draws 
from  different  disciplines.  Disciplinary  boundaries  are  not  seen  as 
constraints;  rather,  interdisciplinary  research  is  encouraged  (e.g. 
Arthur,  2008).  Throughout  this  dissertation,  we  will  use  literature 
stemming  from  psychology,  sociology,  economy  and  management. 
However,  as  our  primary  discipline  is  work  and  organizational 
psychology,  it  will  predominantly  focus  on  the  concerns  and  wellbeing 
of  the  individual  employee  (rather  than,  for  instance,  on  the  tension 
between structure and agency).  
 
A  third  conceptual  difference  among  the  definitions  of  career  cited 
above is that some of them refer to ‘jobs’ and ‘roles’, whereas others talk 
about  ‘experiences’.  These  choices  of  words  indicate  whether  career  is 
considered  from  an  objective  or  a  subjective  lens.  One  of  the  most 
noteworthy  contributions  of  the  postmodern  careers  literature  is  the 
recognition of the importance of the subjective career, that is, the sense 
individuals  make  of  their  personal  career  histories  (Hall  &  Chandler, 
2005).  Now  that  the  assumptions  of  career  are  becoming  increasingly 
ambiguous, individuals’ personal sensemaking   processes are moving to 
the  forefront  (Arnold  &  Jackson,  1997).  Whereas  the  objective  face  of 
career is  generally concerned with  observable career  attainments such 
as  pay,  promotions  and  functional  level  (Nicholson,  2000),  the 
subjective  career  refers  to  an  individual’s  idiosyncratic  perceptions  of 
his  or  her  own  career  and  the  resulting  feelings  of  satisfaction  or 
dissatisfaction  (Greenhaus,  Parasuraman  &  Wormley,  1990).  Although 
virtually  all  career  scholars  acknowledge  the  distinction  between  the 
objective  and  the  subjective  career,  Hall  and  Chandler  (2005),  among 
others, stress that careers are two‐sided and that both perspectives add 
value to our understanding of careers. One of the main contributions of 
7
adopting  a  dual  viewpoint  of  career  is  that  it  helps  researchers  avoid 
‘reification’ – the fallacy of treating dynamic social constructs like career 
as  if  they  were  real,  material,  verifiable  entities  (Evetts,  1992). 
Traditional, generally positivistic, career theory may not (or no longer) 
be  appropriate  to  grasp  the  complex  and  dynamic  realities  of 
postmodern‐day careers (Savickas, 2000).  
 
Theoretical evolutions 
 
Although career theory has notably evolved over the last four decades, 
several authors claim that it is still in a state of “developmental infancy” 
(Patton  &  McMahon,  2006,  p.  7).  Career  theory  has  been  criticized  for 
lacking  in  comprehensiveness  and  coherence  (Brown,  2002;  Savickas, 
2000),  and  for  neglecting  contextual  issues  while  emphasizing 
individual  variables  (Brown,  2002).  Also,  career  theory  has  been 
described  as  overly  segmented,  both  when  considering  individual 
models and the disciplinary field as a whole (Arthur et al., 1989; Arthur, 
2008).  
 
Career theories can be categorized according to several parameters (for 
an overview, see Patton & McMahon, 2006, p. 10). Patton and McMahon 
(2006), in their review of the literature, utilize ‘content’ and ‘process’ as 
their vantage point for categorizing career theories.  
 
Theories that focus on the content of career choice are typically focused 
on  characteristics  intrinsic  to  the  individual  or  emanating  from  the 
context  within  which  the  individual  functions  (Patton  &  McMahon, 
2006).  Trait  and  factor  theory  (e.g.  Parsons,  1909;  Holland,  1973)  is  a 
prominent example of a ‘content’ theory. Career choice, according to this 
theory, involves matching individuals to jobs with the goal of satisfying 
their  needs  and  achieving  satisfactory  performance  levels.  Over  time, 
there has been an evolution from the rather static approach of trait and 
factor theory, where a person is matched to an occupation, to the more 
dynamic  approach  of  person‐environment  fit  (e.g.  Dawis  &  Lofquist, 
1984).  The  latter  theory  implies  ongoing  reciprocal  adjustment  as 
environments  are  influenced  by  individuals  and  vice  versa  (Patton  & 
McMahon, 2006). 
 
‘Process’ theories (also known as developmental theories), on the other 
hand,  emphasize  career  stages  and  change  over  time  (Patton  & 
McMahon, 2006). A particularly well‐known process theory is life‐span 
life‐space  theory  (Super,  1953).  This  theory  stresses  the  importance  of 
also  attending  to  non‐work  variables  in  studies  of  career.  Work  roles, 
8
family  roles,  educational  roles  and  community  roles  all  interact  in 
shaping individual careers; furthermore, the significance of each role for 
a  certain  individual  depends  on  his  or  her  personal  makeup  and  the 
developmental stage the person is in (Herr, 1997). 
 
More  recent  theoretical  work  has  attempted  to  take  both  content  and 
process  variables  into  account,  as  well  as  the  role  of  cognition  in  the 
process    (Patton  &  McMahon,  2006).  Systems  theory,  for  example, 
incorporates  individual  systems  as  well  as  higher‐order  systems  into 
one  model  of  career  (e.g.  McMahon  &  Patton,  1995).  At  the  individual 
level,  variables  such  as  gender,  age,  interest,  abilities,  and  personality 
influence  career  outcomes.  At  the  macro  level,  career  is  shaped  by  the 
family, the community, labor  market trends, geographical location, and 
other  contextual  variables.  The  interaction  between  these  different 
systems is characterized by recursiveness, ongoing change, and chance 
factors. An essential feature is that changes in any one part of the system 
impact the system as a whole (Amundson, 2005). 
 
The  current  careers  literature  identifies  two  main  avenues  for 
advancing career theory: first, allotting a more central role to the notion 
of constructivism, and second, striving for convergence and theoretical 
integration of the many different career theories (Patton, 2008). As for 
the challenge of theory integration, some career scholars have come to 
the  conclusion  that  the  complexities  that  occur  within  and  between 
career  actors’  traits  and  environments  are  simply  too  complicated  to 
capture in theoretical models and that therefore, we should stop trying 
to do so, and look at careers at the individual level alone (Brown, 2002). 
This fits in perfectly with the notion of constructivism, which states that 
reality can only be understood as a subjective experience (e.g. Hayes & 
Oppenheim, 1997). Constructivism represents an epistemological stance 
that  distances  itself  from  ‘one  true  career  reality’  and  instead, 
emphasizes  the  self‐organizing  and  self‐managing  position  of  the 
individuals having the careers. Thus, the constructivist view of career 
reflects  perfectly  the  historical  and  definitional  evolutions  that  have 
been  taking  place  over  the  last  decades  (Savickas,  2000).  Despite 
Brown’s  (2002)  statement  that  the  divide  between  positivistic  and 
constructivist  approaches  to  career  is  causing  “convergence  among 
theories  and  the  development  of  an  integrative  theory  to  be  less  likely 
today  than  ever”  (p.  15),  other  authors  (e.g.  Patton  &  McMahon,  2006) 
are  convinced  that  this  gap  can  be  bridged,  and  that  both  perspectives 
can contribute to a holistic understanding of career.  
 
 
9
The talent management paradox  
 
Now that we have demarcated the general evolutions that have shaped 
the context within which careers are enacted, we move one step further 
to  include  the  (limited)  literature  on  talent  management  into  our 
argumentation.  We  will  go  much  deeper  into  the  various  definitions 
ascribed to talent management later on in this Introduction, but for now, 
we adopt the following working definition of talent management:  
 
Talent management is the systematic attraction, identification, development, 
engagement/retention  and  deployment  of  those  individuals  who  are  of 
particular value to an organization, either in view of their ‘high potential’ for 
the future or because they are fulfilling business/operation‐critical roles (UK 
Chartered Institute of Personnel and Development, 2009, p. 2).  
 
Putting  the  literature  on  talent  management  alongside  postmodern 
careers  literature,  a  twofold  ‘talent  management  paradox’  seems  to 
arise.  First  of  all,  talent  management  is  simultaneously  depicted  as 
utterly outdated (in the careers literature, e.g. Baruch & Peiperl, 1997), 
and  as  more  pivotal  than  ever  for  the  competitive  advantage  of 
organizations  (in  the  management  literature,  e.g.  Buckingham  & 
Vosburgh,  2001).  Second,  talent  management  practices  seem  to  be 
aimed at retaining exactly those employees who are most likely to leave 
(e.g. Tulgan, 2001).  
 
Postmodern careers literature, which commonly advocates the ‘death’ of 
the  traditional‐organizational  career,  describes  talent  management  as 
“at  best  an  anachronism,  and  at  worst  a  false  promise  used  to  keep 
valuable  employees in organizations” (Baruch &  Peiperl, 1997, p. 356). 
The  authors  argue  that  organizations  operating  within  the  current 
turbulent  economic  environment  can  no  longer  promise  long‐term 
employment  to  employees,  let  alone  rapid  upward  progression  on  the 
corporate  ladder.  This  means  that  the  very  principle  underlying  talent 
management programs – that hard work and the display of exceptional 
talent will be rewarded by internal career opportunities – is threatened 
(Tulgan, 2001).  
  
At  the  same  time,  however,  the  management  literature  refers  to  the 
second  ‘war  for  talent’  as  the  critical  human  resource  management 
(HRM) challenge of the  early  21st century (Michaels, Handfield‐Jones  & 
Axelrod, 2001). The war for talent is believed to be the result of mainly 
demographic and psychological contract drivers. As for the demographic 
situation, structural shortages are starting to surface in the labor market 
due  to  reduced  birthrates,  a  larger  relative  percentage  of  older 
10
employees,  and  massive  retirements  (e.g.  Guthridge,  Komm  &  Lawson, 
2008;  Sels,  Van  Woensel  &  Herremans,  2008).  With  respect  to  the 
psychological  contract  between  employers  and  employees  (i.e. 
“individuals’ beliefs regarding the terms and conditions of the reciprocal 
exchange  agreement  between  themselves  and  their  employers”, 
Rousseau, 1989, p. 123), it has been said that it is now characterized by 
decreased employee loyalty and increased  employee cynicism, brought 
about  by  the  fact  that  job  security  is  no  longer  available  to  many 
employees (e.g. Sullivan, 1999).  
 
One  possible  way  to  interpret  the  talent  management  paradox  is  by 
assuming  that  talent  management  practice  is  (hopelessly)  running 
behind  (Baruch  &  Peiperl,  1997).  However,  despite  the  fact  that  both 
talent management and ‘the career’ have been declared dead repeatedly 
over  the  last  few  decades,  a  volume  of  research  indicates  that  claims 
about  the  shift  from  traditional‐organizational  to  more  ‘boundaryless’ 
career  types  have  to  be  put  into  perspective  (e.g.  Granrose  &  Baccili, 
2006; Guest & Mackenzie Davey, 1996; Sullivan, 1999; Walton & Mallon, 
2004). 
 
A  second  take  on  the  talent  management  paradox  involves  assuming 
that traditional‐organizational careers can, in fact, still exist, but only for 
‘privileged’  groups  such  as  employees  identified  as  high  potentials.  We 
opted for the following working definition of ‘high potentials’:  
 
Those individuals within an organization who are recognized, at that point in 
time, as the organization’s likely future leaders (Cope, 1998, p. 15).  
 
However, it is quite plausible that exactly those employees who are still 
in a position to receive internal career benefits such as job security and 
upward  advancement  opportunities  are  also  those  who  are  most  likely 
to thrive in the postmodern career landscape (e.g. because they have the 
highest levels of employability), and are thus most likely to leave (Dyer 
& Humphries, 2002; Tulgan, 2001).  
 
In what follows, we will dig deeper into the different aspects composing 
the  talent  management  paradox  outlined  here.  We  start  out  with  an 
overview of the arguments raised in the postmodern careers literature 
as  to  the  death  of  the  traditional‐organizational  career,  which  we 
counterbalance  by  reviewing  several  lines  of  critique  on  these 
arguments.  We  then  continue  with  an  overview  of  the  literature  on 
talent  management  and  high  potentials,  and  identify  from  all  of  the 
above the research gap that sparked this dissertation.  
 
11
The career is dead, long live the career 
 
“The  career  is  dead,  long  live  the  career!”.  The  title  of  this  renowned 
book  (Hall,  1996)  accurately  captures  the  state  careers  research  is  in 
today.  The  postmodern  careers  literature  appears  to  actively  sponsor 
the  idea  that  while  the  traditional‐organizational  career  is  dead,  the 
‘boundaryless’ career is alive and flourishing (Guest & Mackenzie Davey, 
1996).  The  boundaryless  career,  rather  than  representing  any  single 
form  of  career,  encompasses  “a  range  of  possible  forms  that  defy 
traditional  employment  assumptions”  (Arthur  &  Rousseau,  1996,  p.  6). 
As  careers  researchers  are  uncovering  more  and  more  ‘new’  career 
patterns,  their  vocabulary  and  lexicon  for  describing  these  patterns  is 
evolving  accordingly  (Dickmann,  Brewster  &  Sparrow,  1999).  Apart 
from  boundaryless  careers  (which  seems  to  be  the  most  popular 
concept), there have also been accounts of ‘kaleidoscope’ (e.g. Mainiero 
&  Sullivan,  2005),  ‘postindustrial’  (e.g.  Gershuny,  1993)  and  ‘protean’ 
careers (e.g. Hall, 1976).  
 
The  term  ‘boundaryless  career’  was  used  for  the  first  time  somewhere 
in  the  early  1990s,  and  was  the  theme  of  the  1993  Academy  Of 
Management  conference.  Although,  in  empirical  papers,  boundaryless 
careers  are  often  operationalized  as  careers  characterized  by  inter‐
organizational  mobility  at  one  point  or  another,  its  original  definition 
was meant to be much broader. Arthur and Rousseau (1996) described 
six different but interrelated meanings of the boundaryless career. First, 
boundaryless  careers  are  careers  that  transcend  the  boundaries  of 
different employers. Second, boundaryless careers are careers that draw 
validity  and  marketability  from  outside  the  present  employing 
organization. Third, boundaryless careers are careers that are sustained 
and  supported  by  external  networks.  Fourth,  boundaryless  careers  are 
careers  that  challenge  traditional  assumptions  about  career 
advancement  and  movement  up  through  an  organizational  hierarchy. 
Fifth,  boundaryless  careers  are  careers  in  which  individuals  reject 
opportunities  for  advancement  in  favor  of  personal  or  family  reasons. 
And sixth and final, boundaryless careers are careers that are based on 
the  actor’s  interpretation,  who  may  see  their  career  as  boundaryless 
regardless  of  contextual  constraints.  A  common  factor  among  these  six 
meanings  is  that  each  of  them  refers  to  the  weakening  ties  between 
employees and organizations (Arnold & Cohen, 2007).  
However  mainstream  the  notion  of  the  boundaryless  career  may  have 
become  in  the  postmodern  careers  literature,  signs  are  that  it  may  not 
be completely attuned to the experiences career scholars and actors are 
(still)  having  in  practice.  Below,  we  build  on  three  lines  of  critique 
12
formulated  against  the  proclaimed  pervasiveness  of  boundaryless 
careers in the contemporary career context.  
 
The career is dead… but not very much so. First, it is not at all clear 
whether the boundaryless career literature’s claims about the speed and 
inevitability  of  the  emergence  of  boundaryless  careers  are  justified 
(Guest  &  Mackenzie  Davey,  1996).  Although  there  is  little  discussion 
over  the  fact  that  opportunities  to  pursue  a  stable,  single‐employer 
career  are,  indeed,  in  decline  (Arnold  &  Jackson,  1997),  the  current 
literature  may  overestimate  the  changes  that  have  been  taking  place, 
particularly  about  the  capacity  of  individual  employees  to  enact  their 
careers  as  ‘free  agents’  (Tulgan,  2001).  The  promotion  of  ideals  that 
have either never existed, or are only achievable by a small segment of 
the total labor force, might prove hazardous to individual career actors  
and organizations alike (Evetts, 1992).  
 
Several  studies  conducted  in  countries  all  over  the  world  (e.g. 
Bouffartigue  &  Pochic,  2001;  Granrose  &  Baccili,  2006;  Jacoby,  1999; 
Lundberg  &  Peterson,  1994;  McDonald,  Brown  &  Bradley,  2005; 
Sullivan,  Carden  &  Martin,  1998)  have  concluded  that  “although  the 
boundaries  of  career  have  shifted,  they  have  not  melted  into  thin  air” 
(Walton  &  Mallon,  2004,  p.  77).  One  specific  reservation  about  the 
boundaryless career is the transferability of the concept beyond the US, 
the  cultural  importance  people  attach  to  security  and  the  extent  of 
employee  union  influences  in  the  labor  market  being  central  to  the 
discussion (Hirsch & Shanley, 1996; Meyer, 1995; Sullivan, 1999). 
 
Several authors have criticized the postmodern careers literature for its 
overly  simplistic  ‘dichotomization’  of  traditional‐organizational  careers 
versus  boundaryless  careers  (e.g.  Arnold  &  Cohen,  2007).  Career 
scholars  seem  to  have  the  tendency  to  portray  the  past  as  stable, 
predictable and secure, and the present as turbulent and ever‐changing, 
thereby  exaggerating  the  amount  of  change  that  has  taken  place  and 
forsaking  the  nuance  that  is  needed  to  study  the  complexities  of  the 
current‐day career reality (Arnold & Cohen, 2007; Cullinane & Dundon, 
2006;  Evetts,  1992).  Careers  are  not  actually  ‘boundaryless’.  Rather, 
some  of  the  boundaries  that  used  to  hinder  certain  forms  of  career 
mobility  have  become  more  permeable  (Sullivan,  1999).  Riley  (1995) 
fittingly sums up:  
 
If it is assumed that many career paths and skill accumulation patterns adapt 
and survive in a changing world, then this is as valuable a focus for analysis 
as are the much championed changing patterns. The case is basically that, if a 
true portrait of the labor market is the objective, the stable dynamics of labor 
markets are as important as the changing dynamics (p. 196).  
13
The  career  is  dead…  and  that  is  not  (necessarily)  a  good  thing.  A 
second  line  of  critique  asserts  that,  although  the  boundaryless  career 
literature  enthusiastically  promotes  the  benefits  of  boundaryless 
careers, its discourse is potentially harmful for weaker employee groups 
(e.g.  Dyer  &  Humphries,  2002;  Van  Buren,  2003).  According  to  Arthur 
and  Rousseau  (1996),  both  employees  and  employers  can  benefit 
equally from the new boundaryless career ‘reality’. Employers can gain 
from the flexibility that these new types of career bring about, as well as 
from  the  fact  that  their  employees  must  engage  in  continuous  (self‐
managed)  development  in  order  to  remain  employable.  Employees, 
from their side ‐ although they are now required to carry more personal 
responsibility  for  their  careers  –  have  been  empowered  to  pursue 
personal  fulfillment  without  having  to  adhere  to  the  rules  of 
organizations.  Furthermore,  Arthur  and  Rousseau  (1996)  suggest  that 
having a boundaryless career can increase personal earning potential.  
 
Despite  the  possible  benefits  accompanying  boundaryless  careers, 
downsides  can  be  identified  as  well.  Potential  downsides  include  the 
underemployment of workers, reductions in organizational training and 
development  initiatives,  and  the  lack  of  personal  identification  of 
employees  with  their  organizations  (Arnold  &  Cohen,  2007;  Hirsch  & 
Shanley,  1996;  Sullivan,  1999).  Especially  more  vulnerable  employee 
groups,  such  as  women,  ethnic  minorities,  and  poorly  educated  people 
have  experienced  disproportionate  declines  in  job  stability  (Sullivan, 
1999). Sennett (1998) asserts that “without clear paths, individuals are 
left  vulnerable  to  a  sense  of  aimlessness  which  constitutes  the  deepest 
sense  of  anxiety”  (p.  120).  This  statement,  however  dramatic,  is 
grounded  in  the  notion  that  in  general,  employees  prefer  secure 
employment  relationships  since  there  are  personal  and  financial  costs 
associated  with  changing  jobs  (Van  Buren,  2003).  For  organizations  as 
well, there are downsides to the boundaryless career. Fournier (1998), 
for  example,  states  that  it  encourages  a  ‘consumerist’  career  mentality, 
reducing  organizations  to  tools  merely  there  to  help  equip  employees 
with the resources they need to develop their personal projects.  
 
Gunz,  Evans  and  Jalland  (2000)  have  raised  the  critique  that  the 
boundaryless  careers  literature,  rather  than  offering  a  new  theoretical 
framework  for  studying  careers,  seems  to  present  normative 
prescriptions.  Furthermore,  its  discourse  has  been  said  to  serve  the 
needs  of  the  current‐day  ‘ruthless  economy’  in  that  it  enables 
organizations  to  get  rid  more  easily  of  as  many  costly  permanent 
workers  (and  their  benefits)  as  needed  (Richardson,  2000;  Van  Buren, 
2003).  Traditionally,  there  was  some  element  of  mutuality  and  risk‐
14
sharing  in  the  employment  relationship  between  employees  and 
employers. In the boundaryless career model, however, employees run 
the  risk  that  their  investments  in  firm‐specific  capital  will  not  be 
reciprocated  by  long‐term  internal  employment  prospects  (Van  Buren, 
2003).  Labeling  it  the  ‘new’  career  conveniently  implies  that  these 
changes  are  inevitable  and  have  already  taken  place  (Hallier  &  Butts, 
1999).  
 
The  career  is  dead…  but  not  for  everyone.  As  for  the  third  line  of 
critique,  although  the  boundaryless  career  literature  typically  assumes 
that “the best employees are thinking like free agents” (Tulgan, 2001, p. 
104),  the  opposite  seems  to  be  true.  Rather  than  thinking  like  free 
agents, ‘the best’ are the ones to whom organizations still offer internal 
career  opportunities  (which,  on  average,  they  gladly  accept),  such  as 
entering  in  a  high  potentials  program  (Viney,  Adamson  &  Doherty, 
1997). Indeed, several authors (Noon & Blyton, 2002; Richardson, 2000; 
Van  Buren,  2003)  have  observed  a  growing  ‘bifurcation’  of  the  labor 
market  between  those  who  are  in  a  position  to  benefit  from  the 
postmodern career environment and those who are not. It has been said 
that  the  boundaryless  career  concept  is  only  advantageous  for  those 
employees  who  were  formerly  also  privileged  in  traditional  career 
settings  (i.e.  highly  motivated,  high‐skilled  white‐collar  employees) 
(Dyer  &  Humphries,  2002).  A  two‐tiered  workforce  thus  arises.  The 
small top‐tier of highly sought‐after employees is in a strong bargaining 
position relative to their employers, which enables them to demand and 
receive  traditional  career  benefits  (‘winner‐take‐all’);  the  large  second 
tier  is  composed  of  employees  whose  skills  are  more  replaceable, 
rendering them ‘expendable’ (Van Buren, 2003).  
 
Research  on  the  psychological  contract  has,  indeed,  found  that 
organizations  engage  in  workforce  segmentation,  delineating  at  least 
two  types  of  employees  (i.e.  ‘core’  and  ‘periphery’  employees)  with 
differential  access  to  career  opportunities  (Rousseau  &  McLean  Parks, 
1993).  For  the  core  employees,  the  traditional‐organizational  career  is 
still intact and the hierarchy remains in place, albeit somewhat reduced 
(Guest & Mackenzie Davey, 1996). These core roles are reserved for the 
best people (e.g. those identified as high potentials). Viney et al. (1997) 
report a fitting quote from a HR manager:  
 
What we are trying to hang on to indeed is the concept that the high potential 
does still have a career. We are still recruiting a large number of people, we 
are  investing  in  their  training  and  development,  and  we  do  want  them  to 
hang on with us (p. 179).  
 
15
As  high  potentials  advance  through  the  organization’s  hierarchy,  they 
focus  their  talents  on  maintaining  and  promoting  the  organizational 
culture,  mission,  and  goals,  assigning  work,  and  holding  people 
accountable  for  results.  In  sum,  they  ensure  continuity  in  terms  of  the 
organization’s resources (Jerusalim & Hausdorf, 2007; Tulgan, 2003).  
 
The ‘war for talent’ 
 
Together  with  the  proclaimed  death  of  the  traditional‐organizational 
career, organizational talent management has been said to be outdated 
and  out  of  touch  with  the  current‐day  needs  of  organizations  and 
individual career actors (e.g. Baruch & Peiperl, 1997). However, there is 
a huge gap between the views put forward in  the careers literature with 
regard to talent management, and those put forward in the management 
literature. In fact, ever since a group of McKinsey consultants launched 
the expression ‘war for talent’ in 1997, the topic of talent management 
has  become  increasingly  popular  in  the  management  literature  around 
the world (Michaels et al., 2001).  
 
The war for talent seems to be grounded in two main assumptions. First, 
that the importance of human capital (especially in terms of leadership) 
for  organizations’  competitive  advantage  is  increasing.  Traditional 
sources of competitive advantage are losing their edge as organizations 
today  are  operating  within  the  context  of  a  knowledge  economy. 
Leadership, on the other hand, is a potentially renewable resource that 
cannot easily be copied or stolen by competitors. Furthermore, it is said 
that  only  through  effective  leadership  a  firm  can  identify  and  exploit 
other  sources  of  potential  competitive  advantage,  by  doing  so  staying 
ahead  of  competitors  in  a  rapidly  changing  economic  context  (McCall, 
1992).  
 
The second assumption is, that attracting and retaining talented human 
capital is becoming harder due demographic and psychological contract 
evolutions (Tucker, Kao & Verma, 2005).  
 
Demographic drivers. The majority of employers worldwide (both in 
emerging  economies  such  as  China  and  India  and  in  developed 
economies such as the United States and Western Europe) are reporting 
difficulties  in  findings  and  retaining  talented  employees  (Tarique  & 
Schuler,  2009).  However,  demographic  trends  differ  across  different 
parts  of  the  world.  Whereas  in  a  US  or  European  setting,  the  problem 
lies  with  the  impending  retirement  of  large  groups  of  older  employees 
and  the  fact  that  fewer  and  fewer  young  people  are  entering  the 
16
workforce,  in  other  parts  of  the  world,  there  appears  to  be  an 
oversupply  of  young  workers  (Guthridge  et  al.,  2008).  Although,  in  the 
European  Union,  younger  workers  are  on  average  achieving  higher 
levels of education, demand for educated workers is still not being met 
(Tucker  et  al.,  2005).  As  for  Belgium,  demographic  trends  predict  that 
the  working‐age  population  will  stop  increasing  from  the  year  2015 
onwards.  After  2020,  it  will  start  shrinking,  causing  a  decrease  in 
potential  labor  market  participation  (Sels  et  al.,  2008).  Although 
Western  organizations  have  been  ‘eyeing’  young  talent  from  emerging 
economies  willingly,  question  marks  remain  over  the  transferability  of 
this source of talent (Guthridge et al., 2008). 
 
The  term  ‘war  for  talent’  is  largely  grounded  in  the  ascertainment  that 
demand for skilled people is slowly but surely outpacing supply (Tulgan, 
2001). Furthermore, there is no reason to believe that this trend will go 
into reverse in the foreseeable future (Hiltrop, 1999). 
 
Psychological contract drivers. In the recent careers literature, there 
has  been  a  renewed  interest  in  the  psychological  contract  (Hess  & 
Jepsen,  2009).  In  general,  it  is  assumed  that  as  organizations  are 
increasingly less able to promise stable, long‐term careers, career actors 
are  distancing  themselves  from  their  organizations  in  turn  (Hiltrop, 
1999;  Tulgan,  2001).  Since  ‘employment’  is  no  longer  a  given  for 
employees,  they  must  now  pursue  ‘employability’  and  acquire 
transferable  skills  that  make  them  more  attractive  to  the  labor  market 
as a whole. As a consequence, employee loyalty towards their employing 
organizations  is  said  to  have  decreased  dramatically  (Sullivan, 
1999).This means that, on top of the demographic shifts that are causing 
a  decrease  in  labor  market  participation  in  its  entirety,  psychological 
contract dynamics are making it harder to retain talented employees.  
 
However, instead of concluding from all of this that talent management 
is  no  longer  workable  (e.g.  Baruch  &  Peiperl,  1997),  the  opposite 
conclusion  is  equally  valid.  Rather  than  stating  that  organizations  with 
stable,  established  career  structures  have  missed  out  on  important 
developments, one could argue that they are in a better position to face 
the war for talent. After all, research has indicated that many ambitious 
young  managers  still  think  in  traditional  career  terms  (Guest  & 
Mackenzie  Davey,  1996).  Organizations  that  rely  primarily  on  external 
hiring  to  ‘buy’  the  knowledge  they  lost  by  way  of  turnover  and 
retirements might find that strategy to be decreasingly effective. Before 
the  demographic  shift,  when  the  talent  supply  exceeded  demand, 
external  hiring  may  have  made  sense  as  a  primary  strategy.  However, 
17
the talent pool is now shrinking, and the demand for talent is escalating. 
Even  if  leadership  talent  could  be  bought,  the  competitive  costs  will 
continue to increase (Calo, 2008). 
 
What is ‘talent management’, anyway? 
 
One  of  the  more  unfortunate  characteristics  of  the  talent  management 
literature is its tendency to slide off into rather vague rhetoric (Collings 
&  Mellahi,  2009;  Lewis  &  Heckman,  2006).  For  instance,  Ashton  and 
Morton  (2005)  declare  that  “good  talent  management  is  of  strategic 
importance”,  although  they  cannot  provide  “a  single  consistent  or 
concise definition” (p. 28). A recent UK survey by the Chartered Institute 
of  Personnel  and  Development  (2009)  found  that  while  51%  of  the 
organizations  surveyed  engaged  in  some  form  of  talent  management, 
only  20%  of  them  worked  with  a  formal  definition  of  the  concept. 
Findings such as these raise the question of how talent management can 
be used to the strategic benefit of organizations when they cannot even 
succeed in establishing a definition for it. As a result, talent management 
runs  the  risk  of  being  merely  a  management  ‘fad’  organizations  run 
along  with  because  all  of  their  competitors  are  doing  so.  Furthermore, 
many  organizations  seem  unwilling  to  explicitly  delineate  what  talent 
management  does  and  does  not  mean  as  they  believe  that  talent 
management  is  a  ‘mindset’  rather  than  a  bundle  of  HRM  practices 
(Creelman,  2004),  or  because  they  like  to  use  the  term  ‘talent’  as  a 
euphemism  for  ‘people’  in  light  of  employer  branding  (Lewis  & 
Heckman, 2006).  
 
Defining ‘talent management’ and ‘high potentials’ 
 
Below,  we  provide  an  overview  of  seven  explicit  definitions  of  talent 
management  we  found  in  the  recent  literature  (on  top  of  our  working 
definition  adopted  from  the  UK  Chartered  Institute  of  Personnel  and 
Development, 2009): 
 
Talent  management  is  defined  as  ‘integrated’  when  several  of  the  following 
eight categories of initiatives interrelate with each other to  create a holistic 
approach  to  talent:  recruitment,  retention,  professional  development, 
leadership/high  potential  development,  performance  management, 
feedback/measurement,  workforce  planning  and  culture  (Morton,  2004,  p. 
3).  
 
Talent management encompasses managing the supply, demand, and flow of 
talent through the human capital engine (Pascal, 2004, p. 9).  
 
Talent management is the systematic attraction, identification, development, 
engagement/retention  and  deployment  of  those  individuals  who  are  of 
18
particular value to an organization, either in view of their ‘high potential’ for 
the future or because they are fulfilling business/operation‐critical roles (UK 
Chartered Institute of Personnel and Development, 2009, p. 2).  
 
In its broadest sense, the term can be seen as the identification, development, 
engagement,  retention  and  deployment  of  talent,  although  it  is  often  used 
more  narrowly  to  describe  the  short‐  and  longer‐term  resourcing  of  senior 
executives and high performers (Warren, 2006, p. 26).  
 
High  potential  identification  and  development  (also  known  as  talent 
management)  refers  to  the  process  by  which  an  organization  identifies  and 
develops  employees  who  are  potentially  able  to  move  into  leadership  roles 
sometime in the future (Jerusalim & Hausdorf, 2007, p. 934).  
 
At its heart, talent management is simply a matter of anticipating the need for 
human capital and then setting out a plan to meet it (Cappelli, 2008, p. 1).  
 
We  define  strategic  talent  management  as  activities  and  processes  that 
involve  the  systematic  identification  of  key  positions  which  differentially 
contribute  to  the  organization’s  sustainable  competitive  advantage,  the 
development  of  a  talent  pool  of  high  potentials  and  high  performing 
incumbents to fill these roles, and the development of a differentiated human 
resource  architecture  to  facilitate  filling  these  positions  with  competent 
incumbents  and  to  ensure  their  continued  commitment  to  the  organization 
(Collings & Mellahi, 2009, p. 2).  
 
(Global)  talent  management  is  about  systematically  utilizing  HRM  activities 
to  attract,  develop,  and  retain  individuals  with  high  levels  of  human  capital 
(e.g.  competency,  personality,  motivation)  consistent  with  the  strategic 
directions  of  the  (multidimensional)  enterprise  in  a  dynamic,  highly 
competitive, and global environment (Tarique & Schuler, 2009, p. 7).   
 
A  first  observation  is  that  some  of  these  definitions  are  quite  vague. 
Furthermore,  most  of  the  above  definitions  of  talent  management  are 
quite  broad,  encompassing  almost  all  domains  of  traditional  HRM. 
Moreover,  several  of  these  definitions  refer  to  ‘talent’  as  the  target 
audience  for  talent  management  without  explicitly  stating  which 
employees  in  organizations  should  be  considered  as  talented.  The 
reason  we  adopted  the  UK  Chartered  Institute  of  Personnel  and 
Development  (2009)  definition  as  our  working  definition  is  because, 
although  broad,  it  provides  specifics  about  both  which  HRM  practices 
should be involved and which employee groups should be targeted. 
 
Lewis  and  Heckman  (2006),  in  their  review  of  the  relevant  literature, 
identified  three  streams  of  talent  management  definitions.  The  first 
stream  defines  talent  management  as  a  collection  of  typical  HRM 
practices,  such  as  recruitment,  selection,  development,  career 
management  and  succession  planning.  The  second  stream  focuses  on 
the  projection  of  staffing  needs  and  the  progression  of  employees 
19
through  the  organization’s  leadership  pipeline.  The  third  stream 
considers  talent  generically,  that  is,  without  regard    for  organizational 
boundaries  or  specific  positions.  Collings  and  Mellahi  (2009)  add  a 
fourth  stream  to  the  list,  which  emphasizes  the  identification  of  key 
positions,  rather  than  key  people,  in  organizational  structures.  This 
latter interpretation of talent management, however, strikes us as rather 
odd  considering  the  historical  shift  in  HRM  from  job‐based  to 
competency‐based levels of analysis (e.g. Lawler, 1994).  
 
More problematically, however, is that the majority of the definitions of 
talent management do not define what ‘talent’ is and, consequently, on 
which employees talent management should be focusing. The segments 
of the workforce to which the term ‘talent’ might apply can range from a 
select  group  of  potential  future  leaders  to  the  whole  of  the  workforce 
(Garrow  &  Hirsch,  2008).  For  instance,  a  study  by  Leigh  (2009) 
uncovered  that  just  under  half  of  the  twenty  companies  interviewed 
indicated  that  they  understood  ‘talent’  to  mean  everyone  in  the 
organization.  
 
The diversity versus the elitist perspective. To our  knowledge, there 
has  been  no  research  to  date  about  the  possible  advantages  and 
disadvantages of investing selectively in talent, versus investing in talent 
in  general  (Lewis  &  Heckman,  2006).  At  one  extreme,  talent 
management  can  be  seen  as  encompassing  all  employees  in  the 
organization  (Ingham,  2006).  The  ‘diversity’  perspective  on  talent 
management advocates that all employees have some form of talent and 
that,  consequently,  talent  management  should  be  aimed  at  developing 
all  employees  to  the  best  of  their  abilities  (Bossuyt  &  Dries,  2008).  No 
matter  how  appealing  the  diversity  perspective  on  talent  management 
sounds, more arguments are found in the talent management literature 
in favor of a more ‘elitist’ perspective.  
 
For  instance,  Boudreau  and  Ramstad  (2005)  argue  that  identifying 
pivotal  talent  pools  –  those  groups  for  which  small  improvements  in 
quality  or  quantity  generate  large  returns  on  measures  of  strategic 
interest  –  should  be  a  central  feature  of  any  talent  management 
program.  Similarly,  Collings  and  Mellahi  (2009)  assert  that  focusing 
mainly  on  high  potentials  facilitates  a  more  deliberate  exploitation  of 
organizational  resources.  Chuai,  Preece  and  Iles  (2008)  describe 
employee  segmentation  as  exactly  that  feature  of  talent  management 
that  distinguishes  it  from  general  HRM,  which  is  more  egalitarian  by 
nature. In contrast, talent management perceives the needs of core and 
periphery  employees  as  noticeably  different  (Ledford  &  Kochanski, 
20
2004).  Finally,  Lin  (2006)  views  employee  segmentation  as  a  logical 
application  of  labor  economics.  If  not  for  segmentation,  the  author 
reasons, all employees would be treated as equally valuable, regardless 
of their performance, competence, potential or other characteristics that 
might  distinguish  them  from  each  other.  Such  an  approach  might  then 
create  unnecessarily  high  costs  in  terms  of  recruitment,  selection, 
training and development, and reward management (Lin, 2006). 
 
High  potentials.  In  the  elitist  perspective,  the  term  ‘talent’  (as 
referring  to  an  individual)  is  more  or  less  equated  with  the  term  ‘high 
potential’,  referring  as  a  rule  to  employees  with  leadership  potential 
(Jerusalim  &  Hausdorf,  2007).  We  should  add,  however,  that  in 
organizations  whose  core  business  is  grounded  in  manufacturing  or 
technology,  talent  management  can  also  be  strongly  aimed  towards 
expert profiles, instead of solely towards future leaders (e.g. Rothwell & 
Poduch,  2004).  The  most  common  focus  of  talent  management, 
however, addresses the future supply of leaders through the leadership 
pipeline (Garrow & Hirsch, 2008).  
 
In  defining  what  a  high  potential  is,  an  important  consideration  is  that 
while  ‘talent’  refers  to  a  personal  attribute  or  characteristic,  ‘high 
potential’  is  a  label  given  to  employees  by  employers  based  on 
assessments  of  their  performance  and  potential  (i.e.  demonstrations  of 
talent)  (Vinkenburg  &  Pepermans,  2005).  For  the  concept  ‘high 
potential’,  as  well,  several  different  definitions  are  found  in  the 
literature. Earlier in this Introduction, we already cited the definition of 
Cope (1998), which we consider concise and to the point. Below, we list 
some more definitions: 
 
High potentials are employees with ultimate potential who will move up the 
hierarchy into increasingly important management positions and eventually 
reach a position close to the top. They advance and change roles faster than 
their  peers;  their  careers  are  carefully  monitored  and  managed;  they  are  a 
small and elite group; they are on a secret list, so that they can be moved on 
and  off  according  to  the  list‐keepers  judgment;  they  are  healthy  and 
dedicated, and the company can count on their stamina and their willingness 
to make the necessary personal sacrifices to continue their fast‐paced career 
paths (Derr, Jones & Toomey, 1988, p. 275). 
 
High  potentials  are  very  ambitious,  creative,  independent  and  intelligent 
employees  that  possess  leadership  potential.  Managing  these  employees 
demands  a  variety  of  strategies,  and  mismanaging  them  will  cost  the 
organization a great deal (Gritzmacher, 1989, p. 422).  
 
High potentials are people identified as being able and willing to ascend the 
corporate ladder (Iles, 1997, p. 347).  
 
21
High potentials advance at a younger age than their peers, and they are part 
of  an  ascendancy  group  considered  to  be  the  company’s  future  leadership 
pool (Roussillon & Bournois, 2002, p. 58). 
 
Those individuals within an organization who are recognized, at that point in 
time, as the organization’s likely future leaders (Cope, 1998, p. 15). 
 
High potential employees embody passion and are characterized by a quick 
movement through various roles in a company, a carefully monitored career 
path and an elite, but usually secretive, status. As the future leaders of their 
organizations,  high  potentials  slide  into  new  positions,  receive  special 
coaching and mentoring, and are expected to deliver superior performances 
(Snipes, 2005, p. 54).  
 
Looking at these definitions, it is clear that the concept of ‘high potential’ 
is  closely  linked  to  the  traditional  understanding  of  career  and  career 
success.  Altman  (1997)  states,  in  reference  to  the  adjective  ‘high’:  “an 
adjective  indicating  success  in  obtaining  desired  outcomes,  with  ‘high’ 
denoting magnitude as well as level, an inference to the echelons of the 
pyramidal hierarchy”. As for potential, it “denotes possibilities, promise 
and  latent  action.  Potential  implies  a  framework  within  which  to 
develop and materialize” (p. 324).  
 
Issues with high potential identification 
 
One step further from defining what ‘high potentials’ are, is establishing 
the criteria organizations should use to decide which of their employees 
should  be  identified  as  high  potentials.  The  model  of  high  potential 
identification  described  by  most  large  organizations  is  very  much 
similar to the sponsored mobility model of career progression that has 
had a long history in the management literature (e.g. Rosenbaum, 1979). 
According  to  this  model,  the  ultimate  career  ‘destinies’  of  employees 
within their organizations need to be determined as early as possible. At 
the  very  beginning  of  their  careers,  those  who  are  ‘destined’  for  upper 
management need to be singled out from average employees, and given 
differential  treatment  in  terms  of  organizational  socialization  and 
training (Ishida, Su & Spilerman, 2002).  
 
However,  many  organizations  struggle  with  the  early  identification  of 
leadership potential (Briscoe & Hall, 1999; Spreitzer, McCall & Mahoney, 
1997).  The  simplest  (but  also  the  most  naïve)  solution  is  to  equate 
potential  with  current  job  performance.  However,  organizations  are 
strongly  advised  against  rating  potential  and  performance  at  the  same 
time, and by the same rater (e.g. the direct supervisor), in the appraisal 
cycle  due  to  the  risk  of  ‘halo  bias’  (e.g.  Fields,  2002).  Halo  bias  occurs 
when a rater adapts his or her ratings on one of two (or more) separate 
22
assessments  so  that  the  overall  assessment  appears  consistent. 
Research has demonstrated that supervisors are finding it very hard to 
rate  an  employee  low  on  potential  when  he  or  she  has  achieved 
excellent  performance  ratings  over  the  course  of  the  evaluation  period 
(Pepermans,  Vloeberghs  &  Perkisas,  2003;  Segalla,  Rouziès  &  Flory, 
2001).  
 
Another  typical  pitfall  in  high  potential  identification  is  ‘right  stuff’ 
thinking.  The  right  stuff  is  whatever  leads  to  a  continuous  high 
performance,  also  in  new  or  challenging  situations  (McCall,  1998). 
Organizations  typically  attempt  to  capture  it  in  competency  profiles, 
which  are  often  based  on    the  profiles  of  their  currently  successful 
executives.  Such  an  approach  has  been  called  the  selection  perspective 
on  high  potential  identification;  an  alternative  approach  is  the 
development perspective (Briscoe & Hall, 1999).  
 
The  selection  versus  the  developmental  perspective.  The  selection 
perspective on high potential identification departs from the notion that 
we  should  identify  high  potentials  based  on  their  resemblance  to 
successful  predecessors.  In  other  words,  organizations  following  this 
perspective  assume  that  knowing  what  made  past  leaders  within  the 
organization successful makes it easier to identify future leaders (Reilly, 
2008).  Traditionally,  organizations  looked  for  innate  attributes  (e.g. 
intelligence,  personality  traits)  in  their  high  potential  identification 
procedures.  More  recently,  however,  and  mirroring  a  general  trend 
within the HRM field (e.g. Lawler, 1994), there has been a shift towards 
competency‐based  high  potential  identification  (Briscoe  &  Hall,  1999). 
The  advantage  of  the  competency‐based  approach  over  the  innate 
abilities  approach  is  that  it  does  not  necessarily  assume  that  “leaders 
are  born,  not  made”  (a  typical  assumption  of  traditional  leadership 
theory, Cawthon, 1996, p. 44). Rather, competencies might be developed 
and improved through experience and training (Reilly, 2008). However, 
the  competency  approach  to  high  potential  identification  has  been  the 
subject of recent critique, as well. For instance, McCall (1998) states that 
the  selection  perspective  on  talent  management  unrealistically  expects 
that  junior  staff  is  able  to  visibly  exhibit  competencies  typical  of 
successful,  experienced  executives.  Furthermore,  by  identifying  high 
potentials  based  on  the  competency  profiles  of  currently  successful 
executives, “we risk choosing people who fit today' s model of executive 
success rather than the unknown model of tomorrow” (Spreitzer et al., 
1997, p. 6), since competency frameworks need to be revised every few 
years  un  order  to  keep  up  with  unfolding  business  and  leadership 
challenges (Briscoe & Hall, 1999).  
23
 
An alternative approach is to adopt a developmental perspective to high 
potential identification. The premise of this perspective is that if people 
learn,  grow  and  change  over  time  (and  consequently,  acquire  new 
competencies),  then  comparing  with  the  competencies  of  50‐year‐olds 
will  not  be  totally  informative  (Spreitzer  et  al.,  1997).  The 
developmental  perspective  departs  from  the  idea  that  the  ability  to 
learn  from  experience  is  the  most  important  indicator  of  leadership 
potential (McCall, 1998; Spreitzer et al., 1997). Briscoe and Hall (1999) 
describe  two  overarching  ‘metacompetencies’  that  affect  an  individual 
employee’s  capacity  to  develop  new  competencies  when  the 
environment  requites  so:  ‘adaptability’  and  ‘identity’.  Adaptability,  as  a 
metacompetency,  helps  an  individual  to  identify  what  he  or  she  must 
learn  in  order  to  reach  personal  goals  concerning  future  performance, 
and  the  motivation  that  is  needed  to  engage  in  learning.  Additionally, 
the identity metacompetency relates to an individual’s self‐concept and 
his  or  her  ability  to  form  accurate  self‐perceptions  based  on  feedback 
from  (learning)  experiences.  Postmodern  careers  require  that  career 
actors  develop  the  ability  to  self‐correct  in  response  to  new  and 
unfamiliar  demands  from  their  environment,  “without  waiting  for 
formal  training  and  development  from  the  organization”  (Hall  &  Moss, 
1998, p. 31).  
 
Issues with high potential retention 
 
“You  go  through  a  long  selection  process  to  identify  people  who  fit 
positions  in  your  organization  chart,  bring  them  in  with  sweetheart 
deals,  spend  lots  of  time,  energy,  and  money  training  them.  Then  what 
happens?  Sometimes  they  stay,  and  sometimes  they  go”.  In  this  quote, 
Tulgan  (2001,  p.  50)  adequately  sums  up  the  concerns  organizations 
have  about  the  return  on  investment  their  talent  management 
procedures might (or might not) generate. Organizations are sometimes 
hesitant to invest in talent management at all since they are afraid that, 
in an era of ‘job hopping’, their high potentials will leave sooner or later 
anyway  (Boudreau  &  Ramstad,  2005).  Galpin  and  Skinner  (2004),  in 
their  survey  of  employees  identified  as  high  potentials,  found  that  50 
per cent of them expected to leave their current organization within two 
to  five  years,  and  82  per  cent  within  five  years.  Given  that  high 
potentials  are  identified  in  light  of  their  long‐term  rather  than  their 
immediate contributions, and given the degree of investments involved 
in  talent  management,  this  is  clearly  an  issue  of  concern  for 
organizations (Calo, 2008). 
24
The  reasons  why  high  potentials  leave  their  organizations  often  relate 
back to their bargaining position in the labor market – an issue already 
mentioned  earlier  in  this  Introduction,  in  the  section  on  labor  market 
‘bifurcation’  (Van  Buren,  2003).  Put  simply,  high  potentials  leave 
organizations  to  pursue  other  career  opportunities  because  they  can 
(Tulgan, 2001). High potentials are typically employees who are highly 
motivated,  highly  skilled  and  highly  employable.  Therefore,  apart  from 
being in the  core employee segment  organizations want to retain, they 
will also be highly sought after by other organizations. As a result, “the 
winner takes it all” (Van Buren, 2003, p. 134), and high potentials might 
find themselves in the comfortable position of receiving the majority of 
internal career opportunities within their organizations without having 
to offer loyalty in return (Tulgan, 2001).  
 
The  question  is  whether  these  dynamics  need  necessarily  discourage 
organizations  of  investing  in  talent  management.  Several  authors 
believe they should not. Calo (2008) describes how consistently relying 
on external hires to fill leadership positions causes lower organizational 
loyalty, both at the leadership level (which is filled with external hires) 
of  the  organization  and  at  the  lower  levels  of  the  organization,  where 
employees  perceive  that  they  are  being  systematically  passed  over  for 
promotion  opportunities.  Furthermore,  the  symbolic  value  of 
organizational  career  management  programs  should  not  be 
underestimated.  For  instance,  Pfeffer’s  (1981)  symbolic  action  theory 
suggests  that  the  mere  presence  of  organizational  policies  and 
management  practices  can  influence  employee  attitudes.  Talent 
management  seeks  to  develop  internal  talent,  which  should  be  viewed 
more favorably by employees than hiring from outside the organization. 
Organizational practices that are discretionary (i.e. not required by law, 
not  a  result  of  union  negotiations)  and  have  not  been  institutionalized 
across all organizations can signal the firm’s core values and  operating 
philosophy.  Organizations  that  invest  greater  time  and  resources  in 
human resource management activities are signaling to employees that 
they  care  for  them  and  are  willing  to  offer  them  ‘special  treatment’ 
(Pfeffer,  1981).  In  addition,  social  exchange  theory  proposes  that, 
subsequently, as one party (the organization) benefits to another party 
(the  high  potentials),  a  sense  of  obligation  is  created  that  requires  the 
latter  party  to  reciprocate  in  some  way  (Eby,  Allen  &  Brinley,  2005). 
Nonetheless, organizations still report en masse that they are struggling 
with  the  ‘retention’  aspect  of  talent  management    (Vloeberghs, 
Pepermans & Thielemans, 2005).  
 
 
 
 
 
 
25

 
 
 
 
Figure I. Research framework.  
26
Dissertation outline 
 
Identification of a research gap 
 
The (limited) literature on talent management is grounded in anecdote 
and  conjecture  rather  than  in  actual  data  collected  from  real‐life 
organizations  and  career  actors  (Collings  &  Mellahi,  2009).  Lewis  and 
Heckman (2006) assert:  
 
Given  the  state  of  affairs  in  the  practitioner  literature,  it  is  distressing  that 
talent  management  does  not  appear  to  be  a  term  with  currency  in  the 
academic literature (p. 143).  
 
Over the course of the last few decades, there have only been a handful 
of  empirical  studies  addressing  talent  management  issues  (e.g. 
Pepermans  et  al.,  2003;  Spreitzer  et  al.,  1997;  Viney  et  al.,  1997, 
Vloeberghs et al., 2005). Furthermore, we do not know of one empirical 
article to date explicitly linking talent management to the developments 
in the careers literature. However, from our Introduction it may be clear 
that  there  is  a  significant  need  and  opportunity  for  more  empirical 
research  on  talent  management.  Clarification  is  needed  on  all  fronts: 
how  should  talent  management  be  defined;  which  criteria  should  be 
used to identify high potentials; is it realistic to set retention targets as 
indicators  of  return  on  investment  in  talent  management  considering 
the current career environment?  
 
Although  one  single  dissertation  cannot  possibly  provide  conclusive 
answers  to  all  of  the  above  questions,  we  have  attempted  to  offer 
empirical  data  and  insights  on  each  of  the  themes  discussed  in  the 
Introduction.  By  doing  so,  we  hope  to  encourage  more  research  on 
‘postmodern‐day  talent  management’  in  the  future,  so  that  much‐
needed theory can be developed to support organizations and individual 
career actors alike (Collings & Mellahi, 2009). This dissertation aims to 
contribute to the field in a threefold manner:  
 
(1)  by  integrating  the  careers  literature  with  the  talent  management 
literature,  thus  creating  a  framework  for  understanding  talent  management 
within the postmodern career context;  
 
(2) by building on empirical data collected from different talent management 
stakeholders  (i.e.  general  managers,  HRM  professionals,  and  individual 
employees identified as high potentials); and 
 
(3) by examining cross‐level effects, in that organization‐level  variables  (i.e. 
the classification of employees into different ‘talent categories’) are related to 
individual‐level variables (i.e. career antecedents and outcomes). 
27
The structure of this dissertation 
 
This  dissertation  is  composed  out  of  six  sections,  the  first  and  the  last 
being  the  Introduction  and  the  Discussion.  Between  these  two  general 
sections,  eight  papers 1   divided  over  four  thematic  chapters,  are 
presented consecutively.  
 
At  the  beginning  of  each  paper,  after  the  abstract,  the  topicality  of  the 
paper  is  indicated  by  linking  it  back  to  Figure  I 2 ,  which  provides  an 
overview of the overall research framework.   
 
The  first  chapter,  Exploring  ‘real’  high  potential  careers,  presents  an 
exploratory  qualitative  study  encompassing  the  entire  research 
framework. The study looks at both organizational‐level and individual‐
level  variables  and  presents  interview  data  collected  from  both  HR 
managers and individuals identified as high potentials.   
 
The  second  chapter,  High  potential  identification,  presents  three 
different  studies  looking  into  the  criteria  organizations  use,  or  should 
use, in assessments of leadership potential. The first study focuses solely 
on  the  organizational  level,  whereas  the  second  and  third  study 
incorporate self‐rated data from individual employees identified as high 
potentials and compares these to data from employees not identified as 
high potentials. Both qualitative and quantitative data are reported.  
 
The  third  chapter,  Career  outcomes  in  high  potential  careers,  presents 
two  quantitative  studies  looking  into  individual  career  outcomes  of 
being  identified  as  a  high  potential  (or  not).  This  chapter  studies  the 
‘bargaining  position’  of  high  potentials  and  its  consecutive  effects  on 
their commitment and satisfaction. 
 
Finally,  the  fourth  chapter,  Constructing  ‘career’,  presents  two  studies 
digging  deep  into  the  meanings  attributed  to  ‘career’  and  ‘career 
success’  within  the  postmodern  career  context.  Both  qualitative  and 
quantitative methods are used.  
 
A more detailed overview of the research questions and methodologies 
adopted  in  each  separate  paper  can  be  found  in  Table  I,  after  the 
                                                            
1  Although,  as  a  rule,  we  apply  the  APA  reference  style  throughout  this  dissertation, 

several of the included papers’ references are in Harvard style, as they  were  published 


in, or submitted to, journals that demand Harvard style referencing.    
2  The  figures  and  tables  presented  in  the  Introduction  and  in  the  Discussion  are 

numbered  with  Roman  numerals;  the  figures  and  tables  presented  in  the  separate 
papers are numbered with decimal numerals, counting from ‘1’ again in each paper. 
28
reference  list  below.  We  will  refer  back  to  Table  I  in  the  Discussion, 
where  we  will  present  a  similar  table  summarizing  the  main  findings 
and publication output for each paper. 
 
References 
 
Adamson,  S.J.,  Doherty,  N.  &  Viney,  C.  (1998).  The  meanings  of  career  revisited:  Implications  for 
theory and practice. British Journal of Management, 9, 251‐259. 
Altman,  Y.  (1997).  The  high  potential  fast‐flying  achiever:  Themes  from  the  English  language 
literature 1976‐1995. Career Development International, 2 (7), 324‐330. 
Amundson,  N.  (2005).  The  potential  impact  of  global  changes  in  work  for  career  theory  and 
practice. International Journal for Educational and Vocational Guidance, 5, 91‐99.  
Arnold, J. & Cohen, L. (2008). The psychology of careers in industrial and organizational settings: A 
critical but appreciative analysis. In G.P. Hodgkinson & J.K. Ford (Eds.), International review of 
industrial and organizational psychology (pp. 1‐44). London, UK: Wiley.  
Arnold,  J.  &  Jackson,  C.  (1997).  The  new  career:  Issues  and  challenges.  British  Journal  of  Guidance 
and Counselling, 25(4), 427‐433. 
Arthur, M. B. (2008). Examining contemporary careers: A call for interdisciplinary inquiry. Human 
Relations, 61 (2), 163‐186. 
Arthur,  M.B,  Hall,  D.T.  &  Lawrence,  B.S.  (1989).  Generating  new  directions  in  career  theory:  The 
case  for  a  transdisciplinary  approach.  In  M.B.  Arthur,  D.T.  Hall  &  B.S.  Lawrence  (Eds.), 
Handbook of career theory (pp. 7‐25). New York, NY: Cambridge University Press. 
Arthur, M.B. & Rousseau, D.M. (1996). The boundaryless career: A new employment principle for new 
organizational era. New York, NY: Oxford University Press.  
Ashton, C. & Morton, L. (2005). Managing  talent for competitive advantage. Strategic HR Review, 4 
(5), 28‐31. 
Barley,  S.R.  (1989).  Careers,  identities,  and  institutions:  the  legacy  of  the  Chicago  School  of 
Sociology. In M.B. Arthur, D.T. Hall & B.S. Lawrence (Eds.), Handbook of career theory (pp. 41‐
59). New York, NY: Cambridge University Press. 
Baruch,  Y.  &  Peiperl,  M.  (1997).  High‐flyers:  Glorious  past,  gloomy  present,  any  future?  Career 
Development International, 2 (7), 354‐358. 
Baruch,  Y.  &  Rosenstein,  E.  (1992).  Career  planning  and  managing  in  high  tech  organizations. 
International Journal of Human Resource Management,3(3), 477‐496. 
Bossuyt, T. & Dries, N. (2008). Talent management en flexibele loopbaanpaden voor de werknemers 
van morgen [Talent management and flexible career tracks for tomorrow's workforce]. In H. 
Calmeyn, K. De Witte, & J. Weverbergh (Eds.), Licht op leren 2008: Leren en ontwikkelen in een 
talentgerichte maatschappij [Spotlight on learning 2008: Learning and developing in a talent­
oriented society] (pp. 55‐90). Leuven, Belgium: LannooCampus.  
Boudreau,  J.W.  &  Ramstad,  P.M.  (2005).  Where’s  your  pivotal  talent?  Harvard  Business  Review,  83 
(4), 23‐24. 
Bouffartigue, P. & Pochic, S. (2001). Cadres nomads: Mythes et réalités [Nomad managers: Myths and 
realities]. Sociologie du Travail, research report.  
Briscoe, J.P.  & Hall, D.T. (1999).  Grooming and picking leaders using  competency frameworks: Do 
they  work?  An  alternative  approach  and  new  guidelines  for  practice.  Organizational 
Dynamics, 28 (2), 37‐52.  
Brown, D. (2002). Career choice and development. San Francisco, CA: Jossey‐Bass. 
Buckingham,  M.  &  Vosburgh,  R.M.  (2001).  The  21st  century  human  resources  function:  It's  the 
talent, stupid! Human Resource Planning, 24 (4), 17‐23. 
Calo,  T.J.  (2008).  Talent  management  in  the  era  of  the  aging  workforce:  The  critical  role  of 
knowledge transfer. Public Personnel Management, 37 (4), 403‐416.  
Cappelli, P. (2008, March). Talent management for the twenty‐first century. Harvard Business 
Review, 1‐8.  
Cawthon, D.L. (1996). Leadership: The Great Man theory revisited. Business Horizons, 10, 44‐
48. 
Chuai, X., Preece, D. & Iles, P. (2008). Is talent management just ‘old wine in new bottles’ ? The 
case of multinational companies in Beijing.  Management Research News, 31 (12), 901‐
911.  
Collin,  A.  &  Watts,  A.G.  (1996).  The  death  and  transfiguration  of  career  –  and  of  career  guidance? 
British Journal of Guidance and Counselling, 24, 385‐398. 
29
Collings,  D.G.  &  Mellahi,  K.  (2009).  Strategic  talent  management:  A  review  and  research  agenda. 
Human Resource Management Review, early online, corrected proof.  
Cope, F. (1998). Current issues in selecting high potentials. Human Resource Planning, 21 (3), 15‐17. 
Creelman, D. (2004, September). Return on investment in talent management: Measures you can put 
to work right now. Washington, DC: Human capital institute research report.  
Cullinane,  N.  &  Dundon,  T.  (2006).  The  psychological  contract:  A  critical  review.  International 
Journal of Management Reviews, 8 (2), 113‐129. 
Dalton,  G.W.  (1989).  Developmental  views  of  careers  in  organizations.  In  M.B.  Arthur,  D.T.  Hall  & 
B.S.  Lawrence  (Eds.),  Handbook  of  career  theory  (pp.  89‐109).  New  York,  NY:  Cambridge 
University Press.  
Dawis,  R.V.  &  Lofquist,  L.H.  (1984).  A  psychological  theory  of  work  adjustment.  Minneapolis,  MN: 
University of Minnesota Press. 
Derr, C.B., Jones, C. & Toomey, E. (1988). Managing  high potential employees: Current practices in 
thirty‐three US corporations. Human Resource Management, 27 (3), 273‐290. 
Dickmann,  M.,  Brewster,  C.  &  Sparrow,  P.  (1999).  International  human  resource  management:  A 
European perspective. New York, NY: Routledge.  
Dyer,  S.  &  Humphries,  M.  (2002).  Normalising  workplace  change  through  contemporary  career 
discourse.  In  L.  Morrow,  I.  Verins  &  E.  Willis  (Eds.),  Mental  health  and  work:  Issues  and 
perspectives (pp. 48‐62). Bedford Park, Asutralia: Auseinet.  
Eby,  L.T.,  Allen,  T.D.  &  Brinley,  A.  (2005).  A  cross‐level  investigation  of  the  relationship  between 
career  management  practices  and  career‐related  attitudes.  Group  and  Organization 
Management, 30 (6), 565‐596. 
Evetts, J. (1992). Dimensions of career: Avoiding reification in the analysis of change. Sociology, 26, 
1‐21. 
Fields, G. S. (2002). Predicting potential for promotion: How the data in human resource information 
systems can be used to help organizations gain competitive advantage (CAHRS Working Paper 
#02‐14). Ithaca, NY: Cornell University, School of Industrial and Labor Relations, Center for 
Advanced Human Resource Studies. 
Fournier,  V.  (1998).  Stories  of  development  and  exploitation:  Militant  voices  in  an  enterprise 
culture. Organization, 5 (1), 55‐80.  
Galpin,  M.  &  Skinner,  J.  (2004).  Helping  high  flyers  fly  high:  Their  motives  and  developmental 
preferences. Industrial and Commercial Training, 36 (3), 113‐116.  
Garrow,  V.  &  Hirsch,  W.  (2008).  Talent  management:  Issues  of  focus  and  fit.  Public  Personnel 
Management, 37 (4), 389‐402.  
Gershuny,  J.  (1993).  Post‐industrial  career  structures  in  Britain.  In  G.  Esping‐Andersen  (Ed.), 
Changing classes (pp. 136‐170). London, UK: Sage.  
Granrose,  C.S.  &  Baccili,  P.A.  (2006).  Do  psychological  contracts  include  boundaryless  or  protean 
careers? Career Development International, 11 (2), 163‐182.  
Greenhaus, J.H. (1987). Career management. Hinsdale, IL: The Dryden Press. 
Greenhaus,  J.H.,  Parasuraman,  S.  &  Wormley,  W.M.  (1990).  Effects  of  race  on  organizational 
experiences,  job  performance  evaluations,  and  career  outcomes.  Academy  of  Management 
Journal, 33 (1), 64‐86. 
Gritzmacher,  K.J.  (1989).  Staying  competitive  through  strategic  management  of  fast‐track 
employees. National Productivity Review, 8 (4), 421‐432. 
Guest,  D.  &  Mackenzie  Davey,  K.  (1996,  February).  Don’t  write  off  the  traditional  career.  People 
Management, 22‐25.  
Gunz, H., Evans, M. & Jalland, M. (2000). Career boundaries in a ‘boundaryless’ career world. In M. 
Peiperl, M.B. Arthur, R. Goffee & T. Morris (Eds.), Career frontiers: New conceptions of working 
lives (pp. 24‐54). Oxford, UK: Oxford University Press.   
Guthridge,  M.,  Komm,  A.B.  &  Lawson,  E.  (2008).  Making  talent  a  strategic  priority.  The  McKinsey 
Quarterly, 1, 49‐59.  
Hall, D. T. (1976). Careers in organizations. Glenview, IL: Scott Foresman. 
Hall,  D.T.  (1996).  The  career  is  dead,  long  live  the  career:  A  relational  approach  to  careers.  San 
Francisco, CA: Jossey‐Bass.  
Hall, D.T. (2002). Careers in and out of organizations. Thousand Oaks, CA: Sage.  
Hall,  D.T.  &  Chandler,  D.E.  (2005).  Psychological  success:  When  the  career  is  a  calling.  Journal  of 
Organizational Behavior, 26 (2), 155‐176.  
Hall,  D.T.  &  Moss,  J.E.  (1998).  The  new  protean  career  contract:  Helping  organizations  and 
employees adapt. Organizational Dynamics, 26(3), 22‐37. 
30
Hallier, J. & Butts, S. (1999). Employers’ discovery of training: Self‐development, employability and 
the rhetoric of partnership. Employee Relations, 21 (1), 80‐94. 
Hayes, R. L., & Oppenheim, R. (1997). Constructivism: Reality is what you make it. In T.L. Sexton & 
B.L. Griffin (Eds.), Constructivist thinking in counseling practice, research, and training (pp. 19‐
40). New York, NY: Teachers College Press. 
Herr,  E.L.  (1997).  Super’s  life‐span  life‐space  approach  and  its  outlook  for  refinement.  Career 
Development Quarterly, 45, 238‐246.  
Hess,  D.M.   &  Jepsen,  N.  (2009).  Career  stage  and  generational  differences  in  psychological 
contracts. Career Development International, 14 (3), 261‐283.  
Hiltrop, J. M. (1999). The quest for the best: Human resource practices to attract and retain talent. 
European Management Journal, 17 (4), 422‐430. 
Hirsch,  P.  &  Shanley,  M.  (1996).  The  rhetoric  of  boundaryless  ‐  or  how  the  newly  empowered 
managerial class bought into its own marginalization. In M.B. Arthur & D.M. Rousseau (Eds.), 
The boundaryless career: A new employment principle for a new organizational era (pp. 218‐
233). New York, NY: Oxford University Press.  
Holland, J. (1973). Making vocational choices: A theory of careers. Englewood Cliffs, NJ: Prentice‐Hall.  
Hughes, E.C. (1937). Institutional office and the person. American Journal of Sociology, 43, 404‐413.  
Iles,  P.  (1997).  Sustainable  high  potential  career  development:  A  resource‐based  view.  Career 
Development International, 2 (7), 347‐353.  
Ingham, J. (2006). Closing the talent management gap. Strategic HR Review, 5 (3), 20‐23.  
Inkson, K. (2004). Images of career: Nine key metaphors. Journal of Vocational Behavior, 65 (1), 96‐
111. 
Ishida, H.,  Su, K.  &  Spilerman,  S. (2002). Models of career  advancement in organizations. European 
Sociological Review, 18 (2), 179‐198.  
Jerusalim,  R.S.  &  Hausdorf,  P.A.    (2007).  Managers'  justice  perceptions  of  high  potential 
identification practices. Journal of Management Development, 26 (10), 933‐950. 
Jacoby, S.M. (1999). Are career jobs headed for extinction? California Management Review, 42, 123‐
145.  
Lawler, E.E. (1994). From  job‐based to competency‐based organizations. Journal of Organizational 
Behavior, 15, 3‐15. 
Ledford,  G.  &  Kochanski,  J.  (2004).  Allocation  training  and  development  resources  based  on 
contribution.  In  L.  Berger  &  D.  Berger  (Eds.),  The  talent  management  handbook:  Creating 
organizational excellence by identifying, developing and promoting your  best people (pp. 218‐
229). New York, NY: McGraw‐Hill.  
Leigh, A. (2009, July). Research topic: Talent management. People Management, 33.  
Lewis,  R.E.  &  Heckman,  R.J.  (2006).  Talent  management:  A  critical  review.  Human  Resources 
Management Review, 16 (2), 139‐154. 
Lin,  W.Z.  (2006,  May).  The  new  key  word  ‘talent  management’  in  retaining  the  top  employees. 
Human Capital Magazine.  
Lundberg,  C.D.  &  Peterson,  M.F.  (1994).  The  meaning  of  working  in  US  and  Japanese  local 
governments at three hierarchical levels. Human Relations, 47, 1459‐1487. 
Mainiero, L.A. & Sullivan, S.E. (2005). Kaleidoscope careers: An alternative explanation for the opt‐
out revolution. Academy of Management Executive, 19 (1), 106‐123. 
Maranda, M. & Comeau, Y. (2000). Some contributions of sociology to the understanding of career.  
In  A.  Collin  &  R.  Young  (Eds.),  The  future  of  career  (pp.  37‐52).  New  York,  NY:  Cambridge 
University Press.  
McCall,  M.W.  (1992).  Executive  development  as  a  business  strategy.  The  Journal  of  Business 
Strategy, 13 (1), 25‐31.  
McCall,  M.W.  (1998).  High  flyers:  Developing  the  next  generation  of  leaders.  Boston,  MA:  Harvard 
Business School Press.  
McDonald,  P.,  Brown,  K.  &  Bradley,  L.  (2005).  Have  traditional  career  paths  given  way  to  protean 
ones?  Evidence  from  senior  managers  in  the  Australian  public  sector.  Career  Development 
International, 10 (2), 109‐129. 
McMahon,  M.  &  Patton,  W.  (1995).  Development  of  a  systems  theory  of  career  development. 
Australian Journal of Career Development, 4, 15‐20.  
Meyer, H. (1995). Organizational environments and organizational discourse: Bureaucracy between 
two worlds. Organization Science, 6 (1), 32‐43.  
Michaels,  E.,  Handfield‐Jones,  H.,  &  Axelrod,  B.  (2001).  The  war  for  talent.  Boston,  MA:  Harvard 
Business School Press. 
31
Miles, R. & Snow, C. (1996). Twenty first century careers. In M.B. Arthur & D.M. Rousseau (Eds.), The 
boundaryless  career:  A  new  employment  principle  for  a  new  organizational  era  (pp.  97‐115). 
New York, NY: Oxford University Press.  
Morton,  L.  (2004).  Integrated  and  integrative  talent  management:  A  strategic  HR  framework.  New 
York, NY: The Conference Board research report. 
Nicholson,  N.  (2000).  Motivation‐selection‐connection:  An  evolutionary  model  of  career 
development.  In  M.  Peiperl,  M.B.  Arthur,  R.  Goffee  &  T.  Morris  (Eds.),  Career  frontiers:  New 
concepts of working life (pp. 54‐75). Oxford, UK: Oxford University Press.   
Noon, M. & Blyton, P. (2002). The realities of work. New York, NY: Palgrave.  
Parsons, F. (1909). Choosing your vocation. Boston, MA: Houghton‐Mifflin.  
Pascal,  C.  (2004)  Foreword.  In  A.  Schweyer  (Ed.),  Talent  management  systems:  Best  practices  in 
technology  solutions  for  recruitment,  retention  and  workforce  planning.  San  Francisco,  CA: 
John Wiley & Sons. 
Patton,  W.  (2008).  Recent  developments  in  career  theories:  The  influences  of  constructivism  and 
convergence.  In  J.A.  Athanasou  &  R.  Van  Esbroeck  (Eds.),  International  handbook  of  career 
guidance (pp. 133‐156). London, UK: Springer. 
Patton,  W.  &  McMahon,  M.  (2006).  Career  development  and  systems  theory.  Connecting  theory  and 
practice. Rotterdam, The Netherlands: Sense Publishers.  
Pepermans,  R.,  Vloeberghs,  D.  &  Perkisas,  B.  (2003).  High  potential  identification  policies:  An 
empirical study among Belgian companies. Journal of Management Development, 22 (8), 660‐
678. 
Pfeffer, J. (1981). Management as symbolic action: The creation and maintenance of organizational 
paradigms. In L.L. Cummings & B.M. Staw (Eds.), Research in organizational behavior (1, pp. 
1‐52). Greenwich, CT: JAI Press. 
Reilly, P. (2008). Identifying the right course for talent management. Public Personnel Management, 
37 (4), 381‐388.  
Remdisch, S. & Dionisius, S. (1998). Training supervisors’ assessment skills – a crucial step in high 
potential selection. European Journal of Work and Organizational Psychology, 7 (4), 559‐564.  
Richardson, M. (2000). A new perspective for counselors: From career ideologies to empowerment 
through work and relationship practices. In A. Collin & R.A. Young (Eds.), The future of career 
(pp. 197‐211). New York, NY: Cambridge University Press. 
Riley, M. (1995). Analysis of managerial experience: Integrating career analysis with labor market 
dynamics. British Journal of Management, 6, 195‐203.  
Rosenbaum,  J.E.  (1979).  Tournament  mobility:  Career  patterns  in  a  corporation.  Administrative 
Science Quarterly, 24, 220‐241.  
Rothwell,  W.  &  Poduch,  S.  (2004).  Introducing  technical  (not  managerial)  succession  planning. 
Public Personnel Management, 33(4), 405–420. 
Rousseau,  D.M.  (1989).  Psychological  and  implied  contracts  in  organizations.  Employee 
Responsibilities and Rights Journal, 2, 121‐139.  
Rousseau,  D.M.  &  McLean  Parks,  J.M.  (1993).  The  contracts  of  individuals  and  organizations. 
Research in Organizational Behavior, 15, 1‐43.  
Roussillon,  S.  &  Bournois,  F.  (2002).  Identifying  and  developing  future  leaders  in  France.  In  C.B. 
Derr, S. Roussillon & F. Bournois (Eds.), Cross­cultural approaches to leadership development 
(pp. 51‐60). Westport, CT: Quorum Books.  
Savickas, M.L. (2000). Renovating the psychology of careers for the twenty‐first century. In A. Collin 
&  R.  Young  (Eds.),  The  future  of  career  (pp.  53‐68).  New  York,  NY:  Cambridge  University 
Press.  
Savickas, M.L., Nota, L., Rossier, J., Dauwalder, J., Duarte, M.E., Guicharde, J., Soresi, S., Van Esbroeck, 
R. & van Vianen, A.E.M. (2009). Life designing: A paradigm for career construction in the 21st 
century. Journal of Vocational Behavior, in press, corrected proof.  
Segalla, M., Rouziès, D.  & Flory, M. (2001). Culture and career advancement in Europe: Promoting 
team players vs. fast trackers. European Management Journal, 19 (1), 44‐57. 
Sels  L.,  Van  Woensel  A.  &  Herremans,  W.  (2008).  Over  rode,  oranje  en  groene  lichten  in  het 
eindeloopbaanbeleid  [About  red,  orange  and  green  lights  in  the  ‘career  ending  policies’]. 
Over.Werk. Tijdschrift van het Steunpunt WSE, 1, 8 ‐33. 
Sennett, R. (1998). The corrosion of character: The personal consequences of work in new capitalism. 
London, UK: Norton.  
Snipes,  J.  (2005,  November).  Identifying  and  cultivating  high  potential  employees.  Ninth  House 
Editorial Series, 1‐6.  
32
Spreitzer, G.M., McCall, M.W. & Mahoney, J.D. (1997). Early identification of international  executive 
potential. Journal of Applied Psychology, 82 (1), 6‐29. 
Sullivan,  S.E.  (1999).  The  changing  nature  of  careers:  a  review  and  research  agenda.  Journal  of 
Management, 25 (3), 457‐484. 
Sullivan,  S.E.,  Carden,  W.A.  &  Martin,  D.F.  (1998).  Careers  in  the  next  millennium:  Directions  for 
future research. Human Resource Management Review, 8 (2), 165‐185.  
Super, D.E. (1953). A theory of vocational development. American Psychologist, 8, 185‐190.  
Super,  D.E.  (1980).  A  life‐span,  life‐space  approach  to  career  development.  Journal  of  Vocational 
Behavior, 16, 282–298.  
Tarique,  I.  &  Schuler,  R.S.  (2009).  Global  talent  management :  Literature  review,  integrative 
framework,  and  suggestions  for  further  research.  Journal  of  World  Business,  46  (2),  early 
online, corrected proof.  
Tucker,  E.,  Kao,  T.  &  Verma,  N.  (2005,  July/August).  Next‐generation  talent  management:  Insights 
on  how  workforce  trends  are  changing  the  face  of  talent  management.  Business  Credit,  107, 
20‐27.  
Tulgan, B. (2001). Winning the talent wars. Employment Relations Today, 23 (1­2), 37‐51.  
UK  Chartered  Institute  of  Personnel  and  Development  (2009).  Talent  management:  An  overview. 
Retrieved on September 20th, 2009 from http://www.cipd.co.uk/subjects/recruitmen/ 
general/talent‐management.htm?IsSrchRes=1.  
Van  Buren,  H.J.  III  (2003).  Boundaryless  careers  and  employability  obligations.  Business  Ethics 
Quarterly, 13 (2), 131‐149. 
Van  Esbroeck,  R.  (2008).  Career  guidance  in  a  global  world.  In  J.A.  Athanasou  &  R.  Van  Esbroeck 
(Eds.), International handbook of career guidance (pp. 23‐44). London, UK: Springer. 
Van  Esbroeck,  R.,  Tibos,  K.  &  Zaman,  M.  (2005).  A  dynamic  model  of  career  choice  development. 
International Journal for Educational and Vocational Guidance, 5, 5‐18.  
Viney, C., Adamson, S. & Doherty, N. (1997). Paradoxes of fast‐track career management. Personnel 
Review, 26 (3), 174‐186. 
Vinkenburg,  C.  &  Pepermans,  R.  (2005).  Top  potentials  in  organisaties  –  Een  inleiding  [Top 
potentials in organizations – An introduction]. In C. Vinkenburg & R. Pepermans (Eds.), Top 
potentials  in  organisaties:  Identificeren,  ontwikkelen,  binden  [Top  potentials  in  organizations: 
Identification, development, engagement] (pp. 3‐16). Assen, The Netherlands: van Gorcum.  
Vloeberghs,  D.,  Pepermans,  R.  &  Thielemans,  K.  (2005).  High  potential  development  policies:  an 
empirical  study among  Belgian  companies.  Journal of Management Development, 24 (6), 546‐
558. 
Walton,  S.,  &  Mallon,  M.  (2004).  Redefining  the  boundaries?  Making  sense  of  career  in 
contemporary New Zealand. Asia Pacific Journal of Human Resources, 42, 75‐95. 
Warren, C. (2006, March). Curtain call: Talent management. People management, 24‐29.  
Wilensky, H.L. (1961). Careers, lifestyles, and social integration. International Social Science Journal, 
12, 553–558. 
 
 
Table I. Overview: research questions and methodology  
 
Chapter  Paper  Research questions  Methodology 
I . Exploring ‘real’ high potential careers 
  1. ‘Real’ high potential careers: An empirical study into the perspectives of organizations and high potentials
    How  do  employees  identified  as  high  potentials  describe  their  personal  career  Interviews
experiences? And how do these contrast  with  the  career  development programs 
described by representatives of their employing organizations? 
II. High potential identification  
  2.  Identification of leadership potential: Is there consensus about ‘the’ criteria?
    Which criteria are considered as most essential for assessing leadership potential  20‐year literature review
by  subject  matter  experts?  Is  there  consensus  between  the  different  parties  Focus groups 
involved in such assessments?  Q‐sorts 
Multidimensional scaling 
Online survey 
33

 
  3. Using emotional intelligence to identify high potentials: A metacompetency perspective 
    What  is  the  utility  of  using  emotional  intelligence  (as  measured  by  the  EQ‐i)  for  Online survey
identifying  high  potentials?  Does  emotional  intelligence  play  a  role  in  high 
potential  careers  mostly  at  the  stage  of  identification,  or  later  on  in  the  career, 
where a lack of emotional intelligence may cause ‘derailment’? And what are the 
relationships  between  being  identified  as  a  high  potential  or  not,  emotional 
intelligence, job performance, and career commitment? 
 
 
 
Table I. Continued  
 
Chapter  Paper  Research questions  Methodology 
 
  4. High potential identification: Examining the developmental perspective
    Is learning agility, indeed, the best indicator of potential, as is put forward by the  Online survey
‘developmental  perspective’  on  high  potential  identification?  Does  it  predict 
potential  above  and  beyond  job  performance?  Is  learning  agility  directly 
assessable  by  the  organization  or  are  its  effects  on  being  identified  as  a  high 
potential or not mediated by actual on‐the‐job learning? And finally, can learning 
agility be developed by experience or career variety? 
 
III. Career outcomes in high potential careers 
  5.    The  role  of  employability  and  firm­specific  capital  in  shaping  the  psychological  contract:  Do  high potentials  and  long­
tenured employees get the better deal? 
     Is  individual  employability  directly  related  to  employee  perceptions  about  their  Online survey
34

psychological contracts? Does  being identified as a high potential or having long 
organizational  tenure  enhance  the  ‘bargaining  position’  of  employees  in  shaping 
their psychological contracts with their employers? 
 
  6. Effects of the high potential label on performance, career success and commitment: A matter of communication?
    What is the impact of being identified as a high potential or not on subsequent job  Online survey
performance,  objective  and  subjective  career  success,  and  career  and 
organizational  commitment?  Are  the  differences  between  high  potentials  and 
non‐high  potentials  intensified  when  organizations  communicate  openly  about 
who  is    identified  as  a  high  potential  or  not?  Is  such  open  communication 
detrimental for the career satisfaction and loyalty of non‐high potentials? 
 
Table I. Continued 
 
Chapter  Paper  Research questions  Methodology 
IV. Constructing ‘career’ 
  7. Exploring four generations' beliefs about career: Is ‘satisfied’ the new ‘successful’?
    Do  people  from  different  generations  experience  different  types  of  career  Online survey 
patterns?  Does the importance attached to organizational security differ between  Online experiment 
generations?    Do  people  from  different  generations  evaluate  career  success 
differently? 
  8. Career success: Constructing a multidimensional model
    What  are  the  different  ways in  which  people  can  define  ‘career  success’  for  Individual interviews
themselves?  To  which  extent  do  these  idiosyncratic  definitions  differ  from  Q‐sorts 
operationalizations commonly found in the careers literature?  Multidimensional scaling 
35

 
 
36
 
37
 
 
 
 
Chapter I. 
Exploring ‘real’ high potential careers 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Paper 1: 
Dries, N. & Pepermans, R. (2008). ‘Real’ high potential careers: An empirical study into 
the perspectives of organizations and high potentials. Personnel Review, 37 (1), 85‐108.  
38
 
39
Paper 1. 
 

‘Real’ high potential careers: An empirical 
study into the perspectives of organizations 
and high potentials 
 
 
This empirical study attempts to make a contribution to career theory in general, and to 
the  literature  on  high  potential  careers  in  particular,  by  examining  the  careers  of  high 
potentials, taking place within the 21st century world of work, from the perspectives of 
the  high  potentials  themselves  as  well  as  those  of  their  organizations.  A  total  of  34 
interviews  were  conducted  within  3  study  samples:  high  potentials  (n  =  14), 
organizational  representatives  employed  by  the  same  organizations  that  provided  the 
high  potential  participants  (n  =  8),  and  organizational  representatives  employed  by 
organizations that  did  not  allow  for  interviewing  of  their  high  potentials  (n  =  12).  The 
study  suggests  that  high  potentials  still  have  traditional‐organizational  careers.  High 
upward  mobility,  low  inter‐organizational  mobility  and  career  self‐management 
emerged  as  key  features  of  high  potential  careers.  Implications  are  spelled  out  with 
respect  to  the  ‘streaming’  of  different  types  of  employees  in  the  workforce  and  the 
importance  of  expectations  management.  As  for  contribution,  not  only  are  the 
viewpoints of individuals largely absent in the literature on high potential careers, the 
majority of publications on the subject matter are also non‐empirical and take a rather 
normative  stance.  The  interview  study  presented  in  this  paper  looks  into  the 
assumptions of ‘real’ high potential careers from the perspectives of the high potentials 
themselves  as  well  as  those  of  their  organizations,  providing  empirical  data  which  is 
interpretive and descriptive rather than normative. 
 
 
 
 

 
40

Figure II. Topicality of Paper 1. 
41
Paper 1.  
 
‘Real’ high potential careers: An empirical study into the 
perspectives of organizations and high potentials 
 
Some  20  years  ago,  Fred  Luthans  accused  management  literature  of 
being  largely  “based  on  a  priori  assumptions  about  what  managers 
actually  do  and  what  they  should  do  to  be  successful”  (Luthans  et  al., 
1985, p. 255). Today, the same seems to be true of the literature on high 
potentials,  i.e.  those  individuals  within  the  organization  who  are 
“recognized,  at  that  point  in  time,  as  the  organization’s  likely  future 
leaders”  (Cope,  1998,  p.  15).  Although  effectively  managing  high 
potentials  and  their  careers  is,  almost  unanimously,  considered  as  one 
of  the  major  challenges  facing  the  21st  century  human  resources 
function  (Buckingham  and  Vosburgh,  2001;  Tulgan,  2001),  empirical 
publications  on  the  topic  are  exceptionally  rare.  While  the  subject 
matter has been tackled by several authors in recent years (e.g. Baruch 
and  Peiperl,  1997;  Spreitzer  et  al.,  1997;  McCall,  1998;  Segalla  et  al., 
2001;  Fields, 2002), remarkably few  studies have built on data coming 
from the actual high potential population itself.  
 
In  the  majority  of  previous  studies,  high  potentials’  direct  supervisors 
were  targeted  as  respondents  or  a  normative  stance  was  taken. 
Normative  publications  –  i.e.  publications  that  are  prescribing  and 
appraising,  rather  than  describing  and  interpretive  –  typically  present 
‘best  practices’  based  on  anecdotal  findings  (e.g.  Ford,  2005;  Miller, 
2006). As a result, discourse on the subject of high potentials and their 
careers  tends  to  be  conjectural  rather  than  based  on  real  empirical 
evidence (Pepermans et al., 2003).  
 
In other studies it is not always clear whether the research sample was 
made up of high potentials or of other (related) types of employees. Cox 
and Cooper (1988) for instance, interviewed managing directors whom 
they  labeled  as  ‘high  flyers’.  This  term,  however,  is  generally  used  to 
designate  successful  managers  –  i.e.  those  that  have  already  ‘arrived’. 
Conversely, the term ‘high potential’ denotes possibilities, promise and 
latent action (Altman, 1997). It may be considered highly inexpedient to 
regard  high  flyers  and  high  potentials  as  interchangeable  research 
populations;  such  practices  contribute  to  the  belief  that  anyone  who  is 
to be labeled as a high potential must be able to display executive‐level 
skills,  knowledge  and  competencies  at  the  time  of  identification,  thus 
completely  ignoring  the  importance  of  learning  from  experience 
(Briscoe and Hall, 1999; Spreitzer et al., 1997). Another population that 
is often confused with the population of  ‘real’  high potentials  is that of 
42
employees on the ‘fast track’ (e.g. Kovach, 1986; Feild and Harris, 1991). 
Fast‐track  development  programs  serve  to  accelerate  the  development 
of potential managers by using frequent job rotations and other special 
opportunities  not  commonly  available  to  other  employees,  all  within  a 
condensed  timeframe  (Larsen,  1997).  However,  many  organizations 
implement  such  development  programs  as  initial  selection  filters,  and 
decide only later which of their fast‐track program participants qualify 
to  receive  the  high  potential  label  (Fields,  2002).  Finally,  some 
researchers  use  samples  made  up of  MBA  students  to  make  inferences 
about  high  potentials  (e.g.  Sabbe  and  Timmerman,  2006),  grounded  in 
the  assertion  that  employees  whose  management  education  is 
sponsored  by  their  organization  are  probably  high  potentials  –  an 
argumentation of questionable validity. 
 
It  seems,  then,  that  there  is  a  significant  need  and  opportunity  for 
researchers to scrutinize high potentials and their careers. But why have 
such studies been exceptionally rare? A possible explanation lies in the 
delicate  nature  of  the  topic  of  high  potentials  and  their  careers,  which 
causes  severe  barriers  for  researchers.  Many  organizations  are 
unwilling  to  expose  their  high  potentials  to  researchers  –  even  though 
the majority amongst them are very much interested in research on the 
matter (and often, are prepared to pay consultancy firms large sums of 
money  to  dissect  their  high  potential  policies).  There  are  still  many 
organizations  that  deem  it  undesirable  to  be  fully  transparent  about 
their high potential policies. A typical belief is that high potentials would 
become  arrogant  and  complacent  if  they  were  to  be  informed  of  their 
status within the organization, which is often referred to as ‘the crown 
prince syndrome’ (Göbel‐Kobialka, 1998). Moreover, organizations fear 
that  employees  who  are  not  labeled  as  high  potentials  will  become 
unmotivated  or  leave  the  organization  when  information  about  the 
organization’s high potential policies is made available to them (Snipes, 
2005).  
 
The 21st century world of work 
 
Before going into further detail on the focus and approach of our study, 
we  wish  to  briefly  outline  the  21st  century  context  affecting  real  high 
potentials  and  their  careers.  Changes,  such  as  business  re‐engineering 
processes,  restructuring,  flattening,  and  downsizing  may  have  caused 
innovation  and  progress,  but  have  also  brought  about  disarray  in  the 
management  of  people  in  the  workplace  (Baruch,  1999).  Employee 
numbers  and  career  opportunities  are  reducing  fast,  leading  to  the 
emergence  of  new  types  of  psychological  contracts  and  the  decline  of 
43
the traditional ‘reward’ of upward mobility (Rousseau, 1995; Iles, 1997). 
Several  authors  (e.g.  Baruch  and  Peiperl,  1997;  Baruch,  1999;  Doyle, 
2000;  Tulgan,  2001;  Kuznia,  2004)  claim  that  organizations  can  no 
longer  promise  long‐term  employment  to  employees,  let  alone  a  rapid 
progression  along  the  organizational  ladder,  and  that  they  should, 
consequently,  refocus  their  efforts  on  motivating  employees  through 
offering  them  greater  skill  improvement  (Arthur  and  Rousseau,  1996). 
As  such,  the  premise  underlying  high  potential  programs  –  that  hard 
work  and  the  display  of  exceptional  talent  are  rewarded  by  a  steady 
progression  in  the  organizational  hierarchy  –  is  undermined  (Baruch 
and Peiperl, 1997; Kuznia, 2004). Baruch and Peiperl (1997) even state 
that  “there  is  no  future  for  Hipos  [high  potentials],  at  least  not  as  we 
have  known  the  phenomenon….  There  are  quite  a  few  Hipos,  and  very 
few places at the top” (p. 354).  
 
Although  the  abovementioned  publications  undeniably  raise  a  number 
of  interesting  and  important  issues,  they  all  share  the  same  weakness: 
they  are  not  based  on  empirical  research,  and  as  such,  rather 
speculative.  Hence,  we  may  ask  ourselves:  is  it  accurate  that  “hipo 
programs are fast becoming at best an anachronism, and at worst a false 
promise used to keep valuable employees in organizations” (Baruch and 
Peiperl, 1997, p. 356)? Are the new career patterns and types that are – 
according  to  the  aforecited  authors  –  emerging  rapidly  in  the  21st 
century  world  of  work  in  fact  irreconcilable  with  the  premise  behind 
high potential careers?  
 
Below,  we  offer  a  brief  review  of  the  literature  on  postmodern  career 
theory  and  career  management,  which  will  be  used  as  a  supportive 
theoretical framework for the interpretation of our study’s findings.  
 
Postmodern career theory 
 
As a result of the postmodern turn in the social sciences during the late 
1980s (Savickas, 1995), attention increasingly shifted from the objective 
to  the  subjective  world  of  work.  Several  authors  posit  that  the 
traditional  (‘organizational’)  career,  determined  by  relatively  stable 
organizational  and  occupational  structures,  is  gradually  being  replaced 
by  more  ‘boundaryless’  career  types,  where  uncertainty  and  flexibility 
are the order of the day (e.g. Arthur and Rousseau, 1996; Collin, 1998). 
The  concept  of  the  boundaryless  career  includes  “a  range  of  possible 
forms  that  defy  traditional  employment  assumptions”  (Arthur  and 
Rousseau, 1996, p. 6), and thus, offers no characterization of any single 
career  type.  Rather,  boundaryless  career  theory  seeks  to  reflect  the 
44
emergent pace of economic change in postmodern society, in which the 
emphasis  lies  on  continuously  changing  career  paths  and  possibilities 
(Littleton et al., 2000).  
 
Postmodern  career  theory  generally  considers  career  to  be  a  social 
construction  rather  than  a  universal  concept  –  thus  defying  traditional 
societal  norms  of  objectively  observable  career  progress  or  success 
(Collin, 1998). As mentioned earlier, postmodern career theory does not 
only carry implications for the definition of careers and career success, 
but  also  for  the  psychological  contracts  that  exist  between  employees 
and  organizations  –  i.e.  “the  perceptions  of  parties  in  the  employment 
relationship  of  their  mutual  obligations”  (Iles,  1997,  p.  349).  While 
traditional  career  models  prescribe  that  people  offer  loyalty  in  return 
for  deferred  rewards  (e.g.  pension  rights,  vacation  time,  promotion 
opportunities,  social  acceptance),  employment  relationships  within 
boundaryless  careers  are  assumed  too  fluid  to  adopt  such  reward 
systems (Littleton et al., 2000). As a result, even stable or fast growing 
organizations would be increasingly unable (or unwilling) to promise as 
well as formally manage career opportunities (Iles, 1997).  
 
Career management literature 
 
Indeed,  contemporary  literature  on  career  management  places 
accountability  primarily  with  the  individual.  Career  management 
includes all interventions to shape careers in organizations, not only by 
the  individuals  concerned,  but  also  formally  and  informally  by  their 
managers  (Mayo,  1991;  Arnold,  1997;  Doyle,  2000).  Traditionally,  in 
high  potential  mobility  programs,  individuals  are  selected  as  early  in 
their  careers  as  possible  and  given  specialized  career  development 
opportunities, allowing them to assume leadership positions as early as 
possible  (Kuznia,  2004).  However,  the  postmodern  perspective  on 
careers highlights more ‘protean’ career types (Hall, 1976, 2002; Briscoe 
and  Hall,  2006).  While  boundaryless  careers  are  characterized  by 
different  levels  of  physical  and  psychological  movement  (beyond 
organizational  boundaries),  protean  careers  emphasize  a  self‐directed 
approach  to  the  career,  in  which  individuals  are  guided  by  their  own 
personal  values  (Briscoe  and  Hall,  2006).  Again,  literature  on  the  topic 
takes  a  postmodern  stance;  personal  choice,  self‐fulfilment  and 
psychological  (intrinsic)  success  are  prioritized  over  organizational 
career management and extrinsic norms of what career success entails 
(Hall, 1976).  
 
45
Recent  literature  has  attempted  to  steer  clear  of  ‘or‐or’  discourse  by 
demonstrating  that  organizational  career  management  (OCM)  and 
career self‐management (CSM) are to be considered as complementary 
rather than supplementary. CSM can never fully substitute OCM; rather, 
both  types  of  career  management  reinforce  one  another.  Those  with 
greater  CSM  skills  appear  to  give  themselves  a  head  start  in  gaining 
access to organizational sources of career support (Sturges et al., 2002; 
De Vos and Buyens, 2005; Forrier et al. 2005).  
 
The  current  study  aims  to  explore  the  careers  of  ‘real’  high  potentials. 
Two  observations  guided  our  research:  first,  the  viewpoints  of 
individuals are largely absent in the literature on high potential careers; 
second,  the  majority  of  publications  on  the  subject  matter  are  non‐
empirical and take a rather ‘normative’ stance. The interview study that 
will be presented in this paper has two corresponding objectives: 
 
(1)  looking  into  the  assumptions  of  real  high  potential  careers  from  the 
perspectives  of  the  high  potentials  themselves  as  well  as  those  of  their 
organizations;  
 
(2) providing empirical data which is interpretive and descriptive rather than 
normative. 
 
Methodology 
 
The  empirical  data  presented  in  this  paper  is  based  on  in‐depth 
interviews  conducted  with  14  high  potentials  and  20  ‘organizational 
representatives’, i.e. HR managers or consultants actively involved in the 
field  of  high  potential  management.  Over  30  organizations,  known  to 
engage  in  high  potential  identification  and  development,  were 
addressed.  Thirteen  of  them  agreed  to  one  or  multiple  interviews, 
although  only  six  of  them  allowed  for  high  potential  employees  to  be 
interviewed.  Furthermore,  two  out  of  these  six  organizations  explicitly 
requested  not  to  refer  directly  to  the  term  ‘high  potential’  during  the 
interviews  with  their  high  potential  employees  –  which  illustrates  the 
ongoing  sensitivity  about  the  topic  in  organizational  settings  (Göbel‐
Kobialka,  1998).  The  study  participants  were  all  employed  in  Belgium 
and  typically  worked  for  large,  multinational  organizations  (employing 
between  750  and  120000  people  worldwide),  situated  in  a  variety  of 
sectoral contexts. 
The  interview  sample  was  split  up  into  three  groups:  high  potentials 
(HP);  organizational  representatives  (OR)  employed  by  the  same 
organizations that provided the HP participants (a); and organizational 
representatives  employed  by  organizations  that  did  not  allow  for 
interviewing of their high potentials (b).  
46
High potentials (HP) 
 
Fourteen  employees,  identified  and  labelled  as  high  potentials  within 
their organizations, were interviewed individually and made up the high 
potential  (HP)  sample  of  this  study.  The  criterion  for  participants’ 
inclusion  in  this  group  was  that  their  organization’s  definition  of  high 
potential  employees  is  in  line  with  the  aforecited  definition  of  Cope 
(1998). Sample demographics for the HP sample are shown in Table 1.  
 
Remarkably,  only  one  participant  in  the  HP  sample  was  a  woman.  A 
discussion of ‘glass ceiling’ issues would take us too far here, although it 
could offer possible explanations for female underrepresentation in high 
potential  programs  (White  et  al.,  1992).  On  the  other  hand,  our  high 
potential  participants  typically  worked  in  male‐dominated  sectors,  an 
unintended  aspect  of  our  data  collection  that  should  not  be  neglected. 
However,  the  mere  existence  of  ‘male‐dominated  sectors’  is  another 
manifestation  of  mobility  barriers  for  women  (e.g.  Barker  and  Monks, 
1998). 
 
Organizational representatives (OR) 
 
In  addition  to  the  interviews  conducted  with  the  HP  sample,  16  HR 
managers  and  4  HR  consultants  were  interviewed  (in  11  cases 
individually,  in  4  cases  by  panel  interview),  making  up  a  group  of  20 
organizational representatives. The participants in the OR sample were 
selected  purposively  to  include  a  range  of  individuals  who  can  be 
considered as experts in the talent management domain; all of them had 
relevant experience in the field, or were (at that time) in a position that 
allowed them to have an overview of their organization’s high potential 
policies.  
 
In  order  to  be  able  to  set  side  by  side  the  views  expressed  by  high 
potentials  and  organizations  in  a  valid  manner,  we  divided  the  OR 
sample  into  two  subsamples:  the  ORa  sample,  which  contained  the 
organizational representatives employed by the same organizations that 
provided the HP participants, and the ORb sample, which was composed 
of  organizational  representatives  employed  by  organizations  that  did 
not allow for high potentials to be interviewed. The conclusions drawn 
further on in this paper are never based on interview data coming from 
the ORb sample alone.  
Table 1. HP interview sample (n = 14) 

Participant code  Organization  Position  Field of work  Tenure  Age  Gender 


HP1.0  1  HR consulting  Junior consultant  Executive search 2.5 yrs  25  Female
HP2.1  2  Finance  Senior manager  Business consulting 8 yrs  32  Male
HP2.2  2  Finance  Senior manager  Business consulting 9 yrs  33  Male
HP2.3  2  Finance  Senior manager  Financial audit 10 yrs  36  Male
HP2.4  2  Finance  Senior manager  Organizational development 10 yrs  36  Male
HP2.5  2  Finance  Manager  Management development 7 yrs  31  Male
HP3.1  3  Electronics  Manager  Business administration 7.5 yrs  29  Male
HP3.2  3  Electronics  Manager  Export 10 yrs  33  Male
HP4.1  4  Telecom  Manager  Field services 5.5 yrs  27  Male
HP4.2  4  Telecom  Manager  Supply chain management 7 yrs  31  Male
HP5.1  5  Distribution  Manager  Department management 2 yrs  25  Male
HP5.2  5  Distribution  Junior officer Corporate HR 3 yrs  27  Male
47

HP5.3  5  Distribution  Manager  Marketing 2.5 yrs  25  Male


HP6.0  6  Security  Manager  Security projects 5 yrs  36  Male
 
 
Table 2. ORa interview sample (n = 8) 

Participant code  Organization  Position  Field of work  Tenure  Age  Gender 


OR1.0  1  HR consulting  Senior consultant Executive search 11 yrs  34  Male
OR2.0  2  Finance  Director Corporate HR  15 yrs  39  Female
OR3.1  3  Electronics  Director Competence management  16 yrs  36  Female
OR3.2  3  Electronics  Director Corporate HR  15 yrs  38  Female
OR4.1  4  Telecom  Director Management development  16 yrs  40  Male
OR4.2  4  Telecom  Director Management development  18 yrs  46  Male
OR5.0  5  Distribution  Manager Training and development 8 yrs  35  Male
OR6.0  6  Security  Director Corporate HR 13 yrs  41  Male
 
 
48
Table 3. ORb interview sample (n = 12) 

Participant code  Organization  Position  Field of work  Tenure  Age  Gender 


OR7.1  7  Materials technology  Manager Management development  20 yrs  49 Male
OR7.2  7  Materials technology  Manager Training and development  15 yrs  38 Male
OR8.1  8  HR consulting  Partner Executive search 18 yrs  47 Male
OR8.2  8  HR consulting  Senior consultant Executive search 8 yrs  33 Female
OR8.3  8  HR consulting  Junior consultant Executive search 1.5 yrs  25 Female
OR9.0  9  Finance  Junior officer Training and development  2.5 yrs  26 Male
OR10.0  10  Finance  Manager Corporate HR 10 yrs  36 Female
OR11.1  11  Materials technology  Manager Corporate HR 22 yrs  52 Male
OR11.2  11  Materials technology  Manager Organizational development  36 yrs  60 Male
OR12.1  12  Energy  Manager Competence management  18 yrs  49 Female
OR12.2  12  Energy  Manager Management development  19 yrs  51 Male
OR13.0  13  Finance  Manager Corporate HR 9 yrs  34 Female
49

 
 
50
Rather,  the  comparison  between  the  HP  data  and  ORa  data  guided  our 
analyses  and  interpretations;  data  from  the  ORb  participants  on  these 
topics  were  then  added  when  available.  Furthermore,  we  thought  it 
would  be  interesting  to  examine  whether  or  not  both  OR  samples 
generated  similar  views,  given  that  they  differ  in  terms  of  openness 
regarding  their  high  potential  policies.  Demographics  for  the  ORa  and 
ORb interview samples are shown in Table 2 and Table 3 respectively. 
 
As  can  be  seen  in  Table  1,  2  and  3,  each  interviewee  (HP/OR)  was 
assigned  a  participant  code  so  that  response  patterns  and  original 
quotes  can  be  traced  back  to  the  participant  in  question  without 
compromizing his or her anonymity. The first digit links the interviewee 
to  his  or  her  employing  organization  (numbered  1  through  6  for 
organizations in which both HP and OR were interviewed, and 7 through 
13  for  organizations  that  only  allowed  for  interviewing  of  OR);  the 
second  digit  refers  to  the  interviewee  him  or  herself  (in  cases  where 
there was more than one interviewee, belonging to the same interview 
sample, within the organization in question).  
 
Interview procedure 
 
Slightly  different  approaches  were  taken  to  interview  each  of  the 
samples.  The  interviews  with  the  HP  sample  were  in‐depth  and  open‐
ended;  this  encouraged  interviewees  to  speak  in  their  own  voices,  and 
consider  at  length  those  topics  that  they  themselves  perceived  as 
relevant  (Duberley  et  al.,  2006).  The  high  potential  participants  were 
invited  to  tell  us  the  stories  of  their  careers,  highlighting  the  critical 
moments,  events  and  choices  that  had  shaped  them.  They  were 
instructed  to  focus  specifically  on  the  events  leading  up  to  their 
identification  as  high  potentials  and  the  way  their  careers  had  evolved 
since  that  time.  In  those  interviews  where  the  use  of  the  term  ‘high 
potential’  was  not  allowed,  it  was  replaced  by  ‘being  perceived  by  the 
organization  as  being  more  successful  or  talented  than  peers’ 
(essentially designating the same thing ‐ it appeared to be the use of the 
term ‘high potential’ per se that seemed to bother certain organizations). 
When no information was spontaneously provided on the interviewee’s 
current career type or on the career management practices affecting his 
or her career, probes were introduced at points where they fit into the 
course  of  the  interview  as  naturally  as  possible  (King,  2004).  Since  we 
were  specifically  keen  to  find  out  if  and  how  high  potentials’  views  on 
their  careers  differ  from  those  of  organizations,  the  element  of 
subjectivity  was  essential  to  the  analysis  of  the  HP  interviews  (Cohen 
and  Mallon,  2001).  In  order  to  investigate  the  viewpoints  of 
51
organizations, a different approach was taken. A more structured topic 
guide  was  used  when  interviewing  the  organizational  representatives 
group  (King,  2004).  As  opposed  to  individuals  telling  us  the  story  of 
their  careers,  this  part  of  the  study  aimed  to  obtain  information  about 
the way organizations manage high potential careers – which is mostly 
through systems and procedures. HR managers and consultants served 
as  organizational  representatives  in  this  study  as  they  are  often 
explicitly involved in the implementation and follow‐up of such systems 
and  procedures  (although  the  extent  to  which  they  are  considered 
strategic business partners within organizations does differ). Interviews 
with  the  OR  sample  started  by  asking  participants  what  their 
organization’s  criteria  were  for  awarding  the  high  potential  label  and 
what competencies employees in this group ought to possess, moving on 
to the policies available for managing the careers of this specific group 
of  employees.  These  policies  were  divided  into  four  broad  domains: 
identification,  training  and  development,  succession  planning  and 
retention  management.  The  data  generated  by  this  –  much  more 
structured – type of interviewing was more objective, i.e. reflected to a 
lesser  extent  the  personal  opinions  of  the  interviewees  than  was  the 
case  in  the  HP  sample.  For  instance,  many  of  the  OR  participants  used 
their  organizations’  high  potential  policy  documents  as  input  for  their 
answers. It is clear that the OR participants did bring their own opinions 
into  the  interviews;  we  believe,  however,  that  the  views  they  put 
forward during the interviews were, on the whole, representative of the 
views  of  their  organizations  as  economic  entities  –  seeing  as  the  OR 
sample  acted  as  policy  contributors  and  spokespersons  for  their 
organizations – while the HP sample acted merely as representatives of 
themselves as individuals.   
 
All  interviews  were  taped  and  transcribed  to  address  issues  of 
credibility  and  confirmability  (Duberley  et  al.,  2006).  Coding  of  the 
interviews  was  carried  out  manually  with  colored  markers;  code  tags 
were marked clearly in margins. We prefer this method over software‐
aided  coding  (e.g.  NUD.IST)  for  reasons  of  ease  and  overview; 
furthermore,  it  is  nowhere  stated  that  manual  coding  would  lead  to  a 
lesser  quality  of  analysis  (King,  2004).  In  using  template  analysis, 
researchers  are  advised  to  use  some  form  of  ‘independent  scrutiny’  of 
their analysis at some stage in the data analysis as a way of checking its 
quality.  We  applied  this  principle  at  the  stage  of  developing  the  initial 
template.  We  (the  two  authors)  both  independently  carried  out 
preliminary coding on a sample of 6 transcripts (3 HP interviews and 3 
OR  interviews),  then  exchanged  our  independent  coding  and  set  aside 
two hours to compare and discuss these, with the aim of agreeing on an 
52
initial template (Duberley et al., 2006). After the first author completed 
the coding process departing from the initial template (which, of course, 
was  to  some  extent  altered  during  the  iterative  process  of  coding  and 
recoding), we both looked for patterns in the distribution of the codes – 
with  some  indication  of  frequency  to  help  us  direct  our  attention  to 
aspects  of  the  data  which  warranted  further  examination  –  thus 
generating  a  framework  within  which  interpretation  of  the  data  could 
take place (King, 2004). 
 
Findings 
 
The  findings  obtained  from  this  interview  study  offer  valuable  insights 
into the assumptions of ‘real’ high potential careers taking place in the 
21st century world of work. The data, acquired from 32 interviews with 
high  potentials  and  organizational  representatives,  have  been  distilled 
into  three  main  themes:  first,  the  interpretations  high  potentials  and 
organizational  representatives  ascribed  to  the  concept  ‘high  potential’; 
second,  the  high  potential  career  types  that  are  portrayed  in  the 
interviews;  and  third,  the  high  potential  career  management  practices 
described  by  the  different  interview  samples.  Below,  each  of  these 
themes is discussed in detail.  
 
Interpretations of the concept ‘high potential’ 
 
Looking  at  the  interpretations  of  the  concept  ‘high  potential’  given  by 
the  HP  sample  and  the  OR  sample,  we  see  that  rather  traditional 
viewpoints are dominant. The sample of organizational representatives 
employed by the same organizations that provided the HP participants 
(ORa)  and  the  sample  of  organizational  representatives  employed  by 
organizations that did not allow for interviewing of their high potentials 
(ORb) both described high potentials as being part of an elite workforce 
segment  (five  ORa  and  five  ORb)  expected  to  advance  upwardly  within 
the  organization  (six  ORa  and  six  ORb).  Similarly,  seven  out  of  the 
fourteen  high  potentials  interviewed  explicitly  stated  that  upward 
mobility is a priority in their careers, and six of them remarked that they 
aspire  to  be  the  organization’s  CEO  one  day,  if  possible  (“But  doesn’t 
everybody?”,  they  commented).  There  were  some  differences  in 
perception  between  the  ORa  and  the  ORb  sample.  For  instance,  seven 
participants  from  the  ORb  sample  (compared  to  only  two  from  the  ORa 
sample) named leadership and strategic contributions on the part of the 
high  potential  as  crucial,  while  six  ORa  participants  (compared  to  zero 
ORb participants) were more inclined to link high potential status to age‐
53
related deadlines (e.g. “Employees over 35 are too old to be identified as 
high potentials”).  
 
Remarkably,  15  out  of  the  20  OR  participants  (five  ORa  and  ten  ORb) 
indicated  that  the  number  one  criterion  serving  as  input  for  the 
identification  of  high  potential  is  current  performance.  In  addition,  11 
out  of  14  high  potentials  felt  that  working  hard  and  being  an  excellent 
performer are the primary prerequisites for obtaining the high potential 
label.  One  could  wonder,  then,  what  separates  ‘high  performers’  from 
‘high potentials’ in the views of our study participants. In this regard, 13 
out  of  14  of  our  HP  interviewees  expressed  the  belief  that  high 
potentials  are  innately  (or  as  a  result  of  early  life  experiences)  more 
talented  than  others,  mostly  in  terms  of  leadership  and  interpersonal 
skills.  Organizational  representatives  on  the  other  hand,  named 
proactiveness and career initiative‐taking (seven ORa and eight ORb) as 
distinctive HP qualities. Possibly a first conflict of interest was detected 
here,  as  it  seems  that  high  potentials  highlight  individual  traits  when 
differentiating  between  high  potentials  and  high  performers,  while 
organizations  emphasized  self‐directed  behavior.  These  differing 
opinions might spell trouble if they cause both parties to hold the other 
accountable for the high potential’s career. We will look deeper into the 
issue  of  accountability  in  the  section  on  high  potential  career 
management.  
 
In  comparing  the  interview  data  from  the  two  samples  composed  of 
organizational  representatives,  it  appears  that  the  ORa  sample 
emphasized  interpersonal  relationships  (e.g.  networking,  coaching, 
customer  focus…)  when  talking  about  high  potential  careers,  while  the 
ORb  sample  highlighted  strategy  and  innovation  (e.g.  long‐term 
orientation,  learning  agility…).  Furthermore,  while  high  potential 
identification  was  mostly  left  to  direct  supervisors,  career  coaches  and 
team members in the view of the ORa sample, the ORb sample reported 
that  their  organizations  use  committees  (including  Board  and 
management members) to identify high potential. All in all, it seems that 
the  high  potential  policies  reported  by  the  ORb  sample  were  more 
formal  and  less  transparent  than  those  reported  by  the  ORa  sample  – 
offering a possible explanation for the difference in ‘openness’ between 
the two groups of organization. However, apart from the above findings, 
the ORa and the ORb sample did not generate widely divergent views on 
high potential careers.  
 
 
 
54
High potential career types 
 
As  mentioned  before,  traditional‐organizational  and  boundaryless 
careers  do  not  represent  single  career  types;  rather,  careers  can  be 
plotted  on  a  continuum  ranging  from  ‘more  traditional’  to  ‘more 
boundaryless’  (Forrier  et  al.,  2005).  Several  aspects  of  careers  were 
taken into account during the interview data analysis in order to be able 
to  make  judgments  about  the  high  potentials’  career  types  (Rousseau, 
1995; Arthur and Rousseau, 1996; Iles, 1997): (1) Does the career take 
place in one single organizational setting?; (2) Are career opportunities 
(e.g.  promotions)  used  as  rewards?;  (3)  Is  the  career  characterized  by 
(rapid)  upward  progression?;  and  (4)  Does  the  emphasis  lie  on  long‐
term  employment  or  on  employability  (i.e.  skill  improvement  aimed  at 
enhancing employee attractiveness to the wider labor market)? 
 
The  low  inter‐organizational  mobility  of  our  14  HP  interviewees  was 
demonstrated  by  the  fact  that  10  of  them  had  only  worked  for  one 
organization during their entire career. The others, who had worked for 
two  or  more  organizations,  had  all  switched  organizations  in  the  first 
few  years  of  their  careers,  and  stabilized  afterwards.  None  of  the  high 
potentials  interviewed  indicated  to  have  any  intention  of  leaving  their 
current organization in the future – although social desirability bias may 
have played a role in this finding (Podsakoff et al., 2003). Furthermore, 
when asked about their future career goals, only four of them claimed to 
have  a  ‘contingency  plan’,  in  case  their  (organization‐specific)  career 
goals would prove unattainable.  
 
Prominent in the interview data were notions of psychological reward. 
The majority of the high potentials interviewed indicated to desire and 
expect  regular  upward  career  moves;  but  such  aspirations  did  not 
appear to originate from a long‐term career perspective, nor from well‐
established career plans. Rather, upward career moves were considered 
as a  type of (short‐term) reward. One of the interviewed high potentials 
remarked: 
 
Young  people  today  are  increasingly  focused  on  themselves,  on  their  work‐
life balance. They are less prepared to make sacrifices …. even so, people are 
still  very  sensitive  about  being  promoted  or  not  ….  In  this  kind  of 
organization,  when  you’re  not  promoted  or  you  don’t  get  a  pay  raise,  [it 
means] they probably think you’re not that good at what you do (HP2.1). 
 
Remarkably, the HP sample linked career success primarily to extrinsic 
reward,  rather  than  to  notions  of  challenge  or  learning,  as  would  be 
expected from postmodern career theory (Collin, 1998). An interviewee 
commented:  
55
….  lateral  moves  can  definitely  be  challenging.  It  takes  up  a  lot  of  time  and 
effort  to  take  on  this  whole  new  role  in  the  organization  and  I’m  sure  it’s 
gratifying when you succeed in a transition like that …. But you don’t really 
feel successful unless you get a real promotion or a raise … and then there’s 
the fact that a lot of people are involved in project work all the time, so every 
time  you  get  started  on  a  new  project,  that’s already  a  lateral move …. So it 
doesn’t really mean anything unless it’s an upward move (HP4.1). 
 
This was not only the view of the HP interview sample; the OR sample 
expressed  a  similar  view.  Six  out  of  the  20  organizational 
representatives  (one  ORa  and  five  ORb)  stated  that  one  of  the  major 
goals  of  their  organization’s  high  potential  policies  was  to  attract, 
motivate  and  retain  talent,  “by  showing  them  that  there  are  plenty  of 
career  opportunities  for  those  who  demonstrate  excellence”. 
Furthermore,  workforce  segmentation  (i.e.  rewarding  excellence 
through  the  assignment  of  ‘special’  labels  to  some  but  not  others, 
accompanied  by  the  necessary  bonuses  and  promotions)  was  seen  as 
the  number  one  means  for  retaining  high  potentials.  Nonetheless,  two 
organizational  representatives  (both  ORa)  asserted  that  high  potential 
policy  makers  should  be  very  careful  not  to  overdo  the  special 
treatment:  
 
It’s dangerous when the system is all about status. Then people start to think 
that you have to be on the list to be somebody …. The problem with that kind 
of mindset is that high potentials usually make up only a really small portion 
of  an  organization’s  population,  so  that  would  imply  that  everyone  else  is 
frustrated (OR4.1). 
 
It  is  not  unlikely  that  most  employees  would  like  to  see  their  efforts 
rewarded  through  promotions  and  pay  raises.  But  as  opportunities  for 
upward  progression  become  increasingly  scarce  (Guest  and  Mackenzie 
Davey,  1996),  organizations  attempt  to  segment  their  workforce, 
ensuring  that  the  career  opportunities  at  hand  are  offered  to  the 
employees  whom  they  most  wish  to  retain.  Besides  being  necessary, 
workforce segmentation is also a very delicate matter:  
 
All  employees  want  to  see  extra  effort  translated  into  extra  rewards  and 
recognition  …  but  when  communication  [about  workforce  segmentation]  is 
too explicit, resentment between colleagues will probably arise. This applies 
especially  to  the  Belgian  culture.  I  get  the  impression  that  in  America  for 
instance, people tend to look up to high potential co‐workers … see them as 
role models  rather than be jealous of them (HP2.1). 
 
As mentioned earlier, 6 out of the 13 organizations that participated in 
our research had more or less open high potential policies; one must not 
forget, however, that this often means that although high potentials may 
be informed about the system, not all employees are. The difficulties in 
56
maintaining  such  ‘half‐open’  communication  policies  become  evident 
when we see that 10 of the organizational representatives in our sample 
(4  ORa  and  6  ORb)  report  using  ‘cyclical’  high  potential  identification 
systems  –  meaning  that  a  person  can  be  a  high  potential  one  year  but 
not the next. It seems that communicating openly to all employees about 
the organization’s workforce segmentation strategies might turn out to 
be the simpler solution in the long run. One HR director remarked:  
 
It  is  standard  procedure  to  start  discriminating  between  employees  some 
years after entry, and everyone is aware of this … There’s really no use being 
secretive about who’s on the list and who isn’t. People will talk to each other 
anyway … they’ll figure it out themselves (OR2.0). 
 
It  seems  that  organizations  are  increasingly  subdividing  their  high 
potential  pools  as  well.  All  of  the  organizational  representatives  in  our 
sample differentiated between different types of high potentials. Fifteen 
of  them  (seven  ORa  and  eight  ORb)  distinguished  all‐round  managerial 
talent  from  experts  and  project  workers;  eleven  (five  ORa  and  six  ORb) 
made  a  distinction  between  high  potentials  based  on  their  functional 
levels;  and  six  of  them  (one  ORa  and  five  ORb)  had  totally  different 
policies for graduate (fast‐track) talent than for managerial talent. This 
indicates  that  high  potential  populations  are  segmented  ‘vertically’  as 
well  as  ‘horizontally’.  Boudreau  and  Ramstad  (2005)  have,  among 
others,  described  how  pivotal  talent  (the  groups  and  individuals  that 
drive the competitive advantage of organizations) can refer to lower‐ as 
well  as  higher‐level  employees  and  to  experts  as  well  as  general 
managers.  In  this  respect,  seven  organizational  representatives  (three 
ORa  and  four  ORb)  asserted  that  not  paying  enough  attention  to  the 
identification  and  development  of  key  experts  is  a  common  pitfall  in 
organizations’ high potential policies, which focus mostly on those with 
managerial  talent.  Another  important  finding  is  that  almost  all 
organizational  representatives  mentioned  that  their  criteria  for  high 
potential  identification  are  stricter  now  than  they  were  in  the  past. 
‘Elitism’ was mentioned by 10 organizational representatives (5 ORa and 
5  ORb)  and  10  high  potentials  as  a  key  feature  of  contemporary  high 
potential policies:  
 
A decade ago, about 10 to 15 percent of our employees were considered high 
potentials. There were more opportunities to climb the ladder back then … and 
we  figured,  if  some  of  them  didn’t  meet  expectations  we  could  always  take 
them off the list. But today we’re much stricter about who gets on the list in the 
first place … it’s about 2 percent worldwide. For instance, we’ll probably only 
pay for an expensive training program when a high potential is asking (OR3.2). 
 
The main reason for the fact that high potentials are offered more career 
opportunities  than  employees  from  other  workforce  segments  is  the 
57
belief  that  internal  successors  usually  perform  better  than  external 
hires. For example:  
 
One  of  the  goals  of  our  high  potential  policy  is  to  be  able  to  grow  talent  in‐
house.  We  do  also  hire  high  profiles  externally,  but  they  usually  show 
adaptation problems. Internal successors fit in with management much quicker 
and master their new jobs much more easily (OR7.1). 
 
Establishing an internal pool of successors (a ‘leadership pipeline’) was 
the number one goal of the high potential policies of the organizations in 
our sample.  Previous research (e.g. Groysberg et al., 2004) has, indeed, 
demonstrated  that  it  is  very  well  possible  that  a  high  potential’s 
performance  ‘plunges’  when  he  or  she  is  recruited  into  a  new 
organization.  Furthermore,  Boudreau  and  Ramstad  (2005)  argue  that 
high  potential  status  is  an  organization‐specific  concept  as  the  type  of 
human  capital  organizations  need  to  support  their  strategies  varies. 
High potentials, too, seem to be aware of the fact that their status may 
well  prove  to  be  organization‐specific.  One  of  our  high  potential 
interviewees commented:  
 
[High potentials] shouldn’t focus too much on vertical growth … you have to 
make  sure  you’re  not  dissatisfied  or  frustrated  all  the  time  because  things 
aren’t moving fast enough. An “if it’s not the way I want it, I’m gone” mindset 
is  dangerous  …  you  might  end  up  disillusioned.  Maybe  other  companies 
cannot  offer  you  the  kind  of  salary  that  you  are  used  to,  especially  when 
you’re  very  high‐level  …  Plus,  different  organizations  might  have  different 
standards for excellence (HP3.2). 
 
High potential career management 
 
Eleven  out  of  the  fourteen  high  potentials  in  our  sample  admitted  to 
continuously  passing  on  information  concerning  personal  career  goals 
and  aspirations  to  management  and  HR.  The  majority  amongst  them 
believed  that  they  owed  their  successful  careers  mainly  to  their  own 
assertiveness:  
 
My major strength is that I am very aware of my position compared to those 
of my co‐workers .… I used to go up to my career coach with entire reports on 
what  I  had  done  and  what  everyone  else  had  achieved  that  term,  I  literally 
handed  him  the  facts  and  figures  ….  I  just  don’t  want  to  depend  on  other 
peoples’  goodwill  or  on  the  improvising  skills  of  my  coach  in  peer  review 
meetings (HP2.5). 
 
Aspects  of  self‐directedness  emerged  spontaneously  during  the 
interviews,  as  in  this  interview  with  an  HR  director  (OR4.1):  “[High 
potentials]  push  against  the  boundaries  of  their  jobs  ….  Typically,  they 
act on their own initiative”. Most high potentials reported to have gone 
58
beyond the boundaries of their job content at one point in their career, 
in  doing  so  often  initiating  some  form  of  change  within  their 
organizations. Some of them had gotten involved in the start‐up of new 
departments or product lines, others had improved existing procedures 
through extensive personal research and cross‐departmental teamwork 
and  yet  others  had  started  training  and/or  mentoring  younger 
employees  on  their  own  initiative.  Several  high  potential  participants 
talked about how their views on career management had evolved during 
their careers:  
 
At first I thought [the organization] would have fixed career tracks set up … 
and all I’d have to do is jump in …. [after some time] I started to realize that if 
I  really  wanted  to  achieve  something,  I’d  have  to  take  charge  of  my  own 
career  ….  [now,  after  being  identified  as  a  high  potential]  I  feel  that  the 
organization  guides  my  career,  but  I’m  the one  who  has  to  live  up  to  it  …. I 
would feel much less satisfied about what I’ve achieved in my career had my 
organization pulled all the strings (HP3.2). 
 
Although  there  was  a  general  consensus  amongst  the  research 
participants in our sample (HP and OR) that high potential careers rely 
primarily on individual initiative, the role organizations play may not be 
neglected.  As  one  HR  manager  (OR5.0)  put  it:  “It  is  the  duty  of  the 
organization  to  create  a  frame  for  growth,  while  it  is  the  duty  of  the 
individual to present himself as a worthy candidate for succession”. 
 
An overview of the most prominent  high potential career  management 
(CM) practices mentioned in the interviews is given in Table 4. We have 
included  all  practices  that,  by  our  own  inference,  emerged  from  the 
interviews  data  as  being  widespread.  We  decided  not  to  include  those 
CM practices that we felt could be classified under the (broadly defined) 
practices  listed  in  Table  4,  nor  practices  that  were  mentioned 
exclusively by the ORb sample. The practices are displayed according to 
CM  domain  (identification,  training  and  development,  succession 
planning  and  retention  management);  furthermore,  in  order  to 
demonstrate the complementary nature of CSM and OCM, we attempted 
to set ‘matching’ practices side by side. 
 
Interestingly,  five  high  potentials  adopted  game‐like  metaphors  when 
describing CSM practices; they used expressions such as ‘playing along’, 
‘withstanding  the  tests’,    ‘collecting  votes’,  ‘playing  your  cards  right’, 
‘hooking on to the right wagon’ and ‘winning the race’.  
Table 4. High potential career management (CM) practices  

  CM practices 
CM domains  CSM  OCM 
Identification  Performance  Performance appraisals
  Initiative  Bottom‐up nomination 
  Visibility  Talent review meetings
  Drive and ambition Development centers
   
Training and development  MBA  Management skills training
  Projects and task forces Challenging assignments
    Early leadership experiences 
  International assignments Job rotation
  Information networking One‐to‐one coaching
   
59

Succession planning  Internal job applications Job interviews


    Assessment centers
  Personal development plans CV database of all employees 
  Political networking Nomination by board members 
   
Retention management  Critical contributions Workforce segmentation
    Reward policies
  Perseverance    Career opportunities
  Realistic goal setting Expectations management 
  Open communication
Notes. CSM = career self‐management, i.e. initiatives or (demonstrations of) achievements instigated by high potentials themselves; 
OCM = organizational career management, i.e. (HRM) practices or policies initiated by organizations. 
 
60
An  interesting  paradox  emerged  from  our  data:  it  seems  that  the  HP 
participants  were,  on  the  one  hand,  sacrificing  a  lot  for  their  careers 
(primarily their private lives) and doing all they could to fall into grace 
with  the  organization’s  makers  (or  at  least  get  their  efforts  noticed)  – 
they  wanted  to  be  part  of  the  ‘in‐group’  –  on  the  other  hand,  they 
seemed to resent their organizations at times. The high potentials in our 
sample seemed to be quite convinced that their organizations were not 
likely  to  do  them  any  favors.  Some  of  them  even  felt  that  they  were 
being  ‘monitored’  for  failure  and  that  some  challenging  assignments 
granted to them by their organizations were really tests in disguise.  
 
Evidently,  high  potential  policies  involve  delicate  matters,  and  keeping 
all those involved satisfied is not an easy task. Therefore, it came as no 
surprise  that  many  OR  interviewees  were  rather  critical  of  their 
organization’s  high  potential  career  management  practices.  As  the 
number  one  pitfall  of  high  potential  policies,  interviewees  mentioned 
‘burning  out’  high  potentials,  i.e.  pushing  them  to  take  up  functions  or 
responsibilities without allowing for an adequate preparation period (5 
ORa  and  9  ORb).  One  high  potential  (HP2.3)  told  us  he  experienced  his 
organization’s  perception  of  his  being  able  to  take  on  any  challenge  at 
any  time  as  rather  stressful:  “…  does  that  mean  that  if  you  can’t  do 
something,  you’re  a  false  positive  ID  or  the  like?”.  Failing  to  manage 
expectations  was  the  second  most  frequently  mentioned  pitfall  in  high 
potential career management (5 ORa and 7 ORb). When organizations do 
not  offer  clear  perspectives  to  high  potentials,  these  might  get 
demotivated and leave: 
 
After five years, I’d had enough. I went up to my boss and told him I wanted 
pay for performance. They didn’t want me to quit so I got to participate in a 
development centre … I got a pay raise and they made some promises about 
future career moves … I like to compare it to a carrot that is dangled in front 
of you. The more you climb the ladder the more explicit their promises get … 
and that’s why you don’t quit, that’s what keeps you going (HP2.3). 
 
A  third  important  pitfall  for  organizations  is  the  incidence  of 
development plateaus (4 ORa and 9 ORb):  
 
Talented  individuals  can  get  stuck  in  the  leadership  pipeline  when 
organizations  do  not  succeed  in  developing  them  properly.  [Organizations] 
usually have a two to three‐year vision, but then their development activities 
reach a plateau … with a lack of high‐level talent as a result … and despite all 
of their investments they will still have to rely on external hires for top‐level 
succession (OR8.1). 
 
Finally, questions were raised about the criteria commonly used in high 
potential  identification  procedures.  Several  OR  interviewees  asserted 
61
that  it  is  very  difficult  to  predict  what  kind  of  human  capital  (or 
competencies)  organizations  will  need  in  the  future,  and  that 
consequently, organizations should definitely try avoiding ‘cloning’ their 
current  management  population  (4  ORa  and  4  ORb).  The  risk  of  bias 
creeping into the identification process was also mentioned (4 ORa and 
3 ORb):  
 
High  potentials  are  either  directly  nominated  by  members  of  the  Board,  or 
have to go through some sort of assessment process. The predictive validity 
of  these  methods  is  however  questionable  ….  First  of  all,  [those  that  are 
involved  in  high  potential  identification  procedures]  are  not  likely  to  admit 
they  were  wrong.  Furthermore,  it  makes  perfect  sense  that  high  potentials 
indeed turn out to be more successful than others … probably because of the 
nomination  itself  and  the  extra  facilities  the  organization  grants  them  as  a 
consequence (OR8.2). 
 
Discussion 
 
Over  the  last  decade  or  so,  several  authors  have  raised  interesting 
questions  about  the  sustainability  of  the  concept  of  ‘the  high  potential 
career’  (e.g.  Baruch  and  Peiperl,  1997;  Baruch,  1999;  Doyle,  2000; 
Tulgan,  2001;  Kuznia,  2004).  It  appears  that  organizations,  facing  the 
economic  pressures  of  the  21st  century  world  of  work,  can  no  longer 
promise  long‐term  employment  to  their  employees,  let  alone  a  rapid 
progression  along  the  organizational  ladder  (Arthur  and  Rousseau, 
1996),  leading  some  authors  to  conclude  that  “there  is  no  future  for 
Hipos,  at  least  not  as  we  have  known  the  phenomenon”  (Baruch  and 
Peiperl,  1997,  p.  354).  However,  the  majority  of  publications  on  the 
topic  of  high  potential  careers  are  non‐empirical  and  take  a  rather 
‘normative’  stance  (i.e.  prescribing  and  appraising,  rather  than 
describing  and  interpretive).  Moreover,  the  viewpoints  of  individuals 
have been grossly neglected in the literature.  
 
The  empirical  study  presented  in  this  paper  attempts  to  make  a 
contribution  to  career  theory  in  general,  and  to  the  literature  on  high 
potential  careers  in  particular,  by  exploring  the  careers  of  ‘real’  high 
potentials  from  the  perspectives  of  the  high  potentials  themselves  as 
well  as  those  of  their  organizations.  In  this  section  we  will  discuss  the 
key  findings  that  have  emerged  from  this  study,  followed  by  some  – 
tentative  –  implications  these  findings  might  carry  for  theory  and 
practice. 
 
It  is  our  belief  that  some  of  the  positions  taken  in  postmodern  career 
theory  are  far  ahead  of  the  realities  of  high  potential  careers.  Several 
recent  studies  have  concluded  that  claims  about  the  speed  and 
62
inevitability  of  the  shift  from  traditional‐organizational  to  more 
boundaryless career types have to be put into perspective (Forrier et al., 
2005)  and  that  the  traditional  career  is  alive  and  well.  As  Guest  and 
Mackenzie  Davey  (1996)  remarked:  “It  is  never  quite  clear  whether 
those  writing  about  the  ‘new  organization’  are  describing  current 
developments, identifying outliers as illustrations of inevitable trends or 
prescribing  the  shape  of  things  to  come  which  any  organization  that 
wishes to survive should heed” (p. 22). Forrier et al. (2005) found that 
only  6  percent  of  the  957  respondents  in  their  study  were  in  a 
boundaryless  career  (in  which  employees  change  employers  and  long 
for  change  and  flexibility)  whereas  60  percent  experienced  bounded 
careers  (in  which  employees  stay  with  their  employer  and  aspire 
stability).  The  authors  questioned  the  portability  of  the  boundaryless 
career concept to other than US settings, the value people place on job 
security  and  the  unionization  of  organizations  (two  cultural  and 
institutional  elements  that  tend  to  reinforce  the  ‘old’  psychological 
contract) being at the heart of the discussion.  
 
Furthermore,  many  authors  departing  from  the  perspective  of 
postmodern career theory seem to assume that “more and more of the 
best people are thinking like free agents” (Tulgan, 2001, p. 37), and that, 
consequently,  traditional‐organizational  career  types  are  no  longer 
wanted by employees. However, recent empirical work has come to the 
conclusion  that  the  majority  of  employees  continue  to  desire  more 
traditional  career  types.  For  instance,  Walton  and  Mallon  (2004) 
concluded  that  “although  the  boundaries  of  career  have  shifted,  they 
have not melted into thin air” (p. 77). Although their study participants 
favoured  career  self‐management,  continuous  learning  and  autonomy, 
they  still  regarded  these  aspects  of  careers  mostly  as  a  means  to 
achieving objective career outcomes such as promotions.  
 
The findings in our study clearly demonstrate that high upward mobility 
and low inter‐organizational mobility are still key features in both high 
potentials’  and  organizations’  views  on  high  potential  careers.  High 
potentials  have  more  traditional  careers,  simply  because  organizations 
prefer to engage internal successors for top management positions, and 
are willing to invest heavily in those that demonstrate the talent and the 
drive to progress within the organization. The need for a stable core of 
talented  employees  who  genuinely  know  the  organization  and  its 
background  is  probably  far  from  evanescent  (Guest  and  Mackenzie 
Davey, 1996; Pepermans et al., 2003). Although many organizations are, 
in  effect,  facing  scarcities  in  promotional  opportunities,  these  are 
countered  by  stricter  high  potential  identification  criteria  and  further 
63
segmentation  within  the  high  potential  population.  Central  to  this 
paper’s argumentation is the notion of ‘workforce segmentation’. In that 
view,  high  potentials  are  those  employees  who  are  indispensable  to 
organizations  (Boudreau  and  Ramstad,  2005).  Organizations  wish  to 
know  who  these  people  are  and  consider  them  separately  from  others 
(identification),  develop  them  strategically  (development)  and  prepare 
them  for  upward  job  moves  (succession  planning).  As  investments  are 
higher  for  this  group  of  employees,  more  efforts  are  made  to  prevent 
them from making inter‐organizational moves (retention management). 
 
It  is  not  our  intention  to  trivialize  the  merits  of  postmodern  career 
theory;  rather,  we  believe  that  although  boundaryless  and  protean 
career  types  apply  to  a  certain  segment  of  the  workforce,  they  do  not 
pertain  to  it  as  a  whole.  Some  authors  (Rousseau  1995;  Larsen,  1997; 
Kuznia  2004)  suggest  that  high  potential  programs  should,  in  the 
current  world  of  work,  place  more  emphasis  on  the  ‘employability’ 
aspect  of  development  as  opposed  to  organization  specific  skills;  our 
research  data,  however,  raises  some  interesting  questions  about  the 
contemporary  psychological  contract  that  is  built  on  the  premise  of 
employability (Arthur and Rousseau, 1996). It seems high potentials are 
still  getting  ‘the  old  deal’  as  they  are  promised  long‐term  career 
perspectives  and  upward  advancement.  But  what  about  all  other 
employees, who are less likely to receive promotions and be targeted for 
retention,  but  apparently  also  less  likely  to  get  proper  training  (in  the 
words of interviewee OR3.2: “…we’ll probably only pay for an expensive 
training  program  when  a  high  potential  is  asking”);  are  they  getting 
neither  the  old  deal,  nor  the  new?  We  hypothesise  that  non‐core 
workers  probably  learn  more  through  lateral  moves  than  high 
potentials  do  –  leaving  aside  fast‐track  rotation  programs  for  high 
potential graduates (Field and Harris, 1991). Through job rotation non‐
core personnel can acquire the employability needed to be attractive to 
other  organizations  (if  necessary  or  desirable)  while  allowing  high 
potential  co‐workers  to  move  up.  Workforce  segmentation  thus  offers 
clear  benefits  to  organizations.  However,  getting  this  notion  across  to 
employees  is  a  delicate  matter,  especially  in  environments  where  high 
potential labels are strongly associated with status and reward.  
 
Several other studies have suggested a ‘streaming’ of different types of 
employees  in  the  21st  century  world  of  work  (e.g.  Tulgan,  2001).  For 
instance,  Sullivan  et  al.  (1998)  posited  that  traditional  organizational 
training and development are still useful for employees belonging to the 
professional  core  of  their  firms,  bearing  in  mind  that  this  group  is 
shrinking due to organizational downsizing. Furthermore, Gratton et al. 
64
(2004)  came  to  the  conclusion  that  only  traditionally  non‐privileged 
employees  had  become  more  boundaryless  with  greater  skills 
development and inter‐organizational mobility, while Viney et al. (1996) 
uncovered  an  apparent  paradox  between  the  evolving  overall  career 
philosophy of short term and ‘no promises’ versus a continuing longer‐
term association with high potentials in organizations.  
 
Figure  1  illustrates  –  in  a  speculative  way  –  how  workforce 
segmentation  might  affect  organizational  career  structures.  The  dotted 
line  marks  the  boundaries  of  the  traditional‐organizational  career, 
where employees come  into the  organization  at  entry level,  work their 
way  up  until  they  reach  their  maximum  potential,  and  leave  the 
organization at retirement. The career structure is shaped like a piramid 
as  each  consectutive  functional  level  is  more  difficult  to  reach,  and 
contains fewer members.  
 

Figure 1. Career types in the 21st century world of work 

In the 21st century world of work, such career structures are no longer 
workable;  rather,  the  working  population  is  segmented  so  that  the 
increasingly scarce amount of organizational (non‐boundaryless) career 
opportunities (and of developmental resources, as well) are reserved for 
two  types  of  core  contributors:  leaders  –  emanating  from  the 
organization’s  high  potential  pool  –  and  experts.  While  leaders  focus 
their  talent  on  maintaining  the  organization’s  culture,  keeping  the 
overall  mission  clear  for  everybody,  assigning  work,  setting  goals,  and 
holding  people  accountable  for  results  (Tulgan,  2001),  key  experts 
65
possess organization‐specific knowledge and skills that are very difficult 
to replace. Establishing stimulating career tracks for key experts – and a 
better  succession  planning  ensuring  timely  transfer  of  expertise  – 
emerged from our interview data as one of the crucial challenges facing 
organizations  today  (which  is  illustrated  by  the  expert  career 
‘development plateau’ depicted in Figure 1).  
 
With  regard  to  high  potential  career  management,  there  was  general 
consensus  between  our  study  participants  that  self‐direction  and 
initiative‐taking  are  key  characteristics  of  real  high  potential  careers. 
Obviously,  organizations  would  support  this  position;  having  assertive 
employees that work hard at demonstrating their potential and creating 
visibility  within  the  organization  would  considerably  facilitate  the 
difficult  task  of  segmenting  the  workforce  (making  it  more  predictable 
and  transparent).  High  potentials,  from  their  side,  like  taking  their 
careers  into  their  own  hands;  they  usually  have  an  aversion  towards 
strictly  delineated  career  tracks,  as  these  might  slow  them  down. 
Furthermore, it appears that they perceive achieving career progression 
through  personal  initiative  as  more  gratifying  (De  Vos  and  Buyens, 
2005).  Nonetheless,  they  do  expect  to  receive  career  guidance  from 
their organizations. It seems that CSM is of central importance prior to 
obtaining  the  high  potential  label  (in  establishing  some  professional 
credibility  and  visibility  within  the  organization)  but  that,  once 
acknowledged by the ‘in‐group’, OCM practices are needed to help high 
potentials  climb  the  ladder  even  further.  This  makes  obvious  sense  as 
high  potentials  are  well  able  to  achieve  performance  goals  and  attend 
networking  events,  but  cannot  participate  in  the  organization’s 
succession planning themselves.  
 
As  for  practical  implications,  ‘managing  expectations’  emerged  as  the 
most important recommendation made by both the high potentials and 
the organizational representatives in our study. Although high potential 
policies  are,  by  definition,  based  on  the  premise  of  high  commitment 
from both parties (McCall, 1998), we see that, in practice, organizations 
are sometimes struggling to keep their end of the deal. When succession 
planning  fails  or  unexpected  change  occurs  in  the  organization’s 
(economical, political …) context, high potentials without a contingency 
plan  may  well  end  up  disillusioned.  In  order  to  avoid  such  unpleasant 
surprises for high potentials and to ensure that no damage is done to the 
organization  –  and  its  image  as  an  employer  –  a  well‐organised 
succession  planning  and  a  transparent  communication  strategy  about 
the  career  opportunities  present  in  the  organization  are  indispensable. 
Furthermore,  high  potentials  themselves  need  to  manage  their  own 
66
expectations  as  well.  It  can  be  observed  that  the  high  potential  career 
track  in  Figure  1  contains  a  ‘bottleneck’,  illustrating  the  fact  that  high 
potentials  need  to  be  realistic  and  practice  some  patience  during  their 
careers, as upward moves may not always come as quickly as desired. 
 
Despite the contributions of the current study to career theory and the 
literature on high potential careers, it is not without limitations. First, it 
is  not  fully  clear  to  what  extent  the  findings  reported  in  this  paper 
possess  external  validity  (i.e.  to  what  extent  the  findings  would  hold 
across different settings, procedures and participants). One could argue 
that the idea of high potential careers being focused on upward mobility 
and  staying  with  one  employer  is  inherently  present  in  Cope’s  (1998) 
definition of a high potential. Consequently, selecting participants based 
on  that  definition  (as  we  did  in  this  study)  would,  then,  typically  bring 
about traditional views on high potential careers. However, none of the 
over  30  organizations  addressed  to  participate  in  the  study  indicated 
that the term high potential is used to designate anything other than a 
potential  candidate  for  internal  succession.  Furthermore,  as  no  ‘non‐
high  potentials’  were  interviewed,  inferences  made  about  differences 
between  high  potentials  and  other  types  of  employees  are  exclusively 
based  on  interview  data  collected  from  high  potentials  and 
organizational  representatives,  and  thus,  partly  speculative.  Finally,  all 
participants  were  employed  in  Belgium  and  typically  worked  for  large 
(in  most  cases  multinational)  companies.  Thirteen  out  of  the  fourteen 
HP participants were men. Issues related to gender and the participants’ 
national as well as organizational culture may play a role in the study’s 
external validity.  
 
Second, interviews contain the danger of common method variance to a 
larger  extent  than  questionnaire  designs,  in  that  they  are  particularly 
prone  to  inducing  socially  desirable  responses.  For  instance,  it  is 
plausible that the organizational representatives were inclined to make 
their  high  potential  policies  out  to  be  better  than  they  really  are. 
Furthermore, characteristics and expectations of the interviewer might 
have influenced the reponse patterns of the interviewees (Podsakoff et 
al.,  2003).  As  the  high  potentials  in  our  sample  were  asked  to 
reconstruct  the  story  of  their  careers  in  hindsight  (knowing  that  they 
are  considered  as  high  potentials  today),  selective  recollection  issues 
may have caused bias in the interview data as well.   
 
As  was  mentioned  before,  the  literature  on  high  potentials  is 
characterized  by  a  lack  of  empirical  scrutiny  and  a  neglect  of  the 
viewpoints  of  high  potential  individuals.  This  study  has  attempted  to 
67
contribute to the literature by offering empirical data collected from real 
high potentials and their organizations. However, more research on the 
topic  is  required  in  order  to  validate  our  findings.  More  specifically,  it 
would  be  interesting  to  do  so  in  a  large‐scale  quantitative  study, 
comparing career data of high potential samples to matching samples of 
non‐high potentials. Additionally, a longitudinal study following a group 
of high potentials throughout their careers would greatly contribute to 
our  insight  in  real  high  potential  career  tracks.  As  such,  our  inferences 
about  the  ‘streaming’  of  different  types  of  employees  could  be  tested. 
Furthermore,  some  other  questions  arise.  How  should  organizations 
segment their workforce? Which criteria can be used for separating one 
group of employees from another, and do they carry predictive validity? 
Do these criteria refer to more or less ‘innate’ talents or to competencies 
that  can  be  developed?  How  can  these  criteria  be  translated  across 
cultures?  Some  studies  have  addressed  questions  similar  to  the  above. 
However,  all  of  them  have  departed  from  the  viewpoints  of 
organizations.  It  appears  there  is  a  great  need  and  opportunity  for 
researchers  to  study  ‘real’  high  potential  careers  by  examining  large 
samples  of  high  potentials  employed  by  different  organizations  at 
different  functional  levels  –  but  this  would  certainly  require  a  greater 
openness of organizations towards research initiatives such as ours.  
 
References 
 
Altman,  Y.  (1997),  “The  high  potential  fast‐flying  achiever:  themes  from  the  English  language 
literature 1976‐1995”, Career Development International, Vol. 2 No. 7, pp. 324‐30. 
Arnold, J. (1997), Managing careers into the 21st century, Chapman, London. 
Arthur, M.B., and Rousseau, D.M. (Eds) (1996), The boundaryless career: A new employment principle 
for a new organizational era, Oxford University Press, Oxford. 
Barker,  P.  and  Monks,  K.  (1998),  “Irish  Women  Accountants  and  Career  Progression:  A  Research 
Note”, Accounting Organizations And Society, Vol. 23 No. 8, pp. 813‐23.  
Baruch,  Y.  (1999),  “Integrated  career  systems  for  the  2000s”,  International  Journal  of  Manpower, 
Vol. 20 No. 7, pp. 432‐57.  
Baruch, Y.  and Peiperl, M. (1997), “High‐flyers: glorious  past, gloomy present, any future?”, Career 
Development International, Vol. 2 No. 7, pp. 354‐58. 
Boudreau,  J.W.  and  Ramstad,  P.M.  (2005),  “Where’s  Your  Pivotal  Talent?”,  Harvard  Business 
Review, Vol. 83 No. 4, pp. 23‐4. 
Briscoe, J.P. and Hall, D.T. (1999), “Grooming and Picking Leaders Using Competency Frameworks: 
Do  They  Work?  An  Alternative  Approach  and  New  Guidelines  for  Practice”,  Organizational 
Dynamics, Vol. 28 No. 2, pp. 37‐52.  
Briscoe,  J.P.  and  Hall,  D.T.  (2006),  “The  interplay  of  boundaryless  and  protean  careers: 
Combinations and implications”, Journal of Vocational Behavior, Vol. 69 No. 1, pp. 4‐18. 
Buckingham, M. and Vosburgh, R.M. (2001), “The 21st century Human Resources function: it's the 
talent, stupid!”, Human Resource Planning, Vol. 24 No. 4, pp. 17‐23. 
Cohen,  L.  and  Mallon,  M.  (2001),  “My  Brilliant  Career?  Using  Stories  as  a  Methodological  Tool  in 
Careers Research”, International Studies of Management and Organization, Vol. 31 No. 3, pp. 
48‐68. 
Collin, A. (1998), “New challenges in the study of career”, Personnel Review, Vol. 27 No. 5, pp. 412‐
25.  
Cope, F. (1998), “Current issues in selecting high potentials”, Human Resource Planning, Vol. 21 No. 
3, pp. 15‐7. 
68
Cox,  C.J.  and  Cooper,  C.L.  (1988),  High  Flyers:  An  Anatomy  of  Managerial  Success,  Basil  Blackwell, 
Oxford. 
De  Vos,  A.  and  Buyens,  D.  (2005),  “Organizational  versus  individual  responsibility  for  career 
management: complements or substitutes”, available at: 
   www.vlerick.be/research/workingpapers/vlgms‐wp‐2005‐18.pdf  (Accessed  March  27th, 
2006).  
Doyle,  M.  (2000),  “Management  development  in  an  era  of  radical  change:  evolving  a  relational 
perspective”, Journal of Management Development, Vol. 19 No. 7, pp. 579‐601. 
Duberley, J., Mallon, M. and Cohen, L. (2006). “Exploring career transitions: accounting for structure 
and agency”, Personnel Review, Vol. 35 No. 3, pp. 281‐96.  
Eby,  L.T.,  Allen,  T.D.  and  Brinley,  A.  (2005),  “A  Cross‐Level  Investigation  of  the  Relationship 
between  Career  Management  Practices  and  Career‐Related  Attitudes”,  Group  and 
Organization Management, Vol. 30 No. 6, pp. 565‐96. 
Feild, H.S. and Harris, S.G. (1991), “Participants' Frustrations in Fast‐track Development Programs”, 
Leadership and Organization Development Journal, Vol. 12 No. 4, pp. 3‐8. 
Fields,  G.S.  (2002),  “Predicting  Potential  For  Promotion:  How  The  Data  In  Human  Resource 
Information  Systems  Can  Be  Used  To  Help  Organizations  Gain  Competitive  Advantage”, 
available  at:  www.ilr.cornell.edu/depts/cahrs/downloads/pdfs/workingpapers/WP02‐
14.pdf (Accessed March 27th, 2007). 
Ford,  M.  (2005),  “Let  talent  bloom:  Cultivate  innovation  by  moving  high  potential  employees 
around, up, and even out of your IT organization”, CIO Magazine, Vol. 18 No. 19, pp. 26‐9.  
Forrier, A., Sels, L. and Verbruggen, M. (2005), “Career Counseling in the New Career Era: A Study 
about the Influence of Career Types, Career Satisfaction and Career Management on the Need 
for Career Counseling”, available at: www.econ.kuleuven.be/fetew/pdf_publicaties/0580.pdf 
(Accessed March 27th, 2007). 
Göbel‐Kobialka,  S.  (1998),  “Reaching  Business  Excellence  Through  Sound  People  Management”, 
European Journal of Work and Organizational Psychology, Vol. 7 No. 4, pp. 549‐56.  
Gratton, L. (2004),  The Democratic Enterprise: Liberating your business  with freedom,  flexibility and 
commitment, Prentice Hall, London. 
Gritzmacher,  K.J.  (1989),  “Staying  competitive  through  strategic  management  of  fast‐track 
employees”, National Productivity Review, Vol. 8 No. 4, pp. 421‐32. 
Groysberg,  B., Ashish N. and  Nitin N. (2004), "The Risky  Business of Hiring Stars", Harvard Business 
Review Vol. 82 No. 5, pp. 1‐10. 
Guest,  D.  and  Mackenzie  Davey,  K.  (1996),  “Don't  write  off  the  traditional  career”,  People 
Management, Vol. 2 No.4, pp. 22‐5. 
Hall, D. T. (1976), Careers in organizations, Scott, Foresman and Company, Glenview. 
Hall, D.T. (2002), Careers In and Out of Organizations, Sage Publications, Thousand Oaks. 
Iles,  P.  (1997),  “Sustainable  high  potential  career  development:  a  resource‐based  view”,  Career 
Development International, Vol. 2 No. 7, pp. 347‐53.  
King,  N.  (2004)  “Using  Templates  in  the  Thematic  Analysis  of  Text”,  in  Cassell,  C.  and  Symon,  G. 
(Eds), Essential guide to qualitative methods in organizational research, Sage Publications Ltd, 
London, pp. 256‐70. 
Kovach,  B.E.  (1986),  "The  derailment  of  fast  track  managers",  Organizational  Dynamics,  Vol.  15 
No.2, pp.41‐8. 
Kuznia,  K.  (2004),  “The  Rhetoric  and  Reality  of  Fast  Track  Management  Development  Programs”, 
available  at:  http://www.midwestacademy.org/Proceedings/2004/papers/Kuznia.doc 
(Accessed March 27th, 2007).  
Larsen,  H.  (1997),  “Do  high‐flyer  programs  facilitate  organizational  learning?”,  Journal  of 
Managerial Psychology, Vol. 12 No. 1, pp. 48‐59.  
Littleton,  S.M.,  Arthur,  M.B.,  Rousseau,  D.M.  (2000),  "The  future  of  the  boundaryless  career",  in 
Collin, A., Young, R.A. (Eds), The Future of Career, Cambridge University Press, New York, NY, 
pp.101‐14.  
Luthans,  F.,  Rosenkrantz,  S.A.  and  Hennessey,  H.W.  (1985),  “What  Do  Successful  Managers  Really 
Do? An Observation Study of Managerial Activities”, The Journal of Applied Behavioral Science, 
Vol. 21 No. 3, pp. 255‐70. 
Mayo, A. (1991), Managing careers, IPM, London.  
McCall, M.W. (1998), High flyers: developing the next generation of leaders, Harvard Business School 
Press, Boston.  
Miller, J. (2006),  “Attend to the talent: The right team can be the secret to success”, CA Magazine, 
Vol. 139 No. 1, pp. 49‐50. 
69
Pepermans,  R.,  Vloeberghs,  D.  and  Perkisas,  B.  (2003),  “High  potential  identification  policies:  an 
empirical study among Belgian companies”, Journal of Management Development, Vol. 22 No. 
8, pp. 660‐78. 
Podsakoff,  P.M.,  MacKenzie,  S.B.  and  Lee,  J.  (2003),  “Common  Method  Biases  in  Behavioral 
Research:  A  Critical  Review  of  the  Literature  and  Recommended  Remedies”,  Journal  of 
Applied Psychology, Vol. 88 No. 5, pp. 879‐903.  
Rousseau,  D.M.  (1995),  Psychological  Contracts  in  Organizations:  Understanding  Written  and 
Unwritten Agreements, Sage, Newbury Park, CA. 
Savickas, M. (1995), “Current theoretical issues in vocational psychology: convergence, divergence, 
and schism”, in Walsh, W.B. & Osipow, S.H. (Eds), Handbook of Vocational Psychology: theory, 
research and practice (2nd Edition), Lawrence Erlbaum Associates, Mahwah, New Jersey, pp. 
1‐34. 
Sabbe,  C.  and  Timmerman,  M.  (2007),  High  potentials:  feiten  en  fabels  [High  potentials:  facts  and 
fictions], Kluwer, Mechelen, Belgium.  
Segalla,  M.,  Rouziès,  D.  and  Flory,  M.  (2001),  "Culture  and  Career  Advancement  in  Europe: 
Promoting Team Players vs.  Fast  Trackers", European  Management Journal, Vol. 19 No. 1, pp. 
44‐57. 
Snipes,  J.  (2005),  “Identifying  and  Cultivating  High  potential  Employees”,  available  at: 
http://www.ninthhouse.com/papers/2%20CLO_HIPOsArticle.pdf  (Accessed  March  27th, 
2007). 
Spreitzer,  G.M.,  McCall,  M.W.  and  Mahoney,  J.D.  (1997),  “Early  Identification  of  International 
Executive Potential”, Journal of Applied Psychology, Vol. 82 No. 1, pp. 6‐29. 
Sturges,  J.,  Guest,  D.,  Conway,  N.  and  Mackenzie  Davey,  K.  (2002),  “A  longitudinal  study  of  the 
relationship between career management and organizational commitment among graduates 
in the first ten years at work”, Journal of Organizational Behavior, Vol. 23 No. 6, pp. 731‐48.  
Sullivan, S.R., Carden, W.A. and Martin, D.F. (1998), “Careers in the next millennium: directions for 
future research”, Human Resource Management Review, Vol. 8 No. 2, pp. 165‐85. 
Tulgan, B. (2001), “Winning the Talent Wars”, Employment Relations Today, Vol. 23 No. 1‐2, pp. 104.  
Viney,  C.,  Adamson,  S.  and  Doherty,  N.  (1997),  “Paradoxes  of  fast‐track  career  management”, 
Personnel Review, Vol. 26 No. 3, pp. 174‐86. 
Vloeberghs, D., Pepermans, R. and Thielemans, K. (2005), “High potential development policies: an 
empirical study among  Belgian  companies”, Journal of Management Development, Vol. 24 No. 
6, pp. 546‐58. 
Walton,  S.  and  Mallon,  M.  (2004),  “Redefining  the  boundaries?  Making  sense  of  career  in 
contemporary New Zealand”, Asia Pacific Journal of Human Resources, Vol. 42 No. 1, pp. 75‐
95. 
White,  B.,  Cox,  C.  and  Cooper,  C.  (1992),  Women’s  Career  Development:  A  Study  of  High  Flyers, 
Blackwell Publishers, Oxford. 

 
70
71
 
 
 
 
Chapter II. 
High potential identification 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Paper 2. 
Dries,  N.  &  Pepermans,  R.  (in  review).  Identification  of  leadership  potential:  Is  there 
consensus about ‘the’ criteria? 
 
Paper 3. 
Dries,  N.  &  Pepermans,  R.  (2007).  Using  emotional  intelligence  to  identify  high 
potentials:  A  metacompetency  perspective.  Leadership  and  Organization  Development 
Journal, 28 (8), 749‐770 .  
 
Paper 4.  
Dries,  N.,  Vantilborgh,  T.  &  Pepermans,  R.  (in  review).  High  potential  identification: 
Examining the developmental perspective. 
72
 
73
Paper 2. 
 

Identification of leadership potential: Is there 
consensus about ‘the’ criteria? 
 
 
In  this  study,  a  multidimensional  model  of  ‘the’  criteria  used  to  identify  leadership 
potential  was  developed.  Building  on  an  extensive  review  of  the  literature,  the  first, 
qualitative  part  of  the  study  revealed  the  77  criteria  considered  most  essential  by 
subject  matter  experts.  Through  multidimensional  scaling,  a  two‐dimensional  model 
(cognition‐conation  versus  extrapersonal‐intrapersonal)  comprising  4  regions 
(information  processing,  learning  agility,  drive,  and  charismatic  leadership)  and  13 
subregions  (input‐seeking  behavior,  insightfulness,  decision  making,  problem  solving, 
willingness  to  learn,  emotional  stability,  personal  adaptability,  drive  for  results, 
persistence,  personal  dedication,  people  management,  inspirational  management,  and 
stakeholder  management)  was  conceptualized.  The  second,  quantitative  part  of  the 
study  found  high  consensus  among  top  managers,  line  managers  and  HR  managers 
about the relevance of the proposed model. 
 
 
 

 
74

Figure III. Topicality of Paper 2. 
75
Paper 2. 
 
Identification of leadership potential: Is there consensus about 
‘the’ criteria? 
 
Within  the  21st  century  business  context,  faced  by  trends  such  as 
globalization,  demographic  changes  in  the  workforce,  digitization  and 
hyper‐competitiveness,  intellectual  capital  has  become  the  prime 
resource  for  organizational  success. The  shift  to  a  knowledge  economy 
demands  a  clear  focus  on  talent  and  leadership  (e.g.  Boudreau  & 
Ramstad, 2005; Frank & Taylor, 2004; Lawler & Mohrman, 2003). As a 
result,  talent  management  has  become  a  ‘hot  topic’  in  organizations 
today  (Buckingham  &  Vosburgh,  2001;  Feild  &  Harris,  1991).  Despite 
the  ever‐increasing  popularity  of  the  concept,  Lewis  and  Heckman 
(2006),  in  their  review  of  the  literature  on  talent  management, 
concluded  that  there  is  a  disturbing  lack  of  clarity  regarding  the 
definition,  scope  and  overall  goals  of  talent  management.  They  found 
three  strains  of  thought  about  talent  management:  talent  management 
as  a  collection  of  human  resources  management  practices;  talent 
management  as  processes  supporting  the  flow  of  employees  through 
jobs  in  an  organization;  and  talent  management  as  focusing  on  talent 
generically, regardless of organizational or functional boundaries.  
 
Although  talent  management  is  generally  considered  to  be  quite  broad 
in  definition,  literature  in  the  field  seems  to  focus  primarily  on 
leadership  potential  –  more  specifically  on  the  concept  of  high 
potentials,  i.e.  “those  individuals  within  an  organization  who  are 
recognized,  at  that  point  in  time,  as  the  organization’s  likely  future 
leaders” (Cope, 1998, p. 15), “people identified as being able and willing 
and  likely  to  ascend  the  corporate  ladder”  (Iles,  1997,  p.  347),  and 
“those  the  company  has  singled  out  as  being  possible  candidates,  in 
time,  for  a  position  on  the  board  of  directors”  (Bournois  &  Rousillon, 
1998,  p.  13).  Some  authors  even  go  as  far  as  equating  talent 
management  with  high  potential  identification  and  development 
(Jerusalim  &  Hausdorf,  2007).  From  this  point  of  view,  talent 
management serves the purpose of selecting high potentials as early in 
their  careers  as  possible  and  then  giving  them  specialized  career 
development  opportunities  allowing  them  to  assume  leadership 
positions as early as possible (Kuznia, 2004). It has been suggested that 
identifying  a  pool  of  high  potentials  can  help  to  ensure  continuity  of 
management  resources  (Hall  &  Seibert,  1992)  and  boost  the  retention, 
motivation  and  performance  of  this  specific  group  (Jerusalim  & 
Hausdorf, 2007).  
 
76
Although  we  feel  that  talent  management  is,  in  fact,  much  broader  in 
scope  than  sponsored  mobility  programs  aimed  solely  at  employees 
with leadership potential, this paper does focus on exactly that aspect of 
it  –  however,  we  wish  to  recognize  that  our  emphasis  on  leadership 
potential does not diminish the individual talents of all employees who 
are  not  on  an  upward  mobility  track.  It  is  our  strong  belief  that  in 
today’s  postmodern  career  environment,  multiple  paths  to  success  are 
possible depending on each individual’s specific talent and on their own 
personal  conception  of  what  success  means  to  them  (Dries  & 
Pepermans,  2008;  Dries,  Pepermans  &  Carlier,  2008).  Nonetheless, 
studies  into  the  ‘rise  and  fall’  of  organizations  are  in  no  doubt  that  the 
quality  of  leadership  greatly  affects  organizational  success.  Although 
competitive advantage does not only depend on an organization’s senior 
leaders, their unique impact on strategy and decision‐making cannot be 
ignored.  Poor  leadership  affects  morale,  gives  rise  to  complacency  and 
to failure to respond to markets and customers (Mayo, 2000).  
 
Whereas before, there was a trend of attracting and hiring ‘stars’ to fill 
leadership  vacancies,  the  current  demographic  situation  (i.e.  structural 
shortage  in  labor  forces  due  to  lower  birth  rates,  a  larger  relative 
percentage  of  older  employees,  and  massive  retirements)  and  changes 
in  psychological  contracts  constructed  by  employees  (Iles,  1997)  are 
causing  organizations  to  be  more  focused  on  early  identification  of 
leadership  potential  in  order  to  develop  and  retain  the  talent  already 
present  in  the  organization.  First  of  all,  career  management  practices 
designed to develop internal talent are viewed much more favorably by 
employees than hiring from outside the organization, and consequently, 
encourage retention (Eby, Allen & Brinley, 2005). Furthermore, research 
has  indicated  that  external  high‐level  hires  often  do  not  live  up  to 
expectations whereas internal successors deliver much more return on 
investment.  Of  course,  a  contingency  approach  needs  to  be  taken; 
external  hires  can  be  required,  for  instance,  when  an  organization’s 
strategy  is  centered  around  innovation  and  renewal  (e.g.  Groysberg, 
Nanda & Nohria, 2004).  
 
The current paper has two main objectives: 
 
(1)  to develop a multidimensional model of ‘the’ criteria essential to the 
identification of leadership potential; 
 
(2) to assess the degree of consensus about the developed model amongst 
different parties involved in assessments of leadership potential.  
 
77
The phrasing of our first objective, to develop ‘the’ model of leadership 
potential,  is  of  course  flavored  with  a  hint  of  sarcasm.  Although 
developing  (and  commercializing)  such  a  generic,  universal  model 
seems to have become the ‘holy grail’ of talent management over the last 
couple  of  decades,  it  is  probably  downright  impossible  (Hollenbeck, 
McCall  &  Silzer,  2006).  Rather,  the  current  paper  aims  to  present  a 
model  of  those  leadership  identification  criteria  considered  by  subject 
matter  experts  as  essential  for  (virtually)  any  organization.  While  a 
comprehensive  model  of  leadership  potential  should,  ideally,  also 
incorporate  situational  variables  –  i.e.  job  variables,  interpersonal 
dynamics,  team  context  variables,  organizational  culture  variables  and 
country  culture  variables  (Hollenbeck  et  al.,  2006)  many  authors  and 
practitioners alike seem to be convinced that there are generic, person‐
centered  elements  all  organizations  look  for  in  future  leaders  (Dries  & 
Pepermans, 2008). 
 
Literature review on leadership potential identification criteria 
 
Although  talent  management  and  the  early  identification  of  leadership 
potential  are,  almost  unanimously,  considered  as  the  major  challenges 
facing  the  21st  century  human  resources  function  (Buckingham  and 
Vosburgh,  2001;  Tulgan,  2001),  empirical  publications  on  these  topics 
are exceptionally rare. Although the subject matter has been tackled by 
several  authors  (e.g.  Baruch  and  Peiperl,  1997;  Lombardo  &  Eichinger, 
2000; Spreitzer, McCall & Mahoney, 1997), remarkably few studies have 
built on empirical data. Normative publications, i.e. publications that are 
prescribing  and  appraising,  rather  than  describing  and  interpretive, 
typically present ‘best practices’ based on anecdotal findings (e.g. Ford, 
2005;  Miller,  2006).  Consequently,  discourse  on  the  subject  of  high 
potentials  and  talent  management  tends  to  be  conjectural  rather  than 
based  on  real  empirical  evidence  (Pepermans,  Vloeberghs  &  Perkisas, 
2003).  
 
In order to collect as much input as possible for the development of our 
model  of  leadership  potential  identification  criteria,  we  conducted  an 
extensive  review  of  the  relevant  literature  from  the  last  20  years. 
Inclusion required that publications had to explicitly describe a number 
of  criteria  that  are,  or  should  be,  used  in  assessments  of  leadership 
potential.  We  searched  a  range  of  established  journals  for  articles 
concerned  with  leadership  potential  over  the  period  1986‐2008.  To 
date,  only  two  empirical  studies  have  explicitly  reported  the 
development and validation of (trademarked) models aimed specifically 
at  high  potential  identification:  the  Prospector  (Spreitzer  et  al.,  1997) 
78
and  the  Choices  (Lombardo  &  Eichinger,  2000).  These  instruments 
depart  from  the  notion  that,  in  today’s  constantly  changing  business 
context,  assessments  of  leadership  potential  should  not  be  based  on 
end‐state  competencies  alone,  and  that  an  element  of  learning  agility 
needs to be incorporated. We also searched for additional terms related 
to leadership potential to be able to provide a more detailed picture of 
relevant research in the past. The additional terms were high potentials 
(e.g. Lombardo & Eichinger, 2000), high flyers (e.g. Dulewicz & Herbert, 
1999),  fast  trackers  (e.g.  Kovach,  1986),  future  leaders  (e.g.  Byham, 
2003), future executives (e.g. Spreitzer et al., 1997), talent (e.g. McCall, 
1992), and promotability (e.g. Cook & Emler, 1999). Our search led to a 
set  of  38  articles.  The  journals  covered  were  Journal  of  Management, 
Academy  of  Management  Journal,  Leadership  Quarterly,  Journal  of 
Occupational  and  Organizational  Psychology,  Administrative  Science 
Quarterly,  Journal  of  Organizational  Behavior,  Career  Development 
International,  Journal  of  Social  Psychology,  Career  Development 
Quarterly, Journal of Vocational Behavior, Human Relations, Organization 
Science,  Human  Resource  Planning,  Organization  Studies,  Journal  of 
Applied Psychology, Personnel Psychology, Journal of Career Development, 
Personnel  Review,  and  Journal  of  Management  Studies.  In  addition,  we 
ran a search in PsycINFO to identify relevant articles published in other 
journals. Furthermore, we added six well‐known books to the list as we 
felt  they  represented  seminal  work  done  in  the  field  of  talent 
management  and  early  identification  of  leadership  potential  (i.e. 
Barham  &  Oates,  1991;  Cox  &  Cooper,  1988;  Kotter,  1988;  Kouzes  & 
Posner,  1987;  Sternberg,  1986;  Zenger  &  Folkman,  2002).  Appendix  A 
provides  an  overview  of  the  articles  (38)  and  books  (6)  identified  as 
relevant in the literature search 3 . Although the presented overview may 
not  be  exhaustive,  we  are  confident  that  it  is,  at  least,  representative  of 
the published work within the field.  
 
In  the  first,  qualitative  part  of  our  study,  we  will  incorporate  the 
identification  criteria  found  in  the  literature  search  into  one  model  of 
leadership potential. Using various qualitative techniques (focus groups, 
Q‐sorts,  and  multidimensional  scaling)  we  will  reduce  the  acquired 
information into a more interpretable and workable, multidimensional, 
model  of  criteria  essential  to  the  early  identification  of  leadership 
potential. 
 
 
 
                                                            
3 A list of 545 leadership potential identification criteria was assembled. It is available 

from the authors upon request. 
79
Different parties involved in assessments of leadership potential 
 
In  previous  studies  of  leadership  potential,  several  parties  have  been 
identified  as  being  (partly)  accountable  for  its  assessment.  A  survey 
study  by  Pepermans  et  al.  (2003)  found  that  top  management, 
immediate  supervisors,  the  HR  department,  line  managers  other  than 
immediate supervisors, and the talented employee him or herself can all 
be  involved  in  assessments  of  leadership  potential  (the  order  in  which 
the  parties  are  listed  denotes  their  degree  of  involvement).  An 
important question in this respect is if these different parties, with their 
different perspectives on, and interests in the process of identifying the 
organization’s  future  leaders  are  able  to  come  to  a  shared 
understanding  of  which  criteria  are  most  relevant  in  assessments  of 
leadership potential. A study performed by Cook and Emler (1999), for 
instance,  found  that  different  perspectives  on  leadership  potential  (i.e. 
bottom‐up  nomination  versus  top‐down  selection)  influence  the 
outcomes  of  potentiality  assessments.  Dries  and  Pepermans  (2008) 
found  that  HR  managers  and  employees  identified  as  high  potentials 
held  different  opinions  about  what  makes  someone  a  high  potential  or 
not (where HR managers stressed career self‐management and initiative 
taking, the high potentials seemed to believe that they possessed ‘innate’ 
leadership  skills).  Another  study,  by  Guthridge,  Komm  and  Lawson 
(2006) concluded that talent management programs often fail to reach 
their goals – and that CEO’s, business unit leaders and HR professionals 
each blame the other parties for this failure.  
 
Consequently,  the  validity  of  multi‐rater  assessments  of  leadership 
potential  may  be  threatened  by  ‘dissensus’  among  the  parties  involved 
(Remdisch  &  Dionisius,  1998;  Ruderman  &  Ohlott,  1990).  Low  inter‐
rater  agreement  is  often  found  in  appraisals  of  performance  and 
potential. This is especially true for managerial jobs, which tend to have 
ambiguous  performance  standards  and  immeasurable  outcomes 
(Lombardo & McCall, 1982). Although most members of an organization 
will  probably  agree  on  generic  role  behaviors  of  leaders  (e.g.  as  in 
Mintzberg,  1973),  implicit  theories  and  self‐interests  may  cause 
differences in the degree of emphasis each role is given. There are three 
commonly  accepted  explanations  for  the  low  agreement  between  any 
two  raters:  selective  perception,  or  differences  in  the  performance 
information available to different raters (higher levels of agreement are 
found between raters at the same hierarchical relationship to the ratee 
than  between  raters  of  different  levels);  variations  in  criteria  used  by 
individual raters, both in the type used and the importance (or relative 
weight)  attributed  to  them,  according  to  their  implicit  theories  of 
80
leadership  potential;  and  unique  rating  tendency  or  style  (leading  to 
different  types  of  rater  bias,  e.g.  leniency,  halo,  or  range  restriction)  of 
different individual raters (Tsui & Ohlott, 1988).  
 
However, in order for HR policies to contribute to the strategic goals of 
an organization, both the consistency of and the consensus about these 
policies  is  essential.  The  former  refers  both  to  the  consistency  of 
espoused  (i.e.  intended)  and  inferred  (i.e.  perceived)  organizational 
values  and  to  the  consistency  of  all  HR  policies  as  a  whole.  The  latter 
means  that  principle  HRM  decision  makers  in  organizations  have  to 
agree  when  setting  strategic  goals  and  designing  HRM  practices  for 
achieving those goals. A ‘weak’ talent management system, for instance, 
will  not  contribute  optimally  to  the  organization’s  goals  because 
individual  processes  might  dominate,  or  collective  sense‐making  may 
result  in  shared  interpretations  that  may  be  inconsistent  with 
organizational strategic goals (Bowen & Ostroff, 2004).  
 
In  the  second,  quantitative  part  of  our  study,  the  multidimensional 
model of leadership potential developed in the first study was presented 
to  a  sample  of  top  managers,  line  managers  and  HR  managers  (i.e.  the 
main  parties  involved  in  assessments  of  leadership  potential  within 
organizations).  We  examined  the  degree  of  consensus  between  these 
different  parties  in  terms  of  the  relative  weight  they  attached  to 
different  criteria  that  may  be  used  to  assess  leadership  potential.  We 
expected to find differences between the three parties based on the fact 
that  they  each  have  access  to  different  types  of  information  about 
employees’  potential  as  well  as  competing  interests  in  the  process. 
Based  on  the  notion  of  information  availability,  we  expected  line 
managers to focus more on performance and achievements, and on the 
effort displayed on a day to day basis. HR managers, on the other hand, 
might focus more on career aspirations, career planning, strengths and 
weaknesses, and development goals, as it is their mission to collect this 
type  of  information  from  various  sources.  Top  managers,  often  not 
present  in  the  immediate  environment  of  lower‐level  employees,  were 
expected to focus on visibility, assertiveness, networking and charisma, 
as  employees  high  in  these  qualities  are  most  likely  to  draw  their 
attention  (Ruderman  &  Ohlott,  1990).  Competing  interests  of  the 
different  parties  may  also  play  a  role.  One  typical  pitfall  in  talent 
management,  for  instance,  is  that  line  managers  are  often  reluctant  to 
lose their best people to another department. HR managers can be seen 
as internal service providers, who have targets to reach (e.g. in terms of 
how many employees should be identified as high potential each year). 
Top  managers  may  be  intolerant  towards  deviant  profiles  and 
81
(unconsciously)  promote  only  those  who  are  ‘clones’  of  themselves. 
Another possible problem is posed by top managers who feel threatened 
by  upcoming  talent,  fearing  that  their  own  position  is  at  risk  (Cook  & 
Emler, 1999).  
 
Development of the model 
 
Methods 
 
Focus  groups.  In  order  to  reduce  the  elaborate  list  of  leadership 
potential identification criteria found in the literature (see Appendix A) 
into a shorter, more workable one suited for our research purposes, we 
organized  a  focus  group  involving  four  senior  practitioners  and  three 
senior  academics  active  in  the  field  of  human  resources  management 
and  talent  management.  Our  research  goals  were  explicated  at  the 
beginning of the focus group session. During the four‐hour session, the 
545 criteria listed in the original articles and books listed in Appendix A 
were scrutinized in order to assess their relevance for the identification 
of leadership potential. Criteria that were removed involved  those that 
were  ambiguous  in  terms  of  meaning,  and  those  that  were  identical  in 
meaning  to  other  criteria.  Furthermore,  some  criteria  that  were 
categorized as relevant but double or triple barreled (i.e. incorporating 
two  or  three  different  meanings  at  the  same  time),  were  split  up  into 
several singular criteria. In a final step, the terminology of each of the 77 
criteria  that  were  withheld  was  standardized  in  order  to  enhance 
interpretability (see Appendix B).  
 
Q­sorts. The resulting 77 leadership potential identification criteria 
were  printed  onto  separate  numbered  cards  and  sent  by  post,  along 
with an instructions letter, to a selected group of experts from the talent 
management  field.  The  experts  were  instructed  to  sort  all  cards  into 
structured piles or ‘Q‐sorts’. They were told to put two cards in the same 
pile if they were similar in meaning, and in a different pile if they were 
dissimilar (Parker, 2006). Furthermore, they were advised to only have 
piles  of  one  card  if  they  were  absolutely  certain  they  perceived  it  as 
dissimilar from all other cards. After sorting all 77 cards into piles, they 
were instructed to regard each pile as a category that required a label to 
capture the shared meaning of the constructs assigned to it. They were 
then asked to make up a document presenting their categorizations (i.e. 
labels and card numbers per category) and return it to the researchers.  
 
Thirty‐two subject matter experts participated in this Q‐sort study. We 
chose to work with experts in the field of human resources management 
82
and  talent  management  as  they  are  familiar  with  the  terminology.  To 
facilitate diverse perspectives half of the sample were students enrolled 
as  Masters  of  Industrial  and  Organizational  psychology  at  a  large 
Belgian  University  (all  students  had  received  several  courses  on  the 
topic  of  human  resources  management);  the  other  half  were  senior 
human  resources  professionals  specialized  in  talent  management 
(either within their organization or as consultants). Eleven of the SMEXs 
were  women  (34%);  twenty‐one  were  men  (66%).  The  age  of  the 
participants varied between 21 and 61 (m = 41.3; sd = 16.37).  
 
Analysis.  Statistical  as  well  as  qualitative  analyses  were  performed 
on  the  categorizations  that  were  obtained  from  the  Q‐sort  study.  The 
goal of these analyses was to determine the underlying structure of the 
77  identification  criteria  in  the  form  of  an  inclusive  model.  To  achieve 
this,  we  investigated  the  appropriate  graphical  configuration  of  the 
criteria, and more specifically, the amount of underlying dimensions and 
clusters (i.e. ‘regions’) in the data by applying multidimensional scaling 
(MDS). Multidimensional scaling is an exploratory technique that helps 
researchers  determine  the  underlying  structure  in  sets  of  data,  and  is 
considered  particularly  useful  for  the  development  of  theory  (Borg  & 
Groenen,  1997).  As  the  aim  of  MDS  is  to  represent  perceived 
dissimilarities  between  constructs  as  metric  distances  in  an  n‐
dimensional  space,  the  data  from  the  Q‐sort  study  needed  to  be 
transformed  into  a  dissimilarity  matrix  first.  For  each  pair  of  criteria, 
dissimilarities  were  calculated  as  the  total  number  of  experts  that  had 
not sorted them into the same category. It logically follows that values in 
the  obtained  dissimilarity  matrix  varied  between  0  (i.e.  all  experts  put 
constructs x and y together) and 32 (i.e. no expert put constructs x and y 
together).  A  classical  ordinal  MDS  analysis  was  carried  out  using  the 
ALSCAL  algorithm  in  SPSS  15.0.  Subsequently,  adequate  labels  were 
determined for the dimensions and regions through content analysis so 
as to add meaning to the model (Figure 1). For a detailed description of 
how to perform MDS analyses in combination with qualitative (content) 
analyses,  see  Derous,  De  Witte  and  Stroobants  (2003),  and  Dries  et  al. 
(2008).  
 
Results 
 
A  two‐dimensional  model  was  found,  consisting  of  4  regions 
(information  processing  [1],  learning  agility  [2],  drive  [3],  and 
charismatic  leadership  [4])  and  13  subregions:  input‐seeking  behavior 
(1),  insightfulness  (2),  decision  making  (3),  problem  solving  (4), 
willingness  to  learn  (5),  emotional  stability  (6),  personal  adaptability 
83
(7),  drive  for  results  (8),  persistence  (9),  personal  dedication  (10), 
people  management  (11),  inspirational  management  (12),  and 
stakeholder management (13) (see Appendix B).  
 
Interpretation  of  the  dimensions  (see  Figure  1).  While  MDS  assures 
that  constructs  which  are  similar  are  close  on  the  MDS  map,  the 
orientation  of  the  axes  is  an  arbitrary  function  of  the  input  data.  Axe 
rotation  of  the  axes  is  allowed  when  it  enhances  interpretability 
(Garson,  2008).  Figure  1  displays  the  final  model  with  the  dimensions 
rotated. The first, horizontal, dimension was called extrapersonal versus 
intrapersonal  (context  versus  self).  Criteria  at  the  extrapersonal  end  of 
this  dimension  belong  to  regions  1  (information  processing)  and  4 
(charismatic  leadership).  The  extrapersonal  end  of  this  dimension 
stands,  more  specifically,  for  leadership  potential  identification  criteria 
referring to processes where input is taken from  the work context (i.e. 
relevant  others  or  situations)  and  output  is  generated  to  enhance  this 
context  (e.g.  organizational  performance).  We  opted  for  the  term 
‘extrapersonal’,  and  not  the  more  common  ‘interpersonal’,  as  these 
criteria  do  not  only  relate  to  relationships  with  others,  but  also  to 
relationships with the broader work context. Criteria belonging to these 
regions are ‘being open to new and diverse people and ideas’ (i24) and 
‘being  able  to  build  high‐performing  teams’  (i63),  among  others.  The 
other  end  of  the  dimension,  intrapersonal,  refers  to  criteria  relating  to 
processes  aimed  at  enhancing  personal  development  and  functioning. 
Examples  of  criteria  at  this  end  of  the  scale  are  ‘chasing  after  variety, 
challenges  and  intellectual  stimulation’  (i19)  and  ‘showing  drive  and 
perseverance’ (i11).  
 
The  second,  vertical,  dimension  was  labeled  cognition  versus  conation 
(head  versus  heart).  Criteria  on  the  cognition  side  of  the  dimension 
belong  to  regions  1  (information  processing)  and  2  (learning  agility). 
The focus on this end of the dimension is on information processing by 
the  individual  (i.e.  comprehension,  inferencing,  decision  making, 
planning  and  learning).  This  is  illustrated  by  the  criteria  belonging  to 
these regions, e.g. ‘reflecting critically on practices and procedures’ (i46) 
and  ‘enjoying  complex  first‐time  problems  and  challenges  associated 
with  new  experiences’  (i28).  Criteria  on  the  conation  side  of  this 
dimension belonged to regions 3 (drive) and 4 (charismatic leadership). 
Examples of criteria at this end of the scale are ‘demonstrating need for 
achievement  (i.e.  being  performance‐oriented)’  (i33)  and  ‘actively 
looking  for  opportunities  to  lead’  (i66).  Therefore,  the  conation  side 
illustrates  a  focus  on  will,  drive  and  motivation,  i.e.  the  powers  that 
direct effort.  
 

 
84

Figure 1. MDS model of all 77 criteria mapped onto 2 dimensions and into 4 regions (dotted lines represent axe rotations) 
85
Assessment of consensus about the model 
 
In order to test our expectations about possible ‘dissensus’ between the 
parties  involved  in  assessments  of  leadership  potential  formulated  in 
the  Introduction  of  this  paper,  we  conducted  a  second  study  using  an 
online survey.  
 
Methods 
 
Measures  and  procedures.  An  online  survey  was  conducted  among 
professionals  engaged  in  assessments  of  leadership  potential.  Three 
groups of respondents were contacted, i.e. top managers, line managers 
and  HR  managers.  Contact  data  came  from  a  large  database  from  the 
research  department  supporting  the  research.  Respondents  were  also 
asked to forward the survey to other professionals active in the field of 
talent  management  within  their  networks  (i.e.  snowball  sampling). 
Respondents  were  instructed  to  indicate  for  each  of  the  77  leadership 
potential  identification  criteria  to  which  extent  (on  a  7‐point  scale 
ranging  from  ‘not  at  all’  to  ‘to  a  very  large  extent’)  they  consider  these 
criteria  as  essential  when  they  themselves  are  asked  for  input  in 
assessments  of  leadership  potential.  Respondents  were  instructed  to 
answer  according  to  their  own  personal  experiences  with  assessments 
of leadership potential and not rely (exclusively) on formalized criteria 
imposed  by  the  organization.  Finally,  respondents  were  asked  to 
indicate  their  gender,  age,  educational  level  and  role  in  their 
organization’s processes concerning assessment of leadership potential 
(roles adapted from Dries & Pepermans, 2008, see Table 1).  
 
Participants.  A  total  of  179  respondents  took  part  in  the  online 
survey:  52  top  managers  (29%),  54  line  managers  (30%)  and  73  HR 
managers (41%). Of these 179 respondents, 60 were women (34%) and 
119  were  men  (66%).  The  age  of  the  respondents  varied  between  23 
and  65  (m  =  44.37;  sd  =  8.25).  As  regards  educational  level,  8%  of 
respondents reported having a high school degree, 23% had a bachelor’s 
degree,  55%  had  a  master’s  degree  and  14%  obtained  a  post‐graduate 
degree.  
 
Table 1 illustrates how top managers, line managers and HR managers 
typically take up different roles within their organizations’ assessments 
of  leadership  potential  (significant  differences  were  discerned  using  χ² 
difference  tests).  Top  managers  in  our  sample  were  more  involved  in 
providing top‐down input on leadership identification policies, and had 
more  decision‐making  capacity  than  line  managers  and  HR  managers.
Table 1. Crosstabs and chi‐square difference tests for the different parties and their roles in assessments of leadership potential 
 
  Top managers  Line managers  HR managers   
(n = 52)  (n = 54)  (n = 73) 
Role in assessments of leadership potential  n  %   n  %   n   %   χ² 
Providing bottom‐up input to management about 
25  48%  40  74%  31  43%  13.38* 
the leadership potential observed in employees 
Providing top‐down input to different people in the 
22  42%  11  20%  26  36%  6.16† 
organization about how to identify leadership potential 
Developing criteria and processes for the identification 
23  44%  10  19%  53  73%  36.80* 
of leadership potential 
Participating in a committee that decides who is identified 
32  62%  21  39%  32  44%  6.11† 
as a potential future leader and who is not 
Having decision‐making capacity (or veto right) about who 
22  42%  9  17%  11  15%  14.54* 
is identified as a potential future leader and who is not 
Notes. % = within‐group percentages; † p < .05; * p < .01.  
86

 
 
87
Their main role seems to be to participating in committees that evaluate 
employees’ leadership potential. Line managers reported to be involved 
mainly in terms of providing bottom‐up information about the potential 
of employees, and did so significantly more than top managers and HR 
managers.  HR  managers’  main  role  was  developing  criteria  and 
processes for assessing leadership potential.  
 
Results 
 
Table  2  offers  an  overview  of  the  means,  standard  deviations  and 
correlations  for  the  entire  sample.  The  correlation  analysis  indicates 
that  all  regions  and  subregions  of  our  developed  model  of  leadership 
potential  correlate  at  the  p  <  .01  level,  with  correlation  coefficients 
around .6.  
 
Factor  structure  and  internal  consistency.  Confirmatory  factor 
analysis  was  performed  using  structural  equation  modeling  (Lisrel 
8.80). Table 3 reports the goodness‐of‐fit statistics for each region. Cut‐
offs  to  assess  goodness‐of‐fit  are:  CFI  must  be  equal  to  or  higher  than 
.95; NNFI must be equal to or higher than .95; and RMSEA must be equal 
to or smaller than .08 (Kline, 2005). Looking at the values in Table 3 we 
can thus conclude that all four regions display good fit with the data.  
 
The  Cronbach’s  alpha  values  per  region  and  subregion  (which  indicate 
internal  consistency)  can  be  found  in  Appendix  B.  The  alphas  for  the 
four  regions  are  high  (above  .70);  for  the  subregions  however,  not  all 
values are above .70 (e.g. for emotional stability) – in interpreting these 
values,  however,  we  must  take  into  account  that  higher  numbers  of 
items in a scale lead to higher alphas (Pallant, 2001).  
 
Consensus  between  parties  involved  in  assessments  of  leadership 
potential.  We  performed  several  ANOVAs  to  test  our  hypotheses  about 
possible  differences  between  top  managers,  line  managers  and  human 
resources managers in terms of the  criteria they use in assessments of 
leadership potential.  
 
However, we did not find any significant differences for the four regions 
information  processing  (F(2,176)  =  .18,ns),  learning  agility  (F(2,176)  = 
.20, ns), drive (F(2,176) = .13, ns) and charismatic leadership (F(2,176) = 
.40,  ns).  We also  performed  ANOVAs  on  each  of  the  13  subregions,  but 
did  not  find  any  significant  differences.  Consequently,  all  expectations 
on possible between‐group differences were discarded.  
88
Table 2. Means, standard deviations and correlations (n = 179) 

  m sd [1]  [2]  [3]  [4]  1 


     Regions     
[1] Information processing 5.74 .61 ‐‐  
[2] Learning agility 5.52 .60 .80* ‐‐  
[3] Drive  5.57 .61 .82* .81* ‐‐  
[4] Charismatic leadership 5.42 .71 .79* .72* .76* ‐‐  
   
   Subregions   
1 Input‐seeking behavior 5.88 .74 .80* .74* .66* .64* ‐‐ 
2 Insightfulness  5.77 .68 .86* .70* .68* .70* .58* 
3 Decision making 5.46 .83 .85* .65* .74* .74* .52* 
4 Problem solving 5.86 .69 .82* .56* .63* .53* .52* 
5 Willingness to learn 5.60 .70 .55* .82* .59* .50* .60* 
6 Emotional stability 5.52 .68 .79* .87* .77* .70* .70* 
7 Personal adaptability 5.44 .73 .72* .88* .73* .68* .62* 
8 Drive for results 5.41 .72 .69* .65* .84* .70* .48* 
9 Persistence  5.63 .78 .68* .68* .85* .56* .60* 
10 Personal dedication 5.66 .62 .74* .76* .88* .71* .61* 
11 People management 5.63 .72 .73* .65* .68* .87* .62* 
12 Inspirational management 5.19 .79 .72* .67* .75* .93* .54* 
13 Stakeholder management 5.46 .84 .69* 63* .65* .91* .58* 
* p < .01 

Discussion 
 
Discussion of the proposed model of leadership potential 
 
The  first  aim  of  the  current  paper  was  to  develop  a  multidimensional 
model  of  ‘the’  criteria  that  are  considered  essential  to  the  early 
identification of leadership potential. A model with four quadrants was 
decided  upon  as  relevant  for  categorizing  the  77  different  criteria 
associated  with  leadership  potential  obtained  from  the  two  studies 
presented in this paper: 
 
(1) cognition­extrapersonal (head­context). Leadership potential criteria 
belonging to this quadrant relate to the ability to extract input from, and 
generate impact on the context within which one functions; the domain 
within  which  these  criteria  are  situated  is  the  cognitive  domain, 
indicating  that  mental  ability  (and  adaptability)  and  information‐
processing capacities are emphasized.  
89
Table 2. Continued  

2  3  4  5  6  7  8  9  10  11  12  13 


       
       
       
       
       
       
       
       
‐‐       
.67*  ‐‐     
.64*  .60*  ‐‐    
.53*  .34*  .38* ‐‐    
.63*  .68*  .60* .54* ‐‐    
.64*  .66*  .47* .57* .70* ‐‐    
.60*  .67*  .53* .48* .56* .63* ‐‐   
.53*  .59*  .53* .46* .70* .59* .51*  ‐‐  
.61*  .65*  .57* .59* .71* .65* .66*  .64* ‐‐  
.63*  .67*  .50* .38* .66* .62* .62*  .49* .65* ‐‐  
.65*  .71*  .48* .44* .66* .62* .72*  .54* .68* .73* ‐‐  
.61*  .62*  .46* .42* .59* .61* .58*  .50* .60* .67* .78* ‐‐ 
* p < .01 

 
Table 3. Goodness‐of‐fit statistics for the four factors of the proposed model  

  Factor 1:  Factor 2:  Factor 3:  Factor 4: 


Information  Learning  Drive  Charismatic 
processing  agility  leadership 
χ²  96.94  253.57  265.26  475.20 
df   84  132  149  272 
p   .16  <.01  <.01  <.01 
RMSEA  .03  .07  .07  .07 
CFI  1.00  .96  .96  .98 
NNFI  .99  .95  .96  .97 
Notes. RMSEA = Root Mean Square Error of Approximation; CFI = Comparative Fit Index; 
NNFI = Non‐Normed Fit Index.  
 
This quadrant encompasses the region information processing, compiled 
of  four  distinct  subregions  of  meaning:  input‐seeking  behavior  (i.e. 
potential in terms of being open to diverse sources of information about 
self  and  situations);  insightfulness  (i.e.  potential  in  terms  of  being  able 
to process information gathered from various sources); decision making 
(i.e. potential in terms of being able to make proper decisions based on 
the  information  available);  and  problem  solving  (i.e.  potential  in  terms 
90
of  being  able  to  solve  the  problems  that  present  themselves  based  on 
the information available); 
 
(2)  cognition­intrapersonal  (head­self).  Leadership  potential  criteria 
belonging to this quadrant refer to the exploration of personal goals and 
the direction of energy towards meeting those goals; the domain within 
which these criteria are situated is the cognitive domain, indicating that 
mental ability (and adaptability) and information‐processing capacities 
are emphasized. This quadrant encompasses the region learning agility, 
compiled  of  three  distinct  subregions  of  meaning:  willingness  to  learn 
(i.e.  potential  in  terms  of  actively  looking  for  novel  experiences  that 
enhance  learning);  emotional  stability  (i.e.  potential  in  terms  of 
maintaining  a stable self‐concept even in stressful or novel situations); 
and personal adaptability (i.e. potential in terms of being open to change 
when novel circumstances require it); 
 
 (3)  conation­intrapersonal  (heart­self).  Leadership  potential  criteria 
belonging to this quadrant refer to the exploration of personal goals and 
the direction of energy towards meeting those goals; the domain within 
which these criteria are situated is the conative domain, indicating that 
motivation,  drive  and  willpower  are  emphasized.  This  quadrant 
encompasses  the  region  drive,  compiled  of  three  distinct  subregions  of 
meaning: drive for results (i.e. potential in terms of being motivated to 
consistently  deliver  high‐quality  results);  persistence  (i.e.  potential  in 
terms of maintaining high energy levels even in difficult circumstances); 
and personal dedication (i.e. potential in terms of displaying a deep and 
intrinsic commitment to relevant goals); 
 
(4) conation­extrapersonal (heart­context). Leadership potential criteria 
belonging to this quadrant relate to the ability to extract input from, and 
generate impact on the context within which one functions; the domain 
within  which  these  criteria  are  situated  is  the  conative  domain, 
indicating  that  motivation,  drive  and  willpower  are  emphasized.  This 
quadrant  encompasses  the  region  charismatic  leadership,  compiled  of 
three distinct subregions of meaning: people management (i.e. potential 
in  terms  of  being  able  to  mobilize  others  towards  a  shared  goal); 
inspirational  management  (i.e.  potential  in  terms  of  having  a  strong 
personal  mission  statement);  and  stakeholder  management  (i.e. 
potential  in  terms  of  being  able  to  identify  relevant  stakeholders  and 
optimize interactions with them).  
 
 
91
Links with existing models of leadership potential. When putting our 
two‐dimensional  (cognition‐conation  versus  extrapersonal‐
intrapersonal)  model  of  leadership  potential  side  by  side  with  the 
criteria found in earlier studies, we see that our proposed model seems 
to  be  rather  inclusive,  in  that  it  incorporates  many  of  the  themes 
identified in earlier research as relevant to the assessment of leadership 
potential.  In  order  to  evaluate  our  model’s  added  value,  however,  we 
will  focus  on  the  leadership  potential  assessment  instruments 
developed by Lombardo & Eichinger (2000) and Spreitzer et al. (1997) 
specifically,  as they were developed  with the exact same  goals  in mind 
as  was  our  model.  Both  instruments,  theoretically,  appear  to  be  aimed 
mainly at assessing the ability to learn from experience – which would, 
logically,  connect  them  most  to  the  subregion  ‘learning  agility’  in  our 
proposed model of learning potential. However, scrutinizing the Choices 
instrument  (Lombardo  &  Eichinger,  2000)  and  the  Prospector 
instrument (Spreitzer et al., 1997) in more detail, we see that the items 
incorporated in these instruments are actually quite diverse. Lombardo 
and  Eichinger  (2000,  p.  324)  discerned  four  factors  within  the Choices 
instrument:  People  Agility,  Results  Agility,  Mental  Agility  and  Change 
Agility.  Inspecting  the  descriptions  of  the  factors  and  the  sample  items 
(p.  325),  it  seems  that  People  Agility  corresponds  mainly  to  our 
conceptions  of  emotional  stability  and  people  management;  Results 
Agility to drive for results; Mental Agility to input‐seeking behavior and 
insightfulness; and Change Agility to personal adaptability in our model. 
The  Prospector  instrument  (Spreitzer  et  al.,  1997)  incorporates  two 
major  themes:  learning‐oriented  themes  and  end‐state  competencies  – 
the authors’ goal was to investigate the incremental predictive power of 
the  ability  to  learn  from  experience  above  and  beyond  the  more 
‘traditional’  end‐state  competencies  (p.  7).  Whereas  their  learning‐
oriented  themes  correspond  mainly  to  the  region  of  learning  agility  in 
our model, and somewhat to the subregion input‐seeking behavior, the 
end‐state competency themes were somewhat more difficult to match as 
they  were  very  diverse  (see  Spreitzer  et  al.,  p.  13).  However,  it  is 
remarkable that in both the Lombardo & Eichinger (2000) model and in 
the Spreitzer et al. (1997) model the element of charismatic leadership 
is  largely  absent.  A  possible  explanation  for  this  is  that  both  groups  of 
authors  wish  to  counter  the  notion  that  leadership  potential  should  be 
assessed by evaluating leadership skills – in their view, it does not make 
sense to expect junior staff to exhibit (be it seminal forms of) leadership 
behavior  at  their  level  already.  Nonetheless,  the  experts  taking  part  in 
the  two  studies  reported  in  this  paper  did  identify  charismatic 
leadership criteria as essential to the assessment of leadership potential. 
The  question  remains,  however,  if  these  criteria  are  indeed  observable 
92
in  junior  staff  who  have  perhaps  not  been  able  to  develop  or 
demonstrate such behaviors yet. We believe that although it may not be 
possible for junior staff to display all of these behaviors at their level yet, 
it is important that they demonstrate the commitment and motivation to 
become leaders in the future. Thus, the orientation towards a leadership 
career path does emerge as a crucial aspect in assessments of leadership 
potential  –  a  point  we  will  elaborate  on  further  in  the  section  on 
Implications for practice.  
 
Consensus  between  parties  involved  in  assessments  of  leadership 
potential.  Contrary  to  the  expected  differences  between  top  managers, 
line managers and HR managers in terms of which criteria of leadership 
potential  they  valued  most,  we  found  that  the  three  parties  displayed 
consensus, in that they all rated all 77 criteria in our model of leadership 
potential as important (i.e. over 5 on a 7‐point scale, with no significant 
differences). A possible explanation for this finding is that the 77 criteria 
that were withheld in the final model were those that were considered 
(by  the  experts  in  the  qualitative  part  of  our  study)  as  most 
differentiating  in  terms  of  leadership  potential.  If  this  list  of  criteria, 
indeed, exists only from criteria that are generally considered important 
in  the  field,  the  consensus  among  the  three  parties  involved  in 
assessments of leadership potential is not especially surprising. It would 
have perhaps been interesting to also incorporate less relevant criteria 
to  see  if  these  would  be  rated  lower  (e.g.  to  counter  response  bias). 
However,  the  findings  in  the  quantitative  part  of  the  study  appear  to 
support  the  model  developed  in  that  all  three  parties  agree  that  these 
criteria are essential to the assessment of leadership potential. Tsui and 
Ohlott (1988) also saw their hypotheses on dissensus about managerial 
effectiveness  rejected  –  their  conclusion  was  similar  to  ours,  i.e.  that 
their model contained only relevant and generalizable criteria.  
 
Limitations 
 
First of all, essential to our methodology of using focus groups, Q‐sorts 
and multidimensional scaling to arrive at our final model of leadership 
potential,  is  the  premise  of  (subjective)  information  reduction.  We 
cannot  tell  with  certainty  if  the  model  is  complete;  furthermore,  we 
cannot  be  sure  whether  different  samples  of  experts  would  have 
generated  an  identical  model  based  on  the  original  literature.  Also,  the 
analyses  leading  to  the  final  model  in  terms  of  deciding  on 
dimensionality, the division of criteria into regions and subregions, and 
the process of attributing meaning through labels were qualitative  and 
subjective  –  implying  that  other  researchers  would  possibly  have  gone 
93
about  differently  in  constructing  the  model.  However,  we  have 
consistently  used  input  from  multiple  sources  and  experts  in  order  to 
objectify  the  outcomes;  it  is  not  so  that  the  findings  reported  in  this 
paper are merely the result of the reasoning of two authors.  
 
Second,  in  this  paper  we  can  not  offer  a  ‘how‐to’  manual  for 
organizations with regard to high potential identification. Our model is 
meant to be descriptive rather than normative, in the sense that our aim 
was  to  describe  which  criteria  are  considered  essential  for  the 
identification  of  leadership  potential  by  (academic  and  practitioner) 
experts, rather than prescribe which criteria should be used. In order to 
do  so,  longitudinal  follow‐up  of  performance  and  advancement 
parameters  of  employees  identified  as  high  potentials  based  on  these 
criteria  would  be  necessary.  However,  this  may  be  a  task  better  suited 
for  management  consultants  than  for  researchers,  as  an  organization‐
specific interpretation of the model will need to be implemented.  
 
Third,  as  with  all  single‐standing  studies,  variables  from  the  research 
context  (i.e.  culture,  language,  sample  selection,  researchers’  beliefs) 
have  undoubtedly  affected  our  model.  Nonetheless,  this  paper  is  the 
result  of  a  substantial  effort  to  collect,  objectify  and  integrate  as  much 
relevant  information  as  possible  on  (the  consensus  about)  leadership 
potential identification criteria.  
 
Finally, although we specifically instructed respondents (on each page of 
the  online  survey)  in  the  survey  study  to  answer  according  to  their 
actual (individual) assessment behavior, it is possible that some form of 
social  desirability  bias  influenced  the  data.  It  is  possible  that 
respondents  were  unwilling  to  admit  that  external  circumstances  (i.e. 
concerns  of  time,  effort,  and  structure)  hinder  them  to  be  able  to 
observe and integrate all 77 criteria into their assessments of leadership 
potential.  The  only  way  to  avoid  such  effects  would  probably  be  to 
conduct participant observation studies instead of survey studies.  
 
Implications 
 
The  significant  overlap  between  our  model  of  leadership  potential  and 
earlier  ones  (Lombardo  &  Eichinger,  2000;  Spreitzer  et  al.,  1997)  does 
not  diminish  the  value  of  each  of  them  separately  –  rather,  it  appears 
that  regardless  of  the  research  methods  or  the  country  where  the 
studies  were  conducted,  several  aspects  of  leadership  potential  are 
repeatedly  pointed  out  as  essential  by  authors  and  practitioners  alike. 
Nonetheless, in order for organizations to develop optimal, ‘customized’ 
94
talent  management  processes  and,  more  specifically,  leadership  tracks, 
contingency  factors  such  as  organizational  strategy,  culture 
(organizational  as  well  as  national)  and  timeframe  should  also  play  an 
important  role.  As  such,  ‘next  practice’  in  terms  of  high  potential 
identification should involve combining generic as well as organization‐
specific  identification  criteria,  end‐state  competencies  as  well  as 
learning  agility  and  a  career  orientation  towards  leadership.  Studies 
separating  generic  criteria  (if  there  are  in  fact  truly  generic  criteria) 
from  aspects  that  should  be  tailored  to  fit  specific  organizational, 
cultural and temporal contexts, would greatly improve knowledge about 
the  topic,  and  contribute  to  assessments  of  leadership  potential  taking 
place in practice. 
 
Another suggestion for further research relates to developability of the 
criteria used in assessments of leadership potential. Research indicates 
that  the  exploitation  of  ‘non‐traditional’  talent  pools  (i.e.  talent  pools 
made  up  of  people  with  lower  educational  levels  and  minority  groups 
such  as  low‐skilled  immigrants,  elderly  employees  and  people  with 
disabilities)  will  be  necessary  to  compete  in  the  ‘war  for  talent’  in  the 
(near)  future.  The  current  demographic  trends,  as  well  as  the  shifts  in 
psychological  contracts  of  employees,  are  causing  a  de‐emphasis  on 
‘elite’ groups on the labor market – and thus, a de‐emphasis on ‘innate’ 
superior competence (Bossuyt & Dries, 2008).  
 
We  wish  to  stress  the  importance  of  incorporating  information  on 
leadership  orientation  when  assessing  leadership  potential.  All  too 
often,  assessments  of  leadership  potential  focus  more  on  abilities  than 
on  orientation,  causing  high‐performers  to  be  ‘forced’  into  leadership 
tracks, while they may prefer a lifestyle with more work‐life balance, or 
a  technical  track  that  does  not  involve  people  management.  Most 
organizational cultures seem to restrict esteem to those who climb the 
hierarchical  ladder.  Therefore,  in  order  to  advance  career‐wise, 
individuals with a strong technical competence career anchor may move 
upwards  hierarchically  to  management  or  supervisory  positions  with 
disastrous consequences for both self and organization as a result of the 
misfit (Garavan & Coolahan, 1996). 
 
Another  important  message  in  the  current  paper  is  that  top 
management,  HR  and  line  management  (and  possibly  yet  other  parties 
within  organizations)  need  to  have  a  shared  opinion,  departing  from 
organizational strategy, on why assessments of leadership potential are 
important  and  which  criteria  to  use  when  conducting  these 
assessments. Only by ensuring that all parties share the same strategic 
95
vision, a strong impact on the organization’s competitive advantage is to 
be  expected  (Bowen  &  Ostroff,  2004).  Nonetheless,  the  different  roles 
and perspectives of the different parties taking part in the process need 
to be recognized. A thorough analysis of the different perspectives and 
interests  of  the  different  parties,  as  well  as  the  different  types  of 
information available to each of them, could provide valuable input for 
future  assessments  of  leadership  potential.  Based  on  such  analyses,  an 
optimal  division  of  roles  within  the  assessment  process  could  be 
established.  Furthermore,  it  could  also  improve  the  consolidation  of 
information  on  employees’  leadership  potential  gathered  from  various 
sources; research indicates that very few organizations to date succeed 
in  establishing  a  workable  database  for  talent  management  purposes 
(Bossut & Dries, 2008). 
 
References 
 
Baruch,  Y.  &  Peiperl,  M.  (1997).  High‐flyers:  Glorious  past,  gloomy  present,  any  future?  Career 
Development International, 2 (7), 354‐358. 
Borg, I. & Groenen, P. (1997). Modern multidimensional scaling: Theory and applications. New York, 
NY: Springer.  
Bossuyt, T. & Dries, N. (2008). Talent management en flexibele loopbaanpaden voor de werknemers 
van morgen [Talent management and flexible career tracks for tomorrow's workforce]. In H. 
Calmeyn, De Witte, K. and Weverbergh, J. (Eds.), Licht op leren 2008: Leren en ontwikkelen in 
een  talentgerichte  maatschappij  [Spotlight  on  learning  2008:  Learning  and  developing  in  a 
talent­oriented society]. Leuven, Belgium: LannooCampus.  
Boudreau,  J.W.  &  Ramstad,  P.M.  (2005).  Where’s  your  pivotal  talent?  Harvard  Business  Review,  83 
(4), 23‐24. 
Bournois, F. & Rousillon, S. (1998). Préparer les dirigeants de demain. Une approche internationale 
de  la  gestion  des  cadres  à  haut  potentiel  [Preparing  tomorrow’s  leaders.  An  international 
approach to managing high potentials]. Paris, France: Editions d’Organizations.  
Bowen,  D.E.  &  Ostroff,  C.  (2004).  Understanding  HRM‐firm  performance  linkages:  The  role  of  the 
"strength" of the HRM system. Academy of Management Review, 29 (2), 203‐221. 
Briscoe, J.P.  & Hall, D.T. (1999).  Grooming and picking leaders using  competency frameworks: Do 
they  work?  An  alternative  approach  and  new  guidelines  for  practice.  Organizational 
Dynamics, 28 (2), 37‐52.  
Buckingham,  M.  &  Vosburgh,  R.M.  (2001).  The  21st  century  Human  Resources  function:  It's  the 
talent, stupid! Human Resource Planning, 24 (4), 17‐23. 
Capowski,  G.  (1994).  Anatomy  of  a  leader:  Where  are  the  leaders  of  tomorrow?  Management 
Review, 84, 10‐17. 
Cope, F. (1998). Current issues in selecting high potentials. Human Resource Planning, 21 (3), 15‐17. 
Cook,  T.  &  Emler,  N.  (1999).  Bottom‐up  versus  top‐down  evaluations  of  candidates'  managerial 
potential: An experimental study. Journal  of Occupational and Organizational Psychology, 72 
(4), 423‐439.  
Derous,  E.,  De  Witte,  K.  &  Stroobants,  R.  (2003).  Testing  the  social  process  model  on  selection 
through expert analysis. Journal of Occupational and Organizational Psychology, 76, 179‐199. 
Dries,  N.  &  Pepermans,  R.  (2008).  "Real"  high  potential  careers:  An  empirical  study  into  the 
perspectives of organizations and high potentials. Personnel Review, 37 (1), 85‐108. 
Dries,  N.,  Pepermans,  R.  &  Carlier,  O.  (2008).  Career  success:  Constructing  a  multidimensional 
model. Journal of Vocational Behavior, 73 (2), 253‐267.  
Eby,  L.T.,  Allen,  T.D.  &  Brinley,  A.  (2005).  A  cross‐level  investigation  of  the  relationship  between 
career  management  practices  and  career‐related  attitudes.  Group  and  Organization 
Management, 30 (6), 565‐596. 
Eichinger, R.W. & Lombardo, M.M. (2004). Learning agility as a prime indicator of potential. Human 
Resource Planning, 27 (4), 12‐15.  
96
Feild,  H.S.  &  Harris,  S.G.  (1991).  Participants'  frustrations  in  fast‐track  development  programmes. 
Leadership and Organization Development Journal, 12 (4), 3‐8. 
Frank,  F.D.  &  Taylor,  C.R.  (2004).  Talent  management:  Trends  that  will  shape  the  future.  Human 
Resource Planning, 27 (1), 33‐41.  
Ford, M. (2005). Let talent bloom: Cultivate innovation by moving high potential employees around, 
up, and even out of your IT organization. CIO Magazine, 18 (19), 26‐29.  
Garavan,  T.N.  &  Coolahan,  M.  (1996).  Career  mobility  in  organizations:  implications  for  career 
development. Journal of European Industrial Training, 20 (4), 30‐40.   
Garson,  G.D.  (2008).  Statnotes:  Multidimensional  scaling.  Available  at: 
http://www2.chass.ncsu.edu/garson/pa765/statnote.htm (accessed October 6th, 2008). 
Groysberg,  B.,  Nanda,  A.  &  Nohria,  N.  (2004).  The  risky  business  of  hiring  stars.  Harvard  Business 
Review, 82 (5), 92‐100. 
Guthridge,  M.,  Komm,  A.B.  &  Lawson,  E.  (2006).  The  people  problem  in  talent  management. 
McKinsey Quarterly, 2, 6‐8.  
Hall, D.T. & Seibert, K.W. (1992). Strategic management development: linking strategy, succession 
planning,  and  managerial  learning.  In  D.H.  Montross  and  C.J.  Shinkman    (Eds.),  Career 
Development: Theory and Practice. Springfield, IL: Charles C. Thomas Press. 
Hollenbeck, G.P., McCall, M.W. & Silzer, R.F. (2006).  Theoretical and practitioner  letters: Leadership 
competency models. The Leadership Quarterly, 17, 398‐413.  
Iles,  P.  (1997).  Sustainable  high  potential  career  development:  A  resource‐based  view.  Career 
Development International, 2 (7), 347‐353.  
Jerusalim,  R.S.  &  Hausdorf,  P.A.    (2007).  Managers'  justice  perceptions  of  high  potential 
identification practices. Journal of Management Development, 26 (10), 933‐950.  
Kruskal, J.B. & Wish, M. (1978). Multidimensional scaling. Beverly Hills, CA: Sage. 
Kuznia,  K.  (2004).  The  Rhetoric  and  Reality  of  Fast  Track  Management  Development  Programs. 
Available  at:  http://www.midwestacademy.org/Proceedings/2004/papers/Kuznia.doc 
(Accessed October 6th, 2008).  
Lawler,  E.E.  &  Mohrman,  S.A.  (2003).  Creating  a  strategic  Human  Resources  organization:  An 
assessment of trends and new directions. Stanford, CA: Stanford University Press.  
Lewis,  R.E.  &  Heckman,  R.J.  (2006).  Talent  management:  A  critical  review.  Human  Resource 
Management Review, 16, 139‐154.  
Lombardo,  M.M.  &  Eichinger,  R.W.  (2000).  High  potentials  as  high  learners.  Human  Resource 
Management, 39 (4), 321‐330.  
Lombardo,  M.M.  &  McCall,  M.W.  (1982).  Leader  on  line:  Observations  from  a  simulation  of 
managerial  work.  In  J.G.  Hunt,  U.  Sekaran  and  C.A.  Schriesheim  (Eds.),  Leadership:  Beyond 
establishment views (pp. 50­67). Carbondale, IL: Southern Illinois University Press.  
Mayo, A. (2000). The role of employee development in the growth of intellectual capital. Personnel 
Review, 29 (4), 521‐533.  
McCall, M.W. (1988). Developing executives through work experiences. Human Resource Planning, 
11 (1), 1‐11.  
McCall, M.W. (1992). Executive development as a business strategy. Journal of Business Strategy, 13 
(1), 25‐31.  
Miller, J. (2006). Attend to the talent: The right team  can be the secret to success. CA Magazine, 139 
(1), 49‐50. 
Mintzberg, H. (1973). The nature of managerial work. New York, NY: Harper & Row.  
Parker  P.  (2006).  Card  sorts:  A  process  of  constructing  personal  narrative.  In  M.  McMahon,  &  W. 
Patton  (Eds.).  Career  Counselling:  Constructivist  approaches  (pp.  176­186).  New  York,  NY: 
Routledge. 
Pepermans,  R.,  Vloeberghs,  D.  &  Perkisas,  B.  (2003).  High  potential  identification  policies:  An 
empirical study among Belgian companies. Journal of Management Development, 22 (8), 660‐
678. 
Remdisch, S. & Dionisius, S. (1998). Training supervisors’ assessment skills – A crucial step in high 
potential selection. European Journal of Work and Organizational Psychology, 7 (4), 559‐564.  
Ruderman,  M.N.  &  Ohlott,  P.J.  (1990).  Traps  and  pitfalls  in  the  judgment  of  executive  potential. 
Center for creative leadership, report 141, 1‐34.  
Spence, I. & Ogilvie, J.C. (1973). A table of expected stress values for random rankings in nonmetric 
multidimensional scaling. Multivariate Behavioral Research, 8, 511‐517. 
Spreitzer, G.M., McCall, M.W. & Mahoney, J.D. (1997). Early identification of international executive 
potential. Journal of Applied Psychology, 82 (1), 6‐29. 
97
Tsui,  A.S.  &  Ohlott,  P.  (1988).  Multiple  assessment  of  managerial  effectiveness:  Interrater 
agreement and consensus in effectiveness models. Personnel Psychology, 41, 779‐803.  
Tulgan, B. (2001). Winning the talent wars. Employment Relations Today, 23 (1), 104.  
Wagenaar,  W.A.  &  Padmos,  P.  (1971).  Quantitative  interpretation  of  stress  in  Kruskal’s 
multidimensional  scaling  technique.  British  Journal  of  Mathematical  and  Statistical 
Psychology, 24, 101‐110. 
Appendix A. Final set of relevant publications selected from the literature (1986‐2008) 
 
Year  Author(s)  Publication   
1986  Kovach   The derailment of fast track managers. Organizational Dynamics, 15 (2), 41‐48.
1986  Sternberg  Introduction: The nature and scope of practical intelligence. In R. J. Sternberg & R. K. Wagner (Eds.), 
Practical intelligence: Nature and origins of competence in the everyday world (pp. 1–10).  
New York, NY: Cambridge University Press. 
1987  Gardner  Leadership development. Leadership papers, 7. Washington, DC: Independent Sector.
1987  Kouzes & Posner  The Leadership Challenge. San Francisco, CA: Jossey‐Bass Publishers. 
1988  Cox & Cooper   High Flyers: An Anatomy of Managerial Success. Oxford, UK: Blackwell Publishing.
1988  Derr, Jones & Toomey  Managing high potential employees: Current practices in thirty‐three US corporations. 
Human Resource Management, 27 (3), 273‐290. 
1988  Kotter  The leadership factor. New York, NY: Free Press.
1988  Lombardo, McCauley &  Explanations of success and derailment in upper‐level management positions. 
Ruderman  Journal of Business and Psychology, 2 (3), 199‐216. 
1988  McCall  Developing executives through work experiences. Human Resource Planning, 11 (1), 1‐11.
98

1989  Gritzmacher  Staying competitive through strategic management of fast‐track employees. 


National Productivity Review, 8 (4), 421‐432. 
1989  Pearson  Six basics for general managers. Harvard Business Review, 67 (4), 94‐101. 
1991  Barham & Oates  The International Manager. London, UK: Business Books.
1991  Hartman, Griffeth, Crino &  Gender‐based influences: The promotion recommendation. Sex Roles, 25, 285–300.
Harris 
1992  McCall   Executive development as a business strategy. Journal of Business Strategy, 13 (1), 25‐31.
1994  Capowski  Anatomy of a leader: Where are the leaders of tomorrow? Management Review, 84, 10‐17.
1994  McCall  Identifying leadership potential in future international executives: developing a concept. 
Consulting Psychology Journal, 46 (1), 49‐63. 
1995  Cockerill, Hunt & Schroder  Managerial competencies: Fact or fiction? Business Strategy Review, 6 (3), 1‐12.
1997  Smith & Peters  Action learning: worth a closer look. Ivey Business Quarterly, 62 (1), 63‐67. 
Appendix A. Continued 
 
Year  Author(s)  Publication   
1997  Spreitzer, McCall &  Early identification of international executive potential. Journal of Applied Psychology, 82 (1), 6‐29.  
Mahoney 
1998  McCredie & Shackleton  The unit general manager: A competency profile. Personnel Review, 29 (1), 106–114.
 
1999  Briscoe & Hall  Grooming and picking leaders using competency frameworks: Do they work? An alternative   
approach and new guidelines for practice. Organizational Dynamics, 28 (2), 37‐52. 
1999  Cook & Emler  Bottom‐up versus top‐down evaluations of candidates' managerial potential: An experimental study. 
Journal of Occupational and Organizational Psychology, 72 (4), 423‐439. 
1999  Dulewicz & Herbert  Predicting advancement to senior management from competencies and personality data: A seven‐
year follow‐up study. British Journal of Management, 10 (1), 13‐22. 
1999  Robertson, Gibbons,  Understanding management performance. British Journal of Management, 10, 5‐12.
Baron, MacIver & Nyfield 
2000  Lombardo & Eichinger  High potentials as high learners. Human Resource Management, 39 (4), 321‐330.
2000  Robertson, Baron,  Conscientiousness and managerial performance. Journal of Occupational and Organizational 
99

Gibbons, MacIver &  Psychology, 73, 171‐80. 
Nyfield 
2001  Mael, Waldman &  From scientific work to organizational leadership: Predictors of management aspiration among 
Mulqueen  technical personnel. Journal of Vocational Behavior, 59 (1), 132‐148. 
2001  Segalla, Rouziès & Flory  Culture and career advancement in Europe: Promoting team players vs. fast trackers. European 
Management Journal, 19 (1), 44‐57. 
2002  Fields  Predicting potential for promotion: How the data in Human Resource information systems can be used 
to help organizations gain competitive advantage.  Available at: 
www.ilr.cornell.edu/depts/cahrs/downloads/pdfs/workingpapers/WP02‐14.pdf (accessed October 
6th, 2008). 
2002  Kesler  Why the leadership bench never gets deeper: Ten insights about executive talent development. 
Human Resource Planning, 25 (1), 32‐44. 
2002  Zenger & Folkman   The extraordinary leader: Turning good managers into great leaders. New York, NY: McGraw‐Hill.
Appendix A. Continued 
 
Year  Author(s)  Publication   
2003  Alldredge, Johnson,  Leadership development at 3M: New process, new techniques, new growth. Human Resource 
Stoltzfus & Vicere  Planning, 26(3), 45‐56. 
2003  Byham  Identifying potential – Are you identifying your top prospects? Executive Excellence, 20(5), 9.
2003  Pepermans, Vloeberghs &  High potential identification policies: An empirical study among Belgian companies. Journal of 
Perkisas  Management Development, 22 (8), 660‐678. 
2004  Eichinger & Lombardo  Learning agility as a prime indicator of potential. Human Resource Planning, 27 (4), 12‐15.
2004  Galpin & Skinner   Helping high flyers fly high: Their motives and developmental preferences. Industrial and 
Commercial Training, 36 (3), 113‐116. 
2004  Ilies, Gerhardt & Le  Individual differences in leadership emergence: Integrating meta‐analytic findings and behavioral 
genetics estimates. International Journal of Selection and Assessment, 12, 207‐219. 
2004  Stern  The right stuff: Are you an effective leader? Canadian Business, 77 (9), 95‐99.
2005  Cheng, Dainty & Moore  Towards a multidimensional competency‐based managerial performance framework: A hybrid 
approach. Journal of Managerial Psychology, 20 (5), 380‐396. 
100

2005  Kuok & Bell   Design, implementation, and evaluation of HR leadership development programs. CAHRS Working 


Paper 05­02. 
2005  McCartney & Campbell  Leadership, management, and derailment: A model of individual success and failure. Leadership and 
Organization Development Journal, 27 (3), 190‐202. 
2005 Snipes Identifying and cultivating high potential employees. Available at: 
http://www.ninthhouse.com/papers/2%20CLO_HIPOsArticle.pdf (Accessed October 6th, 2008). 
2006 Hay Group Potential – for what? What every CEO should know – new insights into selecting the right leaders to 
secure your competitive future. Available at: 
http://www.talentinstitute.co.za/research/files/Potential%20‐%20for%20what.pdf (accessed 
October 6th, 2008). 
2008  Dries & Pepermans  Dries, N. & Pepermans, R. (2008). "Real" high potential careers: An empirical study into the 
perspectives of organizations and high potentials. Personnel Review, 37 (1), 85‐108. 
     
Appendix B. Synopsis of the proposed model of leadership potential 
 
Region  Item   α 
COGNITION‐EXTRAPERSONAL (HEAD‐CONTEXT)
 
1 Information processing    .89
1 Input‐seeking behavior  i23 seeking and using feedback .69
i24 being open to new and diverse people and ideas 
i60 being emotionally intelligent (i.e. possessing certain social capacities) 
2 Insightfulness  i22 being insightful, seeing things from new angles .78
i35 demonstrating strategic thinking 
i43 displaying broad insight into the organization's business and one's own 
role in its goals 
i45 possessing a "helicopter view" (i.e. being multidisciplinary) 
i58 being intelligent (i.e. possessing certain analytical capacities) 
i46 reflecting critically on practices and procedures 
101

3 Decision making  i30 being decisive .70


i68 being able to make decisions rapidly 
i27 being assertive 
4 Problem solving  i2 being able to solve problems well and quickly .64
i67 possessing problem‐solving skills 
i49 being able to cope with complexity 
 
 
Appendix B. Continued 
 
Region  Item   α 
COGNITION‐INTRAPERSONAL (HEAD‐SELF) 
2 Learning agility    .88
5 Willingness to learn  i17 being open to learning .77
i19 chasing after variety, challenges and intellectual stimulation 
i20 seeking out opportunities to learn  
i29 being eager to learn about self, others and ideas 
i42 displaying self‐management in a manner that fosters learning and high 
performance 
i28 enjoying complex first‐time problems and challenges associated with 
new experiences 
6 Emotional stability  i10 being able to deal with stress and ambiguity .63
i15 demonstrating independence 
102

i44 demonstrating emotional stability 
i61 being self‐confident 
i59 being self‐aware of strengths and weaknesses 
7 Personal adaptability  i25 feeling comfortable with turbulent change .79
i26 not being afraid to take risks 
i37 showing adaptability 
i32 demonstrating flexibility 
i47 being change oriented 
i48 being proactive 
i57 displaying personal flexibility and mobility 
   
Appendix B. Continued 
 
Region  Item   α 
CONATION‐INTRAPERSONAL (HEART‐SELF) 
3 Drive    .87
8 Drive for results  i3 seizing opportunities when they present themselves .74
i12 being driven for excellence 
i31 being quality driven 
i33 demonstrating need for achievement (i.e. being performance‐oriented) 
i40 being competitive 
i41 consistently delivering tangible, measurable results above expectations 
i70 demonstrating a drive for results 
9 Persistence  i1 displaying high levels of energy .65
i11 showing drive and perseverance 
i13 persevering under adverse conditions 
103

10 Personal dedication  i18 doing more than just "carry out a job" .78


i6 demonstrating high dedication to the job 
i7 demonstrating intrinsic motivation (i.e. for the work in itself) 
i16 being passionate 
i21 being committed to making a difference 
i52 assuming responsibility/accountability 
i4 having an internal locus of control (i.e. attributing control over events to 
oneself) 
i56 displaying ambition, wanting to grow 
i74 taking initiative 
 
Appendix B. Continued 
 
Region  Item   α 
CONATION‐EXTRAPERSONAL (HEART‐CONTEXT)
4 Charismatic leadership    .93
11 People management  i75 showing commitment .75
i64 being credible 
i34 being able to manage others 
i77 motivating others 
i54 delegating decision making capacity to those best suited (i.e. 
empowerment) 
i76 directing others 
i38 being able to delegate 
i63 being able to build high‐performing teams 
i9 displaying leadership ability 
104

i66 actively looking for opportunities to lead 
i5 being able to set clear objectives 
12 Inspirational management  i65 enhancing one’s own “visibility” (i.e. ensuring one gets noticed by  .88
significant others) 
i71 being able to communicate strategically 
i62 building up professional credibility (i.e. getting results noticed) 
i36 demonstrating influence skills 
i73 knowing how to "sell" ideas 
i39 having cogency (i.e. being able to present strong arguments) 
i55 generating an impact 
i14 using, and not abusing, power 
i53 conveying a vision, inspiring, being charismatic 
13 Stakeholder management  i69 having a focus on the customer and the market .85
i8 having good interpersonal skills 
i51 being able to build long‐term relationships with clients 
i50 possessing networking skills (i.e. being able to build organizational 
relationships) 
i72 being able to adapt communication style and content to an audience 
105
106
 
107
Paper 3. 
 
Using emotional intelligence to identify high 
potentials: A metacompetency perspective 
 
 
This  paper  aims  to  demonstrate  the  utility  of  using  some  indication  of  emotional 
intelligence (EQ) to identify high potentials. Presupposed correspondences between the 
EQ  personal  factors  model  (Bar‐On,  1997)  and  Briscoe  and  Hall’s  (1999) 
metacompetency  model  of  continuous  learning  are  elucidated.  The  study  sample 
consisted of 51 high potentials and 51 ‘regular’ managers, matched onto one another by 
managerial level, gender and age. All participants completed an online survey containing 
Bar‐On’s Emotional Quotient Inventory (1997), Blau’s career commitment scale (1989) 
and  a  self‐anchored  performance  item.  EQ‐i  subscales  assertiveness,  independence, 
optimism,  flexibility  and  social  responsibility  appear  to  be  ‘covert’  high  potential 
identification  criteria,  separating  between  high  potentials  and  regular  managers. 
Furthermore, high potentials display higher levels of job performance and, supposedly, 
less boundaryless career attitudes. Using emotional intelligence – or at least some of its 
subscales  –  in  identifying  high  potential  may  well  contribute  to  the  validity  of  such 
processes.  Furthermore,  the  importance  of  cultivating  positive  emotions  at  work  is 
spelled  out  in  relation  to  high  potential  identification  and  development  policies  and 
retention management. In the majority of studies on high potentials, direct supervisors 
serve as research samples or a non‐empirical, rather normative approach is taken to the 
matter.  The  empirical  study  presented  in  this  paper  is  rather  unique  since  it  departs 
from  the  viewpoints  of  high  potential  individuals,  thus  delivering  added  value  to  the 
study domain.  
 
 
 

 
108

Figure IV. Topicality of Paper 3.
109
Paper 3. 
 
Using emotional intelligence to identify high potentials: A 
metacompetency perspective 
 
In  today’s  rapidly  changing  business  environment  driven  by  human 
capital,  knowing  which  qualities  to  look  for  in  future  leaders  is 
undoubtedly  a  key  competitive  advantage  (Buckingham  and  Vosburgh, 
2001).  Competency  frameworks,  designed  to  meet  this  need  –  and  to 
some extent, suitable instruments for doing so (Bournois and Rousillon, 
1998;  Briscoe  and  Hall,  1999;  Quinn  et  al.,  1990)  –  might  suffer  from 
(predictive) validity issues if they originate from a selection perspective 
alone  (McCall,  1998).  In  this  view,  employees  either  ‘have  it  or  not’, 
implying that young talent should be able to demonstrate competencies 
similar to those of successful executives, be it in a seminal form. As such 
‘talent detection’ competencies are based on past successes rather than 
on  future  challenges,  their  value  for  early  high  potential  identification 
may however be limited (Briscoe and Hall, 1999; Spreitzer et al., 1997).  
 
Alternatively,  Spreitzer  et  al.  (1997)  argue  for  taking  a  developmental 
perspective  to  high  potential  identification.  They  posit  that  executive 
potential  is  best  predicted  by  the  ability  to  learn  from  experience. 
Research  has  demonstrated  that  learning  agility  is,  indeed,  strongly 
related to  advancement  potential (e.g. Lombardo and Eichinger, 2000). 
In a similar vein, Briscoe and Hall (1999) speak of metacompetencies – 
competencies  so  powerful  that  they  affect  an  individual’s  ability  to 
develop  the  competencies  they  will  need  in  the  (unpredictable) future. 
Contemporary careers demand adaptability from employees, that is, the 
ability to self‐correct in response to new and unfamiliar demands from 
their  environment,  “without  waiting  for  formal  training  and 
development from the organization” (Hall and Moss, 1998, p.31).  
 
Emotional intelligence and continuous learning 
 
Strikingly,  it  appears  that  the  multidimensional  construct  of  emotional 
intelligence as defined by Bar‐On (1997) – frequently referred to as trait 
EQ  or  the  EQ  personal  factors  model  (Dulewicz  and  Higgs,  2004)  – 
describes  the  attributes  individuals  need  to  be  able  to  engage  in  such 
continuous  and  generative  learning.  According  to  Bar‐On  (1997), 
emotional  intelligence  (EQ)  is  “an  array  of  noncognitive  capabilities, 
competencies and skills that influence one’s ability to succeed in coping 
with environmental demands and pressures” (p. 14). The author defines 
five  factors  of  EQ,  which  he  believes  contribute  to  success  in  life: 
intrapersonal  functioning  (self‐awareness  and  self‐expression), 
110
interpersonal  skills  (social  awareness  and  interpersonal  relationships), 
adaptability  (change  management),  stress  management  (emotional 
management and regulation) and general mood (self‐motivation) (Bar‐
On, 1997; Bar‐On, 2005).  
 
The idea that emotional intelligence may underlie certain practical skills 
is not new. Some previous studies have focused on the link between EQ 
and  emotional  competency  (Boyatzis  et  al.,  2000;  Goleman,  2001),  a 
concept  that  was  operationalized  by  a  set  of  20  competencies. 
Conversely,  Briscoe  and  Hall  (1999)  describe  no  more  than  two 
metacompetencies:  adaptability  learning  (the  ability  to  react  to  change 
through  learning)  and  identity  learning  (the  ability  to  self‐reflect  and 
assimilate  what  is  learned).  Table  1  demonstrates  possible 
correspondences  between  the  metacompetency  model  and  the  EQ 
personal  factors  model.  Bar‐On’s  (2005)  description  of  the  skills  and 
competencies  assessed  by  each  scale  of  the  EQ‐i  (Emotional  Quotient 
Inventory)  proved  useful  for  matching  his  EQ  scales  (1997)  to  the 
behavioral descriptions of Briscoe and Hall’s metacompetencies (1999). 
However,  the  correspondences  exhibited  in  Table  1  are  tentative  and 
serve  merely  to  demonstrate  the  probability  of  a  connection  between 
emotional  intelligence  (as  measured  by  the  EQ‐i)  and  the  potential  for 
continuous learning.  
 
The  current  study  explores  the  ‘potential’  of  the  EQ  personal  factors 
model  for  predicting  ‘high  potential’  in  employees.  Spreitzer  et  al. 
(1997) observed: “Though the literature has provided evidence that job 
experiences can be a stimulus for learning, it has also suggested that not 
all  people  learn  equally  well  from  the  same  kinds  of  experiences.  As 
such, the identification of executive potential would likely be enhanced 
through knowledge of the variation in people’s ability to acquire needed 
skills,  that  is,  their  ability  to  learn  from  experience.”  (p.  9).  This  paper 
hypothesizes  that,  ideally,  high  potentials  are  those  with  high  levels  of 
emotional intelligence – and consequently, of learning agility (Lombardo 
and Eichinger, 2000). In order to look into this proposition, an empirical 
study was set up to contrast a group of high potential managers with a 
group of regular managers. 
 
Two possible outcomes are anticipated: 
 
(1)  On  average,  high  potential  managers  display  higher  EQ­i  scores  than 
regular  managers  (irrespective  of  functional  level).  If  this  is  so,  then 
emotional intelligence must have played some role in the identification 
of these high potentials. 
Table 1. Presupposed correspondences between subscales of the EQ personal factors model and adaptability and identity metacompetencies 
 
EQ scales a  EQ subscales a    Adaptability and identity metacompetenciesb 
1. Intrapersonal functioning  1. Emotional self‐awareness I Self‐assessment
  2. Assertiveness  A Dialogue skills
  3. Self‐regard  I Being willing to modify self‐perceptions as change occurs
  4. Self‐actualization A Eagerness to accept new challenges in unexplored territory 
    I Engaging in a variety of personal development activities
  5. Independence  I Exploring, communicating and acting on personal values
   
2. Interpersonal skills  6. Empathy  A Openness to new and diverse people and ideas 
  7. Interpersonal relationships I Actively seeking out relationships that evoke learning
  8. Social responsibility I Rewarding subordinates for personal development work
   
3. Adaptability  9. Problem solving I Seeking, hearing and acting on personal feedback
111

  10. Reality testing  A Exploration 
  11. Flexibility  A Flexibility 
   
4. General mood  12. Happiness 
  13. Optimism  I Being open to diverse people and ideas  
   
5. Stress management  14. Stress tolerance A Comfort with turbulent change  
  15. Impulse control
Notes. A = Adaptability metacompetency; I = Identity metacompetency. 
a From “Emotional Quotient Inventory: Technical Manual,” by R. Bar‐On, 1997, Toronto: Multi‐Health Systems. 
b From  “Grooming and Picking Leaders Using Competency Frameworks: Do They Work? An Alternative Approach and New Guidelines for Practice,” 

by J.P. Briscoe and D.T. Hall, 1999, Organizational Dynamics, 28, p. 37‐52. 
 
 
112
As  it  is  unlikely  that  organizations  explicitly  assess  EQ  as  part  of  their 
high potential selection and development policies (Boyatzis et al., 2000), 
it  follows  that  EQ  must  be  some  sort  of  covert  identification  criterion. 
Two facts seem to support this proposition.  
 
First,  several  overt  high  potential  identification  criteria  (i.e. 
performance,  competencies)  appear  to  be  associated  with  EQ. 
McClelland  (1998)  found  that  a  wide  range  of  emotional  competencies 
(and  only  a  narrow  range  of  cognitive  ones)  allow  for  workforce 
segmentation  –  i.e.  for  distinguishing  between  top  performers  and 
average ones. More recently, these emotional competencies (Boyatzis et 
al.,  2000;  Goleman,  2001)  were  shown  to  display  significant  overlap 
with  competencies  commonly  used  in  organizations  to  identify 
potential. For instance, Gowing et al. (2006) demonstrated how the EQ 
competencies  defined  by  Boyatzis  et  al.  (2000)  can  be  aligned  with 
existing competency frameworks in organizations, departing from their 
behavioral definitions. 
 
Second, EQ has often been conceptually linked to models of effective or 
transformational leadership. Goleman (1998a), in his seminal article on 
emotional  intelligence,  stated  that  IQ  and  technical  skills  are  merely 
threshold capabilities for being an effective leader, while EQ is the sine 
qua non. Dulewicz and Higgs (2003, 2005) ascertained that the various 
dimensions  of  the  concept  of  emotional  intelligence  also  appear  in 
models  of  authorities  in  the  field  of  leadership.  Others  have  found  that 
EQ  is  able  to  positively  predict  transformational  leadership  (Barling  et 
al., 2003; Brown and Moshavi, 2005; Gardner and Stough, 2002; Rosete 
and  Ciarrochi,  2005;  Sosik  and  Mergerian,  1999),  leadership 
effectiveness  (Dulewicz  et  al.,  2005;  Goleman  et  al.,  2002;  Rosete  and 
Ciarrochi,  2005)  and  career  advancement  (Dulewicz  and  Higgs,  2000; 
Rosete  and  Ciarrochi,  2005).  Consequently,  as  high  potentials  are 
defined  as  “those  individuals  within  the  organization  who  are 
recognised,  at  that  point  in  time,  as  the  organization’s  likely  future 
leaders” (Cope, 1998, p.15), their identification should be to some extent 
related to their emotional intelligence.  
 
(2)  High  potentials  at  higher  functional  levels  exhibit  higher  EQ­i  scores 
than  regular  managers  and  high  potentials  at  lower  functional  levels. 
Several authors have come to the conclusion that emotional intelligence 
– and interpersonal skill in particular – becomes increasingly important 
(compared to IQ and technical skills) as individuals advance within their 
organizational  hierarchies  (Dulewicz et  al.,  2005;  Goleman  et  al.,  2002; 
Hall,  1999).  It  is  evident  that  as  job  demands  shift,  so  should  the 
113
jobholder’s skills and behaviors. Career derailment is, then, “the failure 
of fit of the individual with the evolving demands of the job over time” 
(Leslie  and  Van  Velsor,  1996,  p.  36)  –  the  authors  note  that  this 
definition  can  also  be  applied  to  more  ‘boundaryless’  career  tracks. 
Could it be that a lack of emotional intelligence is a covert cause of high 
potential  derailment  (rather  than  high  levels  of  EQ  covertly  leading  to 
high potential identification)? If this is the case, then the high potentials 
who have ‘arrived’ at the highest functional levels would be those who 
have displayed above‐average levels  of emotional intelligence;  those at 
lower  functional  levels,  however,  need  not  necessarily  excel  in  this 
respect  since  strong  EQ‐driven  competencies  do  not  become  crucial 
until later in the career.  
 
The  inability  to  evolve  from  an  independent  (autonomous)  to  an 
interdependent (team‐oriented) job actor is generally recognized as the 
number‐one  cause  of  derailment  (Frankel,  1994;  Kovach,  1986). 
Actually,  it  appears  that  nearly  all  recurrently  reported  causes  of 
derailment  are  somehow  related  to  a  lack  of  emotional  intelligence 
(apart  from  the  inability  to  meet  business  objectives).  Most  frequently 
mentioned  are  problems  with  interpersonal  relationships,  arrogance 
and insensitivity to others, the inability to build a team and the inability 
to  develop  or  adapt  (Daft,  1999;  Hall,  1999;  Hogan  and  Hogan,  2001; 
Kovach, 1989; Leslie and Van Velsor, 1996; McCall and Lombardo, 1983; 
Van Velsor and Leslie, 1995). 
 
Linking performance and commitment to EQ and high potential 
status 
 
In addition to examining the EQ‐high potential link, this paper will look 
into  the  (frequently  reported)  relationships  between  job  performance, 
career  commitment,  the  EQ  personal  factors  model  and  high  potential 
status (i.e. the binomial variable high potential/non‐high potential).  
 
Empirical  research  on  individual  job  performance  (Côté  and  Miners, 
2006;  Daus  and  Ashkanasy,  2005;  Goleman,  1998a,  1998b;  Rosete  and 
Ciarrochi,  2005;  Van  Rooy  et  al.,  2005)  as  well  as  group  or  team 
performance  (Daus  and  Ashkanasy,  2005;  Jordan  and  Troth,  2004; 
Offerman  et  al.,  2003)  and  even  organizational  performance  (Goleman, 
1998a;  McClelland,  1998;  Williams,  1994)  has  uncovered  that  these 
variables are all positively related to emotional intelligence. In addition, 
amongst the criteria used for identifying high potential in organizations, 
job performance is probably the pivotal one. An employee who does not 
consistently receive ‘exceeds expectations’ performance ratings, is very 
114
unlikely to be considered for entering any high potential program (Cope, 
1998; Pepermans et al., 2003).  
 
As for career commitment, i.e. “one’s attitude toward one’s profession or 
vocation”  (Blau,  1985,  p.  20),  academic  literature  has  demonstrated  its 
relationships  with  several  factors  relevant  to  career  success  (Spreitzer 
et al., 1997). Employees with higher career commitment display higher 
levels of willingness to make sacrifices for their organizations (Randall 
et al., 1990), are less likely to leave their organizations (Lee et al., 1992), 
and display higher performance (Aryee et al., 1994; Bashaw and Grant, 
1994).  
 
Moreover, some researchers claim that career commitment is influenced 
by emotional intelligence (Brown et al., 2003; Carson and Carson, 1998; 
Matthews et al., 2002), especially when individuals are faced with highly 
complex  managerial  work  that  can  lead  to  high  levels  of  stress. 
According  to  Carmeli  (2003),  emotional  intelligence  enables  people  to 
control their stress levels effectively and thus prevent these from having 
negative  effects  on  their  career‐related  attitudes.  Finally,  strong 
commitment to a profession, a job, or an organization have all been put 
forward  as  crucial  criteria  for  identifying  high  potential  (Bennis  and 
Nannus, 1985; McCall, 1994; Woodruffe, 1993).  
 
The  hypotheses  of  the  study  presented  in  this  paper  are  listed  below. 
Hypothesis  1  links  high  potential  status  to  emotional  intelligence; 
hypotheses  2a  and  2b  look  into  the  relationships  between  job 
performance,  high  potential  status  and  emotional  intelligence;  and 
hypotheses  3a,  3b  and  3c  examine  the  connections  between  career 
commitment, job performance, emotional intelligence and high potential 
status.  
 
H1.  High  potential  managers  display  higher  levels  of  emotional  intelligence 
than their ‘regular’ peers. 
 
H2a.  Emotional  intelligence  is  positively  related  to  individual  job 
performance. 
 
H2b.  High  potential  managers  display  higher  job  performance  levels  than 
their regular peers.  
 
H3a. Career commitment is positively related to individual job performance. 
 
H3b. Emotional intelligence is positively related to career commitment. 
 
H3c.  High  potential  managers  display  higher  levels  of  career  commitment 
than their regular peers. 
115
Method 
 
Participants 
 
A total of 133 managers participated in the study. Fifty‐one of them had 
been,  at  that  time,  identified  as  high  potentials  by  their  organizations; 
the remaining 82 were regular managers. Managers were matched onto 
high  potentials  by  relative  functional  level  (i.e.  their  current  functional 
level,  li,  relative  to  the  highest  possible  functional  level  in  the 
organization, lmax, producing a fraction with 1 as a maximum). Through 
this  matching  procedure,  four  sample  categories  were  created: 
managers  at  lower  managerial  levels  (Ml),  managers  at  higher 
managerial levels (Mh), high potentials at lower managerial levels (HPl) 
and high potentials at higher managerial levels (HPh). Participants were 
then further matched by gender and age. Finally, a randomized number 
generator was used to decide which of the managers to eliminate from 
the  sample,  and  so  attain  equal  sample  sizes  per  category.  The  data 
gathered  from  the  31  ‘no‐match’  managers  were  not  included  in  this 
study. Table 2 provides an overview of sample size, gender distribution, 
and age descriptives per sample category.  
 
The  matching  procedure  described  above  served  internal  validity 
purposes,  i.e.  reduced  the  probability  of  unobserved  bias  in  the  cause‐
effect  relationships  under  study  (Trochim,  2000).  Splitting  the  sample 
by  functional  level  allowed  for  testing  of  the  propositions  on  EQ  and 
derailment  (cross‐sectionally).  Matching  participants  by  gender  ruled 
out  potential  differences  in  EQ  between  male  and  female  participants 
that  might  bias  the  study’s  findings.  An  independent‐samples  t‐test 
exposed  that,  in  this  sample,  women  scored  significantly  higher  than 
men  on  the  EQ‐subscales  empathy  (t(100)  =  ‐2.23,  p  <  .05)  and 
interpersonal relationships (t(100) = ‐2.04, p < .05), but not on the EQ‐i 
overall. Similar results have been reported by Bar‐On (2000) and Eagly 
and Johnson (1990). As the man‐to‐woman ratio is almost equal across 
the four sample categories (79‐21% for lower managerial levels and 78‐
22% for higher managerial levels), however, no internal validity issues 
should  arise  when  studying  male  and  female  participants  together  as 
one group. Finally, age – a typical correlate of EQ (Mayer et al., 1999) – 
was  discarded  as  a  ‘contaminating’  factor  in  this  study  as  the  sample 
categories  were  found  not  to  differ  significantly  in  age  (F(3,97)  =  1.88, 
ns). 
Table 2. Sample size, gender distribution and means and standard deviations of age per sample category 
 
    High potentials (HP)    Managers (M) 
    n  M age SD age n match M age  SD age  n no match
Lower managerial levels (l)a  Men  19  39,26 5,52 19 38,38 5,92  13
  Women  5  39,00 3,08 5 39,00 3,32  5
  Total  24  24   18
         
Higher managerial levels (h) b  Men   21  39,29 2,78 21 41,33 3,44  9
  Women  6  38,83 2,93 6 42,00 3,74  4
  Total  27  27   13
         
Total    51  51   31
Notes. a Li/Lmax < .5  b Li/Lmax ≥ .5  
116
117
Procedure 
 
An online survey study was set up. HR directors from 69 organizations 
known to engage in high potential identification and development were 
addressed  by  e‐mail  and  follow‐up  telephone  calls,  requesting  their 
participation.  Only  three  organizations,  from  the  financial,  insurance 
and telecom sector respectively, eventually agreed to let their managers 
partake in the study – which demonstrates the ongoing discretion about 
high potentials in organizations. In order to reduce the chance of inter‐
organizational  heterogeneity  affecting  the  study  results,  the  study’s 
definition of the concept high potential (Cope, 1998) was stressed in the 
initial mailings; organizations with deviant definitions were not allowed 
to participate.  
 
The  participating  organizations  received  an  e‐mail  with  enclosed  two 
quasi‐identical URLs: one containing the letter a in its character string, 
to be sent out to managers who had been identified as high potentials; 
the other containing the letter b in its character string, to be sent out to 
regular  managers.  The  URLs  lead  to  identical  surveys  but  data  were 
routed  to  two  separate  databases  (in  the  stage  of  data  analysis,  these 
were  merged  and  high  potentials  were  coded  ‘1’  whilst  regular 
managers  received  code  ‘0’).  The  URLs  were  sent  out  by  each 
organization’s HR director following the above guidelines, together with 
an  e‐mail  stating  that  the  organization  had  decided  to  participate  in  a 
study  on  emotional  intelligence,  that  participation  was  voluntary,  and 
that personalized EQ profiles could be obtained after termination of the 
study granted that participants provide their e‐mail address. The utility 
of such profiles within the context of personal development was briefly 
addressed.  
 
This  procedure  ensured  participants’  anonymity  since  no  names  or 
contact data were disclosed outside of their organizations, unless if they 
themselves  chose  to  do  so.  Furthermore,  as  all  participants  received 
identical  instructions  and  surveys,  no  information  on  high  potential 
status  was  revealed  to  employees.  This  latter  intervention  was  very 
important  to  the  participating  organizations  –  mostly  out  of  fear  of 
frustrating the non‐high potentials.  
 
After  closing  the  data  collection,  all  raw  data  were  sent  to  PEN 
Psychodiagnostics  in  the  Netherlands  –  the  association  that  is 
responsible for distribution of the EQ‐i in the Netherlands and Belgium. 
After  being  standardized  against  a  Dutch‐speaking  norm  population  of 
approximately 1100 individuals, the data was then returned to us in the 
118
form of t‐scores. EQ‐i scores are reported as t‐scores with a mean of 100 
and a standard deviation of 15, as is also the convention for IQ (Bar‐On, 
1997).  Note  that  all  analyses  reported  in  this  paper  were  performed 
using these t‐scores, not the raw scores.  
 
Measures 
 
EQ.  In  order  to  measure  participants’  emotional  intelligence,  Bar‐
On’s  EQ‐i  (1997)  was  administered.  This  self‐report  measure  is 
widespread and has been extensively validated (Bar‐On, 2000). Several 
studies  have  demonstrated  reasonable  levels  of  internal  consistency, 
test‐retest  reliability  and  predictive,  convergent  and  discriminant 
validity (Bar‐On, 2000; Dawda and Hart, 2000; Hedlund and Sternberg, 
2000). All 133 EQ‐i items require a response on a five‐point Likert scale 
ranging from ‘very seldom or not true of me’ to ‘very often true of me or 
true of me’. Note that the scales and subscales of the EQ‐i were listed in 
Table 1.  
 
Recently,  several  authors  have  made  the  case  for  using  the  ‘ability’ 
rather than the ‘personal factors’ models of EQ (Côté and Miners, 2006; 
Daus  and  Ashkanasy,  2005;  McEnrue  and  Groves,  2006;  Rosete  and 
Ciarrochi,  2005).  Common  criticism  on  the  EQ‐i  is  that  the  model  on 
which it is based is too broad – i.e. includes dimensions and components 
not  included  in  the  original  definition  of  emotional  intelligence  (Daus 
and  Ashkanasy,  2005;  Salovey  and  Mayer,  1990).  Furthermore,  some 
research  suggests  considerable  overlap  between  the  EQ‐i  measure  and 
established  measures  of  personality,  like  the  NEO‐PI‐R  and  the  16  PF 
(Dawda  and  Hart,  2000;  McEnrue  and  Groves,  2006).  However,  other 
authors (Mayer et al., 2000; Shulman and Hemenover, 2006) emphasize 
that  when  studying  ‘new’  constructs  such  as  EQ,  being  able  to  predict 
even  small  amounts  of  additional  variance  over  known  variables  (i.e. 
personality)  may  be  considered  substantial  advances.  Moreover,  the 
rationale behind the MCSEIT, the most frequently employed measure of 
the  EQ  ability  model,  is  not  irrefutable  either.  Freudenthaler  and 
Neubauer  (2005)  observed:  “Although  emotional  knowledge  about  the 
effectiveness  of  emotion‐related  behavior  can  be  regarded  as  a 
‘maximum‐performance’  indicator  of  an  individual’s  emotional 
management  ability  (assessing  what  an  individual  is  capable  of  when 
highly  motivated),  it  should  not  be  neglected  that  managing  emotions 
effectively does not only require the availability of such knowledge but 
also to behave according to it.” (p. 571). 
 
119
In  any  case,  this  paper  wishes  to  steer  clear  of  ‘or‐or’  discourse  and 
subscribes to the vantage‐point of Van Rooy et al. (2005) who argue that 
both models of EQ have utility depending on the context in which they 
are  used.  The  authors  posit  that,  considering  its  broad  reach,  the 
personal  factors model of EQ is valuable mainly in selection contexts  – 
as is the case in this paper. 
 
  Job  performance.  An  indication  of  individual  job  performance  was 
obtained by a single item, i.e. “When comparing yourself to co‐workers 
who  hold  similar  jobs,  how  would  you  rate  your  own  performance 
(supposing  that  0  is  the  performance  of  your  worst  performing  co‐
worker  and  10  that  of  your  best  performing  co‐worker)?”  Choosing  to 
assess  performance  through  self‐report,  although  subject  to  self‐
enhancement  bias  (Campbell  and  Lee,  1988),  allowed  for  the  entire 
survey  to  be  self‐reported.  The  curreny  departs  solely  from  the 
viewpoints  of  the  high  potential  individuals  themselves,  in  doing  so 
delivering added value to the study domain. In the majority of studies on 
high potentials, direct supervisors serve as research samples  or a non‐
empirical, rather normative approach is taken to the matter (Pepermans 
et  al.,  2003).  Furthermore,  having  participants  ‘self‐anchor’  their 
responses  to  the  job  performance  item  should  enhance  its  validity 
(Kilpatrick and Cantril, 1960) 
 
Career commitment. Career commitment was measured with Blau's 
7‐item  scale  (1989).  All  items  of  this  measure  are  scored  on  a  5‐point 
Likert scale anchored from strongly disagree to strongly agree. Sample 
items  include,  “I  definitely  want  a  career  for  myself  in  this  profession” 
and  “I  like  this  vocation  too  well  to  give  it  up”.  Blau  (1985,  1989) 
demonstrated  the  scale’s  reliability  and  distinctness  from  related 
constructs such as job involvement and organizational commitment. He 
reported  scale  internal  consistencies  of  around  .87  and  a  test‐retest 
reliability of .67 over a seven‐month period (Blau, 1985). The coefficient 
alpha for this measure in the present paper was .80. 
 
Demographics. Participants were asked to indicate lmax and li with 
a  single  figure  (minimum  1),  relying  upon  objective  data  (e.g. 
organizational diagrams). Finally, age and gender were surveyed.  
 
Results 
 
EQ‐i  means  and  standard  deviations  were  scrutinized  for  each  sample 
category prior to testing the hypothesized relations between emotional 
intelligence,  self‐reported  job  performance,  career  commitment  and 
120
high  potential  status.  Only  HPhs’  total  EQ‐i  score  exceeded  115  (m  = 
115.89,  sd  =  13.66).  Total  EQ‐i  scores  for  the  other  sample  categories 
were all higher than 100 but lower  than  115, thus within the ‘average’ 
range.  On  a  subscale  level,  Ml  and  Mh  displayed  below‐average  scores 
on  optimism  (m  =  88.13  and  89.22  respectively),  and  HPl  and  HPh  on 
social responsibility (m = 92.96 and 97.26 respectively). HPl also scored 
below average on empathy (m = 97.75) and impulse control (m = 99.54) 
and HPh scored considerably above average on flexibility (m = 120.11). 
Table  3  presents  means,  standard  deviations  and  correlations  of  the 
study variables. 
 
H1: EQ – high potential status 
 
No  significant  differences  were  found  between  high  potentials’  and 
regular managers’ total EQ‐i scores (t(100) = ‐.44, ns). Figure 1 depicts 
the  observed  means  for  all  four  sample  categories.  Note  that  HPh  do 
display  the  highest  EQ‐i  scores,  although  differences  with  Ml,  Mh  and 
HPl are not significant (F(3,97) = 1.78, ns). HPl, contrary to expectations, 
exhibit the lowest EQ‐i scores.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1. Means plots of EQ‐i scores for HPl, HPh, Ml, and Mh 
 
When  considering  the  data  at  a  subscale  level,  several  significant 
differences  were  observed.  High  potentials  display  higher  levels  of 
assertiveness (t(100) = ‐2.38, p < .05), independence (t(100) = ‐2.28, p < 
.05), optimism (t(100) = ‐3,29, p < .01) and flexibility (t(100) = ‐2.18, p < 
.05).  On  the  subscale  social  responsibility,  however,  managers  were 
found to demonstrate significantly higher scores (t(100) = 2.28, p < .05) 
than  their  high  potential  counterparts.  It  thus  appears  that  the 
aforementioned  propositions  on  EQ  being  a  covert  high  potential 
identification criterion (H1) were only partially supported.  
Table 3. Means, standard deviations and intercorrelations of study variables (n = 133) 
 
  m  sd  1  2  3  4  5  6  7  8  9 
1. Emotional intelligence (total)    112.25  12.81  ‐‐                 
2. Emotional self‐awareness   110.60  13.90  .68**  ‐‐               
3. Assertiveness   110.80  12.16  .62**  .46**  ‐‐             
4. Self‐regard   107.06  13.92  .67**  .34**  .52**  (.77)           
5. Self‐actualization    106.66  11.85  .68**  .52**  .57**  .63**  (.83)         
6. Independence   107.78  12.52  .53**  .19  .51**  .63**  .50**  (.88)       
7. Empathy   101.19  12.85  .39**  .43**  .03  ‐.07  .19  ‐.07  (.81)     
8. Interpersonal relationships   105.18  11.62  .75**  .68**  .38**  .56**  .51**  .28**  .34**  (.85)   
9. Social responsibility   98.59  15.17  .35**  .39**  .01  ‐.05  .25*  ‐.13  .72**  .23*  (.90) 
10. Problem solving   108.49  12.92  .40**  .28**  .13  .04  .25*  .01  .34**  .09  .41** 
11. Reality testing  109.66  12.52  .65**  .42**  .43**  .46**  .36**  .35**  .10  .44**  .02 
12. Flexibility  113.88  13.24  .64**  .39**  .43**  .40**  .37**  .43**  .22*  .42**  .09 
121

13. Happiness  107.48  11.86  .75**  .50**  .48**  .68**  .59**  .42**  .15  .69**  .16 
14. Optimism  99.48  34.57  .29**  .20*  .46**  .29**  .23*  .28**  ‐.17  .19  ‐.26** 
15. Stress tolerance  111.82  11.51  .75**  .37**  .55**  .51**  .48**  .53**  .08  .47**  .07 
16. Impulse control  104.45  16.28  .52**  .31**  .06  .24*  .10  .10  .11  .34**  .15 
17. Job performance  7.09  2.54  .09  .14  .35**  .10  .10  .16  ‐.18  .00  ‐.31**
18. Career commitment  3.70  .68  .21*  .12  .21*  .23*  .25*  .29**  ‐.22*  .02  ‐.17 
Notes. Coefficient alphas are on the primary diagonal; * p <.005; ** p <.001. 
 
 
 
 
 
 
Table 3. Continued 
 
10  11  12  13  14  15  16  17  18 
         
         
         
         
         
         
         
         
         
(.79)         
.25*  (.89)       
.20*  .41**  (.80)     
122

.07  .44**  .38**  (.81)   


.06  .38**  .27**  .58**  (.85) 
.39**  .48**  .59**  .53**  .39**  (.79)
.19  .49**  .37**  .37**  .23*  .33** (.81)
.02  .20*  .17  .18  .88**  .23* ‐.01 ‐‐
.21*  .09  .12  .08  .31**  .40** ‐.04 .28** (.80)

Notes. Coefficient alphas are on the primary diagonal; * p <.005; ** p <.001. 
 
 
 
 
 
 
123
MANOVA  generated  differences  in  flexibility,  happiness  and 
interpersonal relationships between the four sample categories (Figure 
2). Only flexibility (F(3, 97) = 2.72, p < .05) shows the expected pattern; 
for  the  subscales  happiness  (F(3,97)  =  4.36,  p  <  .01)  and  interpersonal 
relationships (F(3,97) = 4.25, p < .01), Ml show the highest scores of all 
sample categories. Remarkably, for overall EQ‐i as well as for happiness 
and  interpersonal  relationships,  high  potentials’  scores  increase  with 
functional  level,  while  managers’  scores  decrease.  The  implications  of 
these  findings  for  the  EQ‐derailment  connection  posited  earlier  in  this 
paper will be discussed below.  
 
H2a: EQ – job performance 
 
Linear  regression  analysis  revealed  that  total  EQ‐i  score  does  not 
significantly  predict  self‐reported  job  performance  (β  =  0.09,  ns).  On  a 
subscale  level,  optimism  (β  =  0.94,  p  <  .01)  and  emotional  self‐
awareness  (β  =.15,  p  <  .05)  were  found  to  be  positive  significant 
predictors of job performance; interpersonal relationships (β = ‐.23, p < 
.01)  and  reality  testing  (β  =  ‐.12,  p  <  .01)  were  found  to  be  negative 
significant  predictors,  indicating  that  higher  scores  on  these  EQ 
subscales  led  to  lower  self‐reported  performance  ratings.  Together, 
these four predictors accounted for 82% (adjusted R²) of variance in the 
job performance score, providing some support for H2a. 
 
H2b: job performance – high potential status 
 
Figure  3  illustrates  the  relationship  between  self‐reported  job 
performance  and  high  potential  status.  High  potentials  were  found  to 
display significantly higher levels of individual job performance than did 
managers  (t(100)  =  ‐4.14,  p  <  .01).  When  comparing  the  four  different 
sample  categories,  HPh  were  found  to  report  significantly  higher  job 
performance scores than Ml (F(3, 97) = 2.72, p < .05) and Mh (F(3, 97) = 
2.72, p < .05). H2b is thus accepted.  
 
H3a: career commitment – job performance 
 
The  career  commitment  scale  score  did  not  significantly  predict  self‐
reported performance (β = 0.09, ns), leading us to reject H3a. Note that 
separate regression analyses were conducted for H2a and H3a. Putting 
career commitment and emotional intelligence in one predictive model 
of  job  performance  proved  ineffective  because  of  the  multicollinearity 
between the two constructs (see also H3b below).  
124
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2. Means plots of flexibility, interpersonal relationships and happiness scores for 
HPl, HPh, Ml, and Mh 
 
 
125
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 3. Means plots of job performance ratings for HPl, HPh, Ml, and Mh 
 
 
H3b: EQ – career commitment 
 
Total score on the EQ‐i significantly predicted career commitment (β = 
.21, p < .05)  in the total sample. As for the 15 subscales of the EQ‐i, all 
subscales  but  the  three  included  in  the  adaptability  scale  (problem 
solving,  reality  testing  and  flexibility)  were  significant  predictors, 
making  up  a  model  with  an  explanatory  power  of  28%  (adjusted  R²). 
H3b was thus, for the most part, supported.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 4. Means plots of career commitment scores for HPl, HPh, Ml, and Mh 
 
 
 
 
126
H3c: career commitment – high potential status 
 
Finally,  the  possibility  of  distinguishing  between  HP  and  M  based  on 
participants’  career  commitment  scores  was  examined.  Differences 
between HP and M were not significant for the career commitment scale 
score  (t(100)  =  ‐.14,  ns).  At  item  level,  HP  rated  “If  I  could  get  another 
job and be paid the same amount, I would probably take it” significantly 
lower  (t(100)  =  2.36,  p  <  .05).  No  significant  differences  were  found 
between the different sample categories (Figure 4). Thus, it appears that 
the evidence for H3c is weak.  
 
Discussion 
 
This paper attempts to demonstrate the utility of using some indication 
of  emotional  intelligence  to  identify  high  potentials.  As  has  been 
demonstrated  in  Table  1,  all  15  subscales  of  the  EQ  personal  factors 
model (Bar‐On, 1997) can be mapped onto the metacompetency model 
proposed  by  Briscoe  and  Hall  (1999).  It  was  therefore  hypothesized 
that,  in  today’s  fast‐moving  business  environment,  possessing  high 
emotional  intelligence  –  and,  consequently,  a  high  potential  for 
continuous  learning  –  should  be  a  crucial  criterion  for  high  potential 
identification.  
 
Using EQ to distinguish between high potentials and ‘regular’ managers 
 
A first indication of the accuracy of the proposed hypotheses emerged as 
findings  revealed  that  high  potentials  at  high  managerial  levels  (HPh) 
displayed  an  EQ‐i  score  of  115.89,  which  is  more  than  one  standard 
deviation  above  average  (Bar‐On,  1997),  while  the  other  sample 
categories (Ml, Mh and HPl) exhibited scores within the average range. 
However, differences in total EQ‐i score were not significant. Support for 
H1 was found at the subscale level alone.  
 
Assertiveness,  independence,  optimism,  flexibility  and  social 
responsibility  appear  to  be  covert  high  potential  identification  criteria, 
separating  between  the  group  of  HP  and  non‐HP.  Strikingly,  these 
attributes fit very nicely into the definition of the protean career, which 
highlights  continuous  learning,  self‐direction  and  choices  based  on 
personal  values  (Briscoe  and  Hall,  2006).  Independence  allows  for 
autonomous  action  (“…without  waiting  for  formal  training  and 
development  from  the  organization”,  Hall  and  Moss,  1997,  p.  31,  see 
earlier). Employees who are independent are self‐reliant and  thus able 
to  take  their  career  into  their  own  hands.  Furthermore,  they  show 
127
themselves  as  leaders,  not  followers,  from  early  on  in  their  career,  as 
they make decisions based on their own personal value system – which 
is,  alongside  self‐direction,  a  core  feature  of  protean  careers  (Briscoe 
and  Hall,  2006).  Assertiveness  helps  HP  express  their  aspirations  and 
ambitions  to  significant  others  within  their  organizations.  Employees 
who  are  assertive  are  not  afraid  to  display  initiative,  achieve  in  ways 
that may be labeled competitive by others and certainly not to ‘stand out 
from  the  crowd’.  Creating  visibility  in  one’s  organization  from  early  on 
in  the  career  is  a  prerequisite  for  being  identified  as  a  high  potential 
(Capowski,  1994;  Feild  and  Harris,  1991).  Flexibility  is  probably  most 
clearly linked to Briscoe and Hall’s (1999) metacompetency framework. 
HPh  display  an  average  score  of  120.11  on  this  subscale  (which  is 
significantly higher than Ml and Mh), demonstrating how essential being 
flexible  is  for  high  potentials.  Being  able  to  adapt  to  changing 
circumstances is, by far, the most crucial competency of the 21st century 
(Briscoe  and  Hall,  1999,  Tett  et  al.,  2000).  Optimism  denotes  always 
looking ahead without being easily demotivated. Optimists tend to turn 
problems  into  learning  opportunities  and  inspire  others  with  their 
ability to put things into perspective (Rottinghaus et al., 2005). Finally, 
social  responsibility  distinguishes  between  high  potentials  and  regular 
managers  –  however,  it  is  the  group  of  managers  that  scores 
significantly  higher  on  this  subscale.  It  appears  that  high  potentials 
show  more  ‘Machiavellism’  than  other  managers  do  (“the  end  justifies 
the  means”).  Probably,  the  competitive  aspect  of  organizational  high 
potential  policies  plays  a  part  in  this  finding,  as  individualism  may  be 
encouraged. Furthermore, since high potentials are considered separate 
from  regular  managers  from  early  on  in  the  career  (receiving  different 
treatment,  training  and  benefits),  it  may  be  no  surprise  that 
identification with co‐workers is low (Dubouloy, 2002).  
 
However,  HPl  who  aspire  to  be  HPh  someday  are  recommended  to 
develop their interpersonal skills. An examination of the EQ differences 
between  the  four  sample  categories  revealed  that  high  potentials  at 
lower managerial levels display lower levels of interpersonal skills than 
managers, but this relationship is inverted at higher managerial levels. A 
possible explanation may be that HPl prioritize individual performance 
over  interpersonal  aspects  of  work  while  Ml  experience  less  pressure 
from higher management and longer socialization periods. Furthermore, 
as  Ml  rate  their  own  job  performance  lowest  of  all  sample  categories, 
relationship  management  (e.g.  networking)  may  be  a  way  of 
compensating  for  average  job  performance  (Luthans  et  al.,  1985).  At 
higher  managerial  levels,  HP  report  better  interpersonal  skills  than  do 
managers  (Mh  display  the  lowest  scores).  A  reference  to  derailment 
128
literature  seems  appropriate  here.  Although  conclusions  should  be 
interpreted  with  caution  (as  the  study  design  is  cross‐sectional,  not 
longitudinal),  it  appears  that  the  high  potentials  at  the  higher 
managerial  levels  are  those  with  the  better  interpersonal  skills  (or: 
those  that  have  learned  to  prioritize  teamwork  over  individualism  and 
competitiveness).  A  possible  explanation  for  the  fact  that  Mh 
demonstrate  the  lowest  interpersonal  skills  scores  of  all  four  sample 
categories  may  be  that  these  managers’  working  relationships  are 
frustrated  because  they  do  not  get  the  same  opportunities  as  do  their 
high potential peers (Cope, 1998). Alternatively, the Mh group may well 
consist  of  experts  who  have  received  promotions  as  a  result  of  their 
expertise,  but  will  never  be  considered  HP  due  to  their  lack  of 
interpersonal skills. The Ml group, on the other hand, may contain some 
managers  who  might  still  be  identified  as  high  potentials.  Happiness 
exhibits  similar  patterns,  with  high  potentials  scoring  lower  than 
regular  managers  at  lower  managerial  levels  and  higher  at  higher 
managerial  levels.  Research  by  Dubouloy  (2002)  uncovered  that  high 
potentials  often  feel  lonely  and  insecure  as  a  result  of  the  lack  of 
feedback  they  receive  and  the  ambiguous  expectations  their 
organizations  hold  towards  them.  It  is  possible  that  such  negative 
feelings  diminish  as  high  potentials  grow  accustomed  to  their  special 
statuses and all the ‘pros and cons’ that go with it. The inversion of the 
happiness‐HP  status  relationship  at  higher  levels  may,  again,  be 
attributed  to  frustrations  on  the  side  of  Mh.  Another  possible 
explanation lies in the fact that high potentials generally make different 
lifestyle  choices  (or  at  least  set  different  priorities)  than  non‐high 
potential  peers  and,  therefore,  suffer  less  of  imbalances  between  work 
and private life (Judge et al., 1994). 
 
The roles of job performance and career commitment 
 
High potentials were found to rate their own performance significantly 
higher  than  regular  managers  did,  at  lower  as  well  as  at  higher 
managerial  levels.  Much  can  be  said  about  having  survey  participants 
rate  their  own  performance;  however,  these  findings  do  raise  some 
interesting  questions.  First  of  all,  if  social  desirability  or  self‐
enhancement bias really plays such a big part in self‐report performance 
ratings  (Campbell  and  Lee,  1988),  why  would  regular  managers,  then, 
rate their performance significantly lower than high potentials? There is 
no  apparent  reason  why  self‐enhancement  bias  would  affect  high 
potentials more than regular managers. It seems more plausible that the 
self‐reported  performance  ratings  in  this  study  are  either  a  proxy  of 
objective  data  available  to  employees  (such  as  the  feedback  they  get 
129
during  a  performance  appraisal  process),  or  a  reflection  of  the 
perceptions  the  study  participants  have  of  themselves  owing  to  their 
knowledge of their HP status. Another angle from which to look at these 
findings is based on cognitive dissonance theory (Festinger, 1956). As it 
is  commonly  known  that  job  performance  is  the  ‘overriding’  high 
potential  identification  criterion  (Cope,  1998;  Pepermans  et  al.,  2003; 
Quinn,  1988),  employees  might  use  information  on  HP  status  to  infer 
how  well  they  perform  compared  to  peers.  When  conclusions  are 
negative,  they  might  solve  the  resulting  psychological  discomfort  by 
instilling  in  themselves  the  mindset  of  prioritizing  interpersonal 
relationships over performance – which is a possible explanation for the 
fact that Ml report higher scores on interpersonal skills than do HPl. 
 
With  regard  to  the  relationship  between  EQ  and  performance,  results 
indicated  that  reality  testing,  optimism  and  emotional  self‐awareness 
make up a significant model. The less realistic and the more optimistic, 
it appears, the higher one’s job performance rating. Reasons for this may 
be twofold: first, it is possible that ‘unrealistic optimists’ overrate their 
own performance when asked to self‐report on it; second, it may well be 
that these attributes enable them to ‘think outside the box’ and come up 
with  creative,  innovative  solutions  that  their  more  conservative  peers 
would  never  have  considered  (Rottinghaus  et  al.,  2005;  Schneider, 
2001). Such innovative ideas would be much valued in today’s business 
environment  and  result  in  higher  performance  ratings  (Luthans  et  al., 
2004).  Finally,  higher  emotional  self‐awareness  might  lead  to  higher 
self‐reported performance ratings as employees who are aware of their 
own  strengths  and  weaknesses  might  benefit  from  this  knowledge  to 
‘market’ themselves as high performers – acting on their strengths and 
drawing attention away from their weaknesses (Spreitzer et al., 1997).  
 
Career  commitment  does  not  strongly  impact  self‐reported  job 
performance; only at item level, higher performance rates were found in 
those  participants  who  gave  low  ratings  to  “If  I  could  get  another  job 
and be paid the same amount, I would probably take it” and “If  I could 
do  it  all  over  again,  I  would  not  choose  to  work  in  this  profession”.  It 
seems  that  those  who  have  once  made  a  very  conscious  choice  for  the 
current  profession,  report  the  highest  performance  ratings,  a  finding 
that makes obvious sense (Aryee et al., 1994; Bashaw and Grant, 1994).  
 
In  the  current  study,  all  subscales  of  the  EQ‐i,  except  for  those  in  the 
adaptability  scale  are  significantly  related  to  the  career  commitment 
scale. Furthermore, high potentials differ from managers in this respect 
for the item “If I could get another job and be paid the same amount, I 
130
would  probably  take  it”,  which  they  rated  lower  than  managers  did. 
Some  reference  to  the  literature  on  boundaryless  and  protean  careers 
seems  relevant  here.  High  potentials  seem  to  experience  protean,  but 
not per se boundaryless careers. Briscoe et al. (2006) argue that protean 
and  boundaryless  careers  are  independent  but  related  constructs: 
people  can  display  protean  attitudes  but  not  prefer  cross‐boundary 
collaboration  and  vice  versa.  They  found  that  being  protean  and 
boundaryless  in  terms  of  career  attitudes  is  not  synonymous  with  job 
mobility  preference  (as  it  is  often  depicted  in  literature)  and  that, 
consequently,  mobility  should  not  be  used  as  the  primary  proxy  for 
either  boundaryless  or  protean  career  attitudes  or  outcomes.  The 
possibility of a connection between career commitment and traditional 
(non‐boundaryless) career attitudes is supported by the fact that the EQ 
subscales within the adaptability scale are those that would, logically, be 
expected to relate most to boundaryless career attitudes.  
 
Implications for practice 
 
Drawing from the findings in this paper, it appears that using emotional 
intelligence (as conceptualized by Bar‐On, 1997) – or at least some of its 
subscales  –  in  identifying  high  potential  may  well  contribute  to  the 
validity of such processes. Concluding that assertiveness, independence, 
optimism and flexibility are attributes high potentials should possess to 
an  above‐average  extent  is,  perhaps,  not  very  new  or  shocking. 
However, the heart of the matter is, are these attributes really assessed in 
organizations today? Briscoe and Hall (1999) described the competency 
models  they  encountered  in  their  in‐depth  study  of  31  leading  North 
American  organizations  as  ‘too  complex’  and  drowning  “in  overly 
detailed competency definitions” (p. 48). The authors found that not one 
of  these  companies  assessed  learning  ability  in  any  way,  even  though 
most of them acknowledged the paramount importance of a continuous 
learning approach to the success of their organizations in the future. A 
similar discourse is found in Spreitzer et al. (1997). The EQ‐i might help 
organizations  and  individuals  alike  assess  the  potential  they  have  for 
continuous  learning  in  a  very  simple  manner.  If  nothing  more,  the  EQ 
profiles  that  emerge  can  provide  valuable  input  into  the  high  potential 
identification process as well as into individuals’ personal development 
plans.  Many  organizations  probably  do  have  competencies  labeled 
independence  or  flexibility  in  their  competency  models.  However, 
looking  deeper  into  the  definitions  of  such  competencies,  their  ad  hoc 
nature  is  generally  revealed.  Such  ‘micro‐competencies’  are  mostly 
unrelated  to  the  meanings  of  independence  and  flexibility  in  the 
metacompetency approach (Briscoe and Hall, 1999).  
131
 
Another  important  implication  for  practice  is  that  organizations  must, 
very carefully, separate high potentials from high performers. Potential 
for promotion and job performance have been found to be correlated in 
previous studies, but correlation coefficients did not exceed .40 (Fields, 
2002).  Although  an  excellent  performance  is  and  should  be  a 
prerequisite for being identified as having high potential, organizations 
must  beware  of  halo  effects  in  their  performance  appraisal  processes, 
causing assessors to align potentiality ratings with performance ratings 
(Fields, 2002). As is seen in the data presented above, a too great focus 
on performance may cause high potentials to pay little attention to the 
development  of  interpersonal  skills,  which  is  a  typical  cause  of 
derailment (Leslie  and  Van  Velsor,  1996;  Van  Velsor and  Leslie, 1995). 
Organizations need to be proactive in this respect, making sure that high 
potentials do not get isolated in the beginning of their careers, fostering 
individualistic  and  competitive  mindsets  that  will  be  difficult  to 
‘unlearn’  later.  Furthermore,  in  light  of  retention  management,  paying 
sufficient  attention  to  employees’  emotional  states  (e.g.  happiness, 
frustration)  might  benefit  organizations  more  than  focusing  on 
compensation  and  benefits.  This  point  is  demonstrated  by  the  finding 
that high potentials rate the item “If I could get another job and be paid 
the same amount, I would probably take it” low, implying that pay is not 
their primary motive for staying with the organization – and that their 
career  preferences  are  perhaps  not  as  boundaryless  as  some  literature 
suggests (Viney et al., 1996). Also, abovementioned findings about high 
potentials’  ‘unhappiness’  and  non‐high  potential  frustrations  are 
manifestations  of  the  risks  inherent  to  HP  identification  policies  based 
on  a  system  of  peer  comparison  or  forced  ranking  appraisal  (Cope, 
1998). The relevance of cultivating positive emotions at work is further 
supported  by  the  finding  that  optimism  and  (a  low  score  on)  reality 
testing show a significant relationship with individual job performance.  
 
Limitations and directions for further research 
 
Despite the contributions of the current study to the literature on high 
potential identification, it is not without limitations. First of all, the small 
sample size per sample category (n = 24, 24, 27, 27) affected the power 
of  our study, which was  between .3 and .4 (depending on sample  size) 
for  the  MANOVA  analyses  and  around  .6  for  the  t‐tests  (n  =  51,  51). 
Generally,  .8  is  accepted  as  a  standard  for  adequacy  (Cohen,  1988). 
Furthermore,  using  data  from  high  potentials  coming  from  different 
organizations may have introduced some additional variance in the data 
(which  enhances  the  chance  of  chance  capitalization  and  thus  might 
132
decrease the power of the study). However, if we would have restricted 
data  collection  to  only  one  organization,  we  would  have  had  even 
smaller  samples.  We  have  tried  to  reduce  the  chance  of  inter‐
organizational  heterogeneity  regarding  what  ‘high  potential’  means  by 
explicitly  communicating  to  organizations  that  they  could  only 
participate in the study if their definition of high potential matched that 
of  Cope  (1998).  However,  this  particular  research  population  is  very 
difficult to reach. Our study was a first attempt to study a high potential 
sample  in  comparison  to  a  non‐high  potential  sample  in  a  way  that 
would  not  compromise  discretion  within  organizations.  It  is,  by  our 
knowledge,  the  first  study  that  surveys  the  high  potential  population 
itself.  However,  in  the  future,  we  hope  to  be  able  to  collect  data  from 
larger samples, in order to obtain higher power. Further research must 
then reveal if the effects that were found can, in effect, be replicated. It 
would be interesting to see if follow‐up studies on larger high potential 
samples  (and  with  a  higher  power)  would  yield  similar  results, 
especially  with  respect  to  effect  size,  which  was  around  .4  for  all  EQ‐i 
subscales (this is a rather small effect size, see Cohen, 1988). However, 
even weak effects may be of empirical importance (Garamszegi, 2006) – 
but again, follow‐up studies using high potentials  as study participants 
would  allow  researchers  to  build  a  much  stronger  case  about  the  link 
between emotional intelligence and high potential.  
 
Second,  we  have  not  applied  a  Bonferroni  correction  to  our  data  even 
though  the  study  incorporates  multiple  comparisons.  As  our  study 
explored a novel topic by means of a survey (where study variables are 
more  complex  and  interrelated  than  in  experimental  designs),  such  a 
correction would be too conservative. Anderson (1961) advocated that 
pilot  studies  should  focus  on  maximizing  their  power,  whilst  Perneger 
(1998)  argues  that  the  Bonferroni  method  and  the  concept  of  study‐
wide  error  rate  applies  only  to  the  ‘general’  or  ‘universal  null 
hypothesis’,  i.e.  that  all  null  hypotheses  are  true  simultaneously,  which 
is not really of interest to us in this study, as we wanted to assess each 
subscale in its own right. Furthermore, applying corrections for multiple 
comparison  increases  the  likelihood  of  Type  II  errors  (i.e.  concluding 
that there are no significant differences between the different groups in 
the study when in fact, there are), leading to a ‘sacrificial loss of power’ 
(Garamszegi, 2006). 
 
Third,  both  performance  and  emotional  intelligence  scores  were 
assessed using self‐report measures, which run a higher risk of inducing 
self‐enhancement  bias.  We  wanted  to  incorporate  some  indication  of 
performance  in  our  study,  and  using  a  measure  that  was  not  self‐
133
reported  would  have  hindered  our  data  collection  even  more  (as  we 
would  have  needed  to  link  archival  data  or  supervisory  ratings  to  our 
anonymous participants). Therefore we opted for a self‐report measure, 
even though this was not an ideal solution. Nevertheless, we do feel that 
the results section of this paper offers an interesting discussion of data 
on  high  potentials’  versus  regular  managers’  performance  scores  – 
although  we  cannot  be  sure  that  the  construct  that  was  measured  was 
actually  ‘performance’,  which  somewhat  limits  the  validity  of  our 
findings.  However,  in  the  EQ‐i,  there  is  some  control  over  ‘fake  good’ 
behaviour as the inventory has two additional validity factors built into 
it: the Positive Impression (or PI) scale and the Negative Impression (or 
NI)  scale  which  control  for  desirable  inclinations  or  impressions  that 
people project. The conversion of raw EQ‐i data sets into standardized t‐
scores  (by  the  publisher)  takes  these  two  validity  factors  into  account. 
An  ‘Inconsistency  Index’  is  calculated  for  each  data  set  and  the  EQ‐i  t‐
scores  are  adjusted  statistically  (downward  or  upward)  taking  into 
account the values of PI and NI – these adjusted t‐scores then make up 
the final data set that is returned to researchers. We believe that this is a 
valid method for reducing the chance of self‐enhancement bias affecting 
the results.  
 
In  the  current  study,  emotional  intelligence  as  measured  by  the  Bar‐ 
On’s  EQ‐i  (1997)  served  mainly  as  a  proxy  for  Briscoe  and  Hall’s 
metacompetency model (1999). As a result, the interpersonal aspect of 
EQ was underplayed in this paper. Even though interpersonal skills are 
extremely important in  high potential careers (as the primary cause  of 
career derailment is a lack of these skills), a broader scope was applied 
in the current study. The EQ‐i, although not without disadvantage – see 
earlier for a discussion of common criticism on the measure – is a very 
broad instrument in nature and fit this scope very well. Aside from the 
apparent  overlap  of  Bar‐On’s  EQ  personal  factors  model  (1997)  with 
identity  and  adaptability  metacompetencies  (Briscoe  and  Hall,  1999), 
using  the  EQ‐i  as  a  proxy  for  learning  agility  (and  letting  participants 
rate  their  own  performance)  allowed  for  the  entire  study  to  be  self‐
reported. Typically, the viewpoint of the individual tends to be neglected 
in  research  on  high  potentials  and  self‐report  studies  are  very 
uncommon due to the fact that this particular population is quite hard to 
reach  (Pepermans  et  al.,  2003).  Obviously,  the  study  domain  would 
benefit  from  more  empirical  findings  coming  from  the  side  of  high 
potential individuals. 
 
It  might  be  interesting  to  validate  the  EQ‐i  against  other  proxies  or 
measures of learning agility and so find support for the findings in this 
134
paper. Additionally, more research needs to be done on the capacity of 
learning  metacompetencies  to  identify  leadership  potential.  Finally,  an 
important  question  in  research  on  high  potential  identification  is  to 
which  extent  certain  attributes  (in  this  case,  EQ  traits)  are 
predispositions or skills that can be developed (e.g. Dulewicz and Higgs, 
2004).  Further  research  needs  to  be  done  in  order  to  confront  this 
question. 
 
References 
 
Anderson,  N.H.  (1961),  “Scales  and  statistics:  parametric  and  nonparametric”,  Psychological 
Bulletin, Vol. 58 No. 4, pp. 305‐16. 
Aryee,  S.,  Chay,  Y.W.,  Chew,  J.  (1994),  “An  investigation  of  the  predictors  and  outcomes  of  career 
commitment in three career stages”, Journal of Vocational Behaviour, Vol. 44 No. 1, pp. 1‐16. 
Barling,  J.,  Slater,  F.,  &  Kelloway,  K.  (2000),  “Transformational  Leadership  and  emotional 
intelligence: an exploratory study”, Leadership and Organization Development Journal, Vol. 21 
No. 3, pp. 157‐61. 
Bar‐On, R. (1997), Emotional Quotient Inventory: Technical Manual, Multi‐Health Systems, Toronto.  
Bar‐On,  R.  (2000),  “Emotional  and  social  intelligence:  Insights  from  the  Emotional  Quotient 
Inventory (EQ‐i)”, in Bar‐On, R. and Parker, J.D.A. (Eds.), Handbook of emotional intelligence, 
Jossey‐Bass, San Francisco, pp. 363‐88. 
Bar‐On,  R.  (2005),  “The  Bar‐On  model  of  emotional‐social  intelligence”,  Psicothema,  Vol.  18  No.  1, 
pp. 13‐25.  
Bashaw, E. R., & Grant, S. E. (1994), “Exploring the distinctive nature of work commitments: their 
relationships  with  personal  characteristics,  job  performance,  and  propensity  to  leave”, 
Journal of Personal Selling and Sales Management, Vol. 14, No. 2, pp. 41‐56. 
Bennis,  W.  &  Nannus,  B.  (1985),  Leaders:  the  Strategies  for  Taking  Charge.  Harper  and  Row,  New 
York. 
Blau, G. J. (1985), “The measurement and prediction of career commitment”, Journal of Occupational 
Psychology, Vol. 58 No. 4, pp. 277‐88. 
Blau, G. J. (1989), “Testing the generalizability of a career commitment measure and its impact on 
employee turnover”, Journal of Vocational Behavior, Vol. 35 No. 1, pp. 88‐103. 
Bournois, F. & Rousillon, S.  (1998),  Préparer les Dirigeants de Demain. Une Approche Internationale 
de la Gestion des Cadres à Haut Potentiel, Editions d’Organizations, Paris.  
Boyatzis,  R.E.,  Goleman,  D.  &  Rhee,  K.  (2000),  “Clustering  competence  in  emotional  intelligence: 
Insights  from  the  emotional  competence  inventory  (ECI)”,  in  Bar‐On,  R.  and  Parker,  J.D.A. 
(Eds.), Handbook of emotional intelligence, Jossey‐Bass, San Francisco, pp. 343‐62. 
Briscoe, J.P. & Hall, D.T. (1999), “Grooming and Picking Leaders Using Competency Frameworks: Do 
They  Work?  An  Alternative  Approach  and  New  Guidelines  for  Practice”,  Organizational 
Dynamics, Vol. 28 No. 1, pp. 37‐52. 
Brown, C., George‐Curran, R., & Smith, M. L. (2003), “The role of emotional intelligence in the career 
commitment  and  decision‐making  process”,  Journal  of  Career  Assessment,  Vol.  11  No.  4,  pp. 
379‐92. 
Brown,  F.W.  &  Moshavi,  D.  (2005),  “Transformational  leadership  and  emotional  intelligence:  a 
potential  pathway  for  an  increased  understanding  of  interpersonal  influence”,  Journal  of 
Organizational Behavior, Vol. 26 No. 8, pp. 867‐71. 
Buckingham,  M.  &  Vosburgh,  R.  (2001),  “The  21st  Century  Human  Resources  Function:  It’s  the 
Talent, Stupid!”, Human Resource Planning, Vol. 24 No. 4, pp. 17‐23. 
Campbell,  D.  &  Lee,  C.  (1988),  "Self  appraisal  in  performance  evaluation:  development  vs. 
evaluation", Academy of Management Review, Vol. 13 No. 2, pp. 302‐14. 
Capowsky,  G.  (1994),  "Anatomy  of  a  leader:  where  are  the  leaders  of  tomorrow?",  Management 
Review, Vol. 84, pp.10‐17. 
Carmeli, A. (2003), “The Relationship between Emotional Intelligence and Work Attitudes, Behavior 
and Outcomes: An Examination  among Senior Managers”,  Journal  of Managerial Psychology, 
Vol. 18 No. 8, pp. 788‐813. 
Carson  K.D.  &  Carson  P.P.  (1998),  “Career  commitment,  competencies  and  citizenship”,  Journal  of 
Career Assessment, Vol. 6 No. 2, pp. 195–208. 
135
Cohen, J. (1988), Statistical power analysis for the behavioral sciences, Erlbaum, Hillsdale, NJ. 
Cope,  F. (1998),  “Current issues in selecting high potentials”, Human Resource Planning, Vol. 21  No. 
3, pp. 15‐6. 
Côté,  S.  &  Miners,  C.T.H.  (2006),  “Emotional  Intelligence,  Cognitive  Intelligence,  and  Job 
Performance”, Administrative Science Quarterly, Vol. 51 No. 1, pp. 1‐28. 
Daft, R.L. (1999), Leadership: Theory and Practice, The Dryden Press, Forth Worth. 
Daus, C.S. & Ashkanasy, N.M. (2005), “The case for the ability‐based model of emotional intelligence 
in organizational behaviour”, Journal of Organizational Behavior, Vol. 26 No. 4, pp. 452‐66. 
Dawda, D. & Hart, S.D. (2000), “Assessing emotional intelligence: reliability and validity of the Bar‐
On  Emotional  Quotient  Inventory  (EQ‐i)”,  Journal  of  Personality  and  Individual  Differences, 
Vol. 28 No. 4, pp. 797‐812.  
Dubouloy, M. (2006),  “Les  hauts  potentiels  et  le  faux‐self”,  Journal  des  Psychologues,  Vol.  236,  pp. 
22‐6. 
Dulewicz, V. & Higgs, M.J. (2000), “Emotional intelligence: a review and evaluation study”, Journal of 
Managerial Psychology, Vol. 15 No. 4, pp. 341‐68. 
Dulewicz, V. & Higgs, M.J. (2003), “A new approach to assessing leadership  dimensions, styles and 
context”, Competency & Emotional Intelligence Quarterly, Vol. 11 No. 2, pp. 24‐32. 
Dulewicz, V. & Higgs, M. (2004), “Can Emotional Intelligence be developed?” International Journal of 
Human Resource Management, Vol. 15 No. 1, pp. 95‐111. 
Dulewicz,  C.,  Young,  M.  &  Dulewicz,  V.  (2005),  “The  relevance  of  emotional  intelligence  for 
leadership performance”, Journal of General Management, Vol. 30 No. 3, pp. 71‐86. 
Eagly,  A.H.  &  Johnson,  B.T.  (1990),  “Gender  and  leadership  style:  A  meta‐analysis”,  Psychological 
Bulletin, Vol. 108 No. 2, pp. 209‐32. 
Feild, H. S. & Harris,  S. G. (1991), “Participants' frustrations in a fast‐track development program”, 
Leadership and Organization Development Journal, Vol. 12 No. 4, pp. 3‐8. 
Festinger, L. (1957), A theory of cognitive dissonance, Stanford University Press, Stanford. 
Fields,  G.  S.  (2002),  “Predicting  potential  for  promotion:  How  the  data  in  human  resource 
information systems can be used to help organizations gain competitive advantage”, CAHRS’ 
Working  Paper  Series,  Working  Paper  02‐14,  July.  Available 
http://www.ilr.cornell.edu/depts/cahrs/downloads/ pdfs/workingpapers/WP02‐14.pdf 
Frankel, L.P. (1994), “Preventing Individuals' Career Derailment”, Employment Relations Today, Vol. 
21, pp. 295‐306. 
Freudenthaler,  H.H.  &  Neubauer,  A.C.  (2005),  “Emotional  intelligence:  The  convergent  and 
discriminant  validities  of  intra‐  and  interpersonal  emotional  abilities”,  Personality  and 
Individual Differences, Vol. 39 No. 3, pp. 569‐79. 
Garamszegi, L.Z. (2006), “Comparing effect sizes across variables: generalization without the need 
for Bonferroni correction”, Behavioral ecology, Vol. 17 No. 4, pp. 682‐87. 
Gardner,  L.  &  Stough,  C.  (2002),  “Examining  the  relationship  between  leadership  and  emotional 
intelligence  in  senior  level  managers”,  Leadership  &  Organization  Development  Journal,  Vol. 
23 No. 2, pp. 68‐78. 
Goleman, D. (1998a), “What makes a leader?”, Harvard Business Review, Vol. 76 No. 6, pp. 93‐102. 
Goleman, D. (1998b), Working with emotional intelligence, Bantam Books, New York.  
Goleman,  D.  (2001),  “Emotional  intelligence:  Issues  in  paradigm  building”,  in  Cherniss,  C.  & 
Goleman, D. (Eds), The emotionally intelligent workplace, Jossey‐Bass, San Francisco, pp. 13‐
26. 
Goleman,  D.,  Boyatzis,  R.,  and  McKee,  A.  (2002),  “The  emotional  reality  of  teams”,  Journal  of 
Organizational Excellence, Vol. 21 No. 2, pp. 55‐65. 
Gowing,  M.K.,  O’Leary,  B.S.,  Brienza,  D.,  Cavallo,  K.  &  Crain,  R.  (2006),  “A  Practitioner’s  Research 
Agenda: Exploring Real‐World Applications and Issues”, in Druskat, V.U., Sala, F. & Mount, G. 
(Eds.),  Linking  Emotional  Intelligence  and  Performance  at  Work,  Lawrence  Erlbaum 
Associates, New Jersey, pp. 245‐66.  
Hall,  D.T.  (1999),  “Accelerate  executive  development  –  at  your  peril!”  Career  Development 
International, Vol. 4 No. 4, pp. 237‐39. 
Hall,  D.T.  &  Moss,  J.E.  (1998),  “The  new  protean  career  contract:  Helping  organizations  and 
employees adapt”, Organizational Dynamics, Vol. 26 No. 3, pp. 22‐37.  
Hedlund,  J.,  &  Sternberg,  R.  J.  (2000),  “Too  many  intelligences?  Integrating  social,  emotional,  and 
practical  intelligence”,  in  Bar‐On,  R.  and  Parker,  J.D.A.  (Eds.),  Handbook  of  emotional 
intelligence, Jossey‐Bass, San Francisco, pp. 136‐68. 
Hogan,  R.  &  Hogan,  J.  (2001),  “Assessing  leadership:  A  view  from  the  dark  side”,  International 
Journal of Selection and Assessment, Vol. 9 No. 1, pp. 40–51. 
136
Jordan,  P.J.  &  Troth,  A.C.  (2004),  “Managing  emotions  during  team  problem  solving:  emotional 
intelligence and conflict resolution”, Human Performance, Vol. 17 No. 2, pp. 195–218. 
Judge,  T.,  Boudreau,  J.,  &  Bretz,  R.  (1994),  “Job  and  life  attitudes  of  male  executives”,  Journal  of 
Applied Psychology, Vol. 79 No. 5, pp. 767‐82. 
Kilpatrick,  F.  P.,  &  Cantril,  H.  (1960),  “Self‐anchoring  scaling:  A  measure  of  individuals'  unique 
reality worlds”, Journal of Individual Psychology, Vol. 16, pp. 158‐73. 
Kovach, B.E. (1986), “The Derailment of Fast‐Track Managers”, Organizational Dynamics, Vol. 15 No. 
2, pp. 41‐8. 
Kovach,  B.E. (1989), “Successful  Derailment: What  Fast‐Trackers Can Learn While They’re Off the 
Track”, Organizational Dynamics, Vol. 18 No. 2, pp. 33‐47. 
Lee, T.W., Ashford, S.J., Walsh, J.P. & Mowday, R.T. (1992), “Commitment propensity, organizational 
commitment,  and  voluntary  turnover:  A  longitudinal  study  of  organizational  entry 
processes”, Journal of Management, Vol. 18 No. 1, pp. 15–32. 
Leslie,  J.  B.  &  Van  Velsor,  E.  (1996),  A  look  at  derailment  today,  Centre  for  Creative  Leadership, 
Greensboro.  
Lombardo,  M.M.  &  Eichinger,  R.W.  (2000),  “High  Potentials  as  High  Learners”,  Human  Resource 
Management, Vol. 39 No. 4, pp. 321‐30.  
Luthans,  F.,  Rosenkranz,  S.  &  Hennessey,  H.  (1985),  “What  do  successful  managers  really  do?  An 
observational  study  of  managerial  activities”,  Journal  of  Applied  Behavioral  Science,  Vol.  21 
No. 3, pp. 255‐70. 
Luthans,  F.  &  Youssef,  C.  (2004),  “Human,  social,  and  now  positivepsychological  capital 
management:  Investing in people for competitive advantage”, Organizational Dynamics, Vol. 
33 No. 2, pp. 143‐60. 
Matthews,  G.,  Zeidner,  M.  &  Roberts,  R.D.  (2002),  Emotional  Intelligence:  Science  and  Myth.  MIT 
Press, Cambridge. 
Mayer,  J.D.,  Caruso,  D.R.  &  Salovey,  P.  (1999),  “Emotional  intelligence  meets  traditional  standards 
for an intelligence”, Intelligence, Vol. 27 No. 4, pp. 267‐98. 
McCall,  M.W.  (1994),  “Identifying  Leadership  Potential  in  Future  International  Executives: 
Developing a Concept”, Consulting Psychology Journal, Vol. 46 No. 1, pp. 49‐63. 
McCall, M.W. (1998), High flyers: developing the next generation of leaders, Harvard Business School 
Press, Boston.  
McCall,  M.  W.  &  Lombardo,  M.M.  (1983),  Off  the  Track:  Why  and  How  Successful  Executives  Get 
Derailed, Centre for Creative Leadership, Greensboro.  
McClelland,  D.  (1998),  “Identifying  Competencies  with  Behavior‐Event  Interviews”,  Psychological 
Science, Vol. 9 No. 5, pp. 331‐40. 
McEnrue, M.P. & Groves, K. (2006), “Choosing Among Tests of Emotional Intelligence: What Is The 
Evidence?”, Human Resource Development Quarterly, Vol. 17 No. 1, pp. 9‐42. 
Offerman, L., Bailey, J. R., Vasilopoulos, N. L., Seal, C. & Sass, M. (2004), “EQ versus IQ: The relative 
contribution  of  emotional  intelligence  and  cognitive  ability  to  individual  and  team 
performance”, Human Performance, Vol. 17 No. 2, pp. 219‐43. 
Pepermans,  R.,  Vloeberghs,  D.  &  Perkisas,  B.  (2003),  “High  potential  identification  policies:  an 
empirical study among Belgian companies”, Journal of Management Development, Vol. 22 No. 
8, pp. 660‐78. 
Perneger,  T.V.  (1998),  “What’s  wrong  with  Bonferroni  adjustments”,  British  Medical  Journal,  Vol. 
316, pp. 1236‐37. 
Quinn, R. E. (1988), Beyond rational management, Jossey‐Bass, San Francisco.  
Quinn, R.E., Faerman, S.R., Thompson, M.P. & McGrath,  M.R. (1990), Becoming a Master Manager:  A 
Competency Approach, John Wiley & Sons, New York.  
Randall,  D.M.,  Fedor,  D.B.,  &  Longenecker,  C.O.  (1990),  “The  Behavioral  Expressions  of 
Organizational Commitment”, Journal of Vocational Behavior, Vol. 36 No. 3, pp. 210‐24. 
Rosete,  D.  &  Ciarrochi,  J.  (2005),  “Emotional  intelligence  and  its  relationship  to  workplace 
performance outcomes of leadership effectiveness”, Leadership & Organization Development 
Journal, Vol. 26 No. 5, pp. 388‐99. 
Rottinghaus,  P.  J.,  Day,  S.  X.,  &  Borgen,  F.  H.  (2005),  “The  Career  Futures  Inventory:  A  measure  of 
career‐related adaptability and optimism”, Journal of Career Assessment, Vol. 13 No. 1, pp. 3‐
24. 
Salovey,  P.  &  Mayer,  J.D.  (1990),  “Emotional  intelligence”,  Imagination,  Cognition,  and  Personality, 
Vol. 9 No. 3, pp. 185‐211. 
Schneider,  S.L.  (2001), “In  search  of  realistic  optimism:  Meaning,  knowledge,  and  warm  fuzzies”, 
American Psychologist, Vol. 56 No. 3, pp. 250‐63. 
137
Shulman,  T.  &  Hemenover,  S.  (2006),  “Is  dispositional  emotional  intelligence  synonymous  with 
personality?”, Self and Identity, Vol. 5 No. 2, pp. 147‐71.  
Sosik, J. J. & Mergerian, L. E. (1999), “Understanding leader emotional intelligence and performance: 
the  role  of  self‐other  agreement  on  transformational  leadership  perceptions”,  Group  and 
Organizational Management, Vol. 24 No. 3, pp. 367–90. 
Spreitzer, G.M., McCall, M.W. & Mahoney, J.D. (1997), “Early identification of international executive 
potential”, Journal of Applied Psychology, Vol. 82 No. 1, pp. 6‐29. 
Tett, R. P., Guterman, H. A., Bleier, A., & Murphy, P. J. (2000), “Development and content validation 
of a “hyperdimensional” taxonomy of managerial competence”, Human Performance, Vol. 13 
No 3, pp. 205–51. 
Trochim, W. (2000), The Research Methods Knowledge Base. Atomic Dog Publishing, Cincinnati. 
Van Rooy, D.L., Viswesvaran, C. & Pluta, P. (2005), “An Evaluation of Construct Validity: What Is This 
Thing Called Emotional Intelligence?”, Human Performance, Vol. 18 No. 4, pp. 445‐62. 
Van Velsor, E. & Leslie, J.B. (1995), “Why executives derail: perspectives across time  and cultures”, 
Academy of Management Executive, Vol. 9 No. 4, pp. 62‐72. 
Viney,  C.,  Adamson,  S.  and  Doherty,  N.  (1997),  “Paradoxes  of  fast‐track  career  management”, 
Personnel Review, Vol. 26 No. 3, pp. 174‐86. 
Williams, D. (1994), Leadership for the 21st Century: Life Insurance Leadership Study, HayGroup, 
Boston. 
 
138
139
Paper 4. 
 

High potential identification: Examining the 
developmental perspective  
 
 
In this paper, we empirically examine the developmental perspective on high potential 
identification,  which  states  that  learning  agility  is  the  best  indicator  of  leadership 
potential.  63  supervisors  (i.e.  ‘raters’)  completed  an  online  survey  about  their 
subordinates  (i.e.  ‘ratees’),  which  were  classified  as  either  high  potential  or  non‐high 
potential.  This  classification  variable  was  not  disclosed  to  the  raters  and  ratees  at  the 
time of survey administration. A learning agility scale was incorporated in the survey, as 
were  measures  of  on‐the‐job  learning,  job  performance  and  career  variety.  We  found 
empirical  support  for  all  hypotheses.  Supervisor‐rated  scores  of  learning  agility  were 
found  to  significantly  predict  whether  a  ratee  was  identified  by  the  organization  as  a 
high  potential  or  not.  This  relationship  was  partially  mediated  by  (actual,  observable) 
on‐the‐job  learning.  Learning  agility  predicted  leadership  potential  above  and  beyond 
job  performance.  Finally,  although  learning  agility  was  found  to  be  related  to  career 
variety,  results  seem  to  indicate  that  it  cannot  be  developed  with  age  or  experience. 
However, since our design was non‐longitudinal, we cannot make causal inferences with 
certainty. To date, there has been very little research about employees identified as high 
potentials.  This  paper  aims  to  contribute  to  the  talent  management  literature  by 
building  arguments  based  on  empirical  data  rather  than  conjecture.  Furthermore,  we 
discuss the selection versus the developmental perspectives on talent management and 
their implications for the ongoing ‘war for talent’.  
 
 
 

 
140

Figure V. Topicality of Paper 4.
141
Paper 4. 
 
High potential identification: Examining the developmental 
perspective 
 
It  is  widely  accepted  that  we  are  currently  operating  in  a  knowledge 
economy  in  which  the  intellectual  capital  of  organizations,  shaped  by 
their  employees’  knowledge  and  skill  levels,  is  a  decisive  factor  for 
organizational  success.  Consequently,  identifying,  developing  and 
retaining  employee  talent  (i.e.  ‘talent  management’)  is  increasingly 
being  recognized  as  a  prime  source  of  sustainable  competitive 
advantage  (Boudreau  and  Ramstad,  2005;  Frank  and  Taylor,  2004; 
Lewis  and  Heckman,  2006).  While  talent  management  is  generally 
considered to be quite broad in definition, literature in the  field seems 
to  focus  primarily  on  leadership  potential  –  more  specifically  on  the 
concept of high potentials, i.e. “those individuals within an organization 
who  are  recognized,  at  that  point  in  time,  as  the  organization’s  likely 
future  leaders”  (Cope,  1998,  p.  15).  It  has  been  suggested  that 
identifying  a  pool  of  high  potentials  can  help  to  ensure  continuity  of 
management  resources  and  boost  the  retention,  motivation  and 
performance of this specific group (Jerusalim and Hausdorf, 2007).  
 
Although,  in  the  current  business  context,  talent  management  is 
considered  more  crucial  than  ever  to  organizational  survival,  signs 
indicate  that  it  is  also  becoming  more  challenging  (Buckingham  and 
Vosburgh,  2001).  The  increase  in  complexity  of  the  business 
environment asks for a different view on talent, especially in the field of 
leadership. Whereas before, a certain technical expertise could suffice to 
give  direction  to  a  team  of  subordinates,  leadership  today  is 
considerably  affected  by  external  factors  and  events.  The  success  of  a 
leader, then, will depend on his or her capacity to handle these events, 
and  to  their  ability  to  learn  from  experience  (Briscoe  and  Hall,  1999; 
McCall, 1998).  
 
High potential identification: From a selection to a developmental 
perspective 
 
The importance of leader adaptability and learning has been underlined 
in  a  number  of  publications  (e.g.  Briscoe  and  Hall,  1999;  Dries  and 
Pepermans,  2007;  McCall,  1998;  Peters  and  Smith,  1996;  Spreitzer, 
McCall  and  Mahoney,  1997).  In  tandem,  the  literature  has  identified  a 
shift  from  a  selection  perspective  to  a  developmental  perspective  on 
high potential identification.  
 
142
According to the selection perspective, employees either ‘have it or not’ 
(‘it’ being some sort of intangible ‘right stuff’ candidates must possess), 
implying that young talent should be able to demonstrate competencies 
similar to those of successful executives. However, the use of a more or 
less fixed set of competencies to identify high potentials is problematic 
because of the notion of certainty on which it is based. First of all, it is 
not  at  all  clear  to  which  extent  organizations  can  expect  junior 
employees  to  exhibit  the  same  competencies  as  senior  managers,  even 
in a seminal form. Second, it is not at all clear whether the competencies 
leading to success today will do so in the unpredictable future. Briscoe 
and  Hall  (1999),  in  their  study  of  31  organizations  engaged  in  talent 
management, found that they took an average of 4.1 years to revise their 
competency  frameworks,  far  too  long  to  keep  up  with  many  of  the 
unfolding  business  and  leadership  challenges  encountered  by  today’s 
organizations. The authors state that organizations would be better off 
focusing  on  so‐called  learning  ‘metacompetencies’,  required  to  develop 
the just‐in‐time competencies their future leaders will need in order to 
adapt to ongoing, short‐term challenges and the personal competencies 
that will help them endure and lead through multiple waves of change. 
The  developmental  perspective  on  high  potential  identification,  then, 
argues that leadership potential is best predicted by the motivation and 
ability to learn from experience (e.g. McCall, 1998).  
 
Table  1  identifies  a  number  of  concepts  from  the  literature,  related  to 
the  concept  of  learning  metacompetency.  On  the  one  hand,  it  lists 
concepts such as ‘general mental ability’ (Wechsler, 1944) and ‘learning 
orientation’  (Porter  and  Tansky,  1999),  which  have  a  long  tradition  in 
the field of education; on the  other  hand, it describes concepts such as 
‘adaptive  competence’  (Hall,  2002)  and  ‘employability’  (Fugate,  Kinicki 
and  Ashforth,  2004),  explicitly  aimed  at  people  in  a  work  setting. 
Furthermore,  some  authors  also  make  a  distinction  between  ‘hard 
learning’  (structural‐technical  learning  of  explicit  knowledge)  and  ‘soft 
learning’  (social‐relational  learning  of  ‘tacit’,  intangible  knowledge  and 
cognitive  styles),  where  the  former  relates  mostly  to  intelligence,  and 
the  latter  mostly  to  adaptability  and  flexibility  (e.g.    Lombardo  and 
Eichinger, 2000).  
 
In this paper, we will use the term ‘learning agility’ (as in Lombardo and 
Eichinger,  2000)  as  an  operationalization  of  learning  metacompetency. 
We  chose  this  operationalization  as  it  is  specifically  grounded  in  the 
work  context,  and  integrates  aspects  of  hard  learning  as  well  as  soft 
learning.  More  background  information  on  the  learning  agility  concept 
will be given in the Methodology section.  
Table 1. Overview of concepts related to ‘learning metacompetency’  
 
Reference    Related concept  Definition  
Wechsler   1944  intelligence/general mental  “the aggregate or global capacity of the individual to act purposefully, to think 
ability  rationally and to deal effectively with his environment” (p. 3) 
Kolb   1984  experiential learning  “the process whereby knowledge is created through the transformation of 
experience; knowledge results from the combination of grasping and transforming 
experience" (p. 41) 
Wagner &  1991  practical intelligence/tacit  “practical know‐how that rarely is expressed openly or taught directly” (p. 1)
Sternberg   knowledge 
Sloan   1994  versatility  “adapting and performing effectively across a wide range of new situations and 
changing contexts” (p. 28) 
Bar‐On   1997  emotional intelligence  “an array of noncognitive capabilities, competencies and skills that influence one’s 
ability to succeed in coping with environmental demands and pressures” (p. 14). 
143

Spreitzer,  1997  the ability to learn from  “seek opportunities to learn, seek feedback, culturally adventurous, open to criticism, 


McCall &  experience  and flexible” (p. 13) 
Mahoney  
Taylor   1997  learning potential  “an individual’s fundamental cognitive capabilities and potentialities used to master 
new cognitively demanding material in a formal educational or training context. 
Learning potential does not focus on acquired skills or abilities but on the ability to 
learn new skills and gain knowledge” (p. 1) 
London &  1999  continuous learning  “seek information to identify skill gaps, recognize areas to improve current 
Smither   performance, keep up with advances in their profession, and anticipate how changes 
elsewhere in the firm and the industry may affect work demands and skill 
requirements (p. 4) 
Table 1. Continued  
 
Reference    Related concept  Definition  
Murphy &  1999  role flexibility  “the willingness and ability to adapt to diverse and changing expectations” (p. 345)
Jackson  
Porter &  1999  learning orientation  “receptiveness to training and development, reactions to feedback, and the potential 
Tansky   to be self‐managing” (p.50)  
Lombardo  2000  learning agility  “the willingness and ability to learn new competencies in order to perform under 
& Eichinger   first‐time, tough, or different conditions” (p. 323) 
Pulakos et  2000  adaptive performance  “handling emergencies or crisis situations; learning work tasks, technologies, and 
al.  procedures ; handling work stress; demonstrating interpersonal adaptability; 
displaying cultural adaptability; solving problems creatively; dealing effectively with 
unpredictable or changing work situations; and demonstrating physically oriented 
adaptability” (p. 617) 
144

Hall   2002  adaptive competence  “developing appropriate behavioral responses to environment” (p. 208)


Fugate,  2004  employability  “a host of person‐centered constructs that combine synergistically to help workers 
Kinicki &  effectively adapt to the myriad of work‐related changes occurring in today’s 
Ashforth   economy” (p. 15) 
White et al.   2005  adaptability  “an effective change in response to an altered situation” (p. 2)
Kaiser,  2007  behavioral flexibility  “a capability to adapt to new, different, or changing requirements. The range of 
Lindberg &  specific behaviors one is capable of enacting is less important here than the generic 
Craig   tendency to vary one’s approach” (p. 41) 
 
145
Research model and hypotheses 
 
Figure  1 offers a schematic overview  of the hypothesized relationships 
examined  in  this  study.  As  mentioned  above,  recent  publications  have 
advocated that possibly the single most important characteristic of high 
potentials, and one that  cuts across all industries and functions, is that 
they recognize the value  of taking on  new projects and are willing and 
able  to  apply  their  new  experiences  immediately.  High  potentials  have 
been found to be risk takers, who expose themselves to a variety of new 
experiences,  and  are  driven  to  learn.  Therefore,  it  has  been  suggested 
that learning agility should be a crucial determinant in identifying high 
potentials (Dries and Pepermans, 2007; Lombardo and Eichinger, 2000; 
Peters  and  Smith,  1996;  Spreitzer  et  al.,  1997).  However,  it  is  unlikely 
that a large fraction of organizations explicitly assess learning agility in 
their  talent  management  programs.  Previous  research  has 
demonstrated  that  organizational  procedures  for  identifying  high 
potentials  are  based  largely  on  indicators  of  performance  and  ‘regular’ 
competencies  (e.g.  Bossuyt  and  Dries,  2008;  Briscoe  and  Hall,  1999; 
Dries and Pepermans, 2008). Rather, it is possible that learning agility is 
some kind of ‘covert’ high potential identification criterion, making up a 
common  factor  among  high  potentials  (Dries  and  Pepermans,  2007). 
This latter view is supported by the finding that most organizations tend 
to identify employees as high potentials at the soonest after two or three 
years after selection, as they first want to observe how well they adapt 
to the specific setting of the organization (Dries and Pepermans, 2008).  
 
H1. Learning agility significantly predicts whether an employee is identified 
as a high potential or not.  
 
As  learning  agility  in  itself  is  a  rather  ‘intangible’  indicator  of  potential 
(i.e.  not  directly  visible  to  supervisors  and  other  significant  parties  in 
the  high  potential  identification  process),  we  posit  that  actual, 
observable  on‐the‐job  learning  will  serve  as  a  mediator  between 
learning agility and being identified as a high potential or not. Learning 
from  on‐the‐job  experiences  is  defined  as  “learning  that  takes  place 
through  participation  in  some  actual  practices  in  the  workplace” 
(Wouters  and  Buyens,  2006,  p.  6).  Two  central  features  can  be 
distinguished.  First,  interactions  with  the  environment,  such  as  task 
accomplishments  and  interpersonal  interaction  are  seen  as  the  major 
sources  of  learning.  Second,  learning  can  take  place  in  and  be 
deliberately  encouraged  by  an  organization,  but  it  is  not  highly 
structured  or  planned  and  control  of  learning  rests  primarily  in  the 
hands  of  the  learner.  It  is  generally  assumed  that  not  all  employees 
benefit equally from on‐the‐job learning experiences.  
 
 
 
 
 
 
 
 
 
 
 
 
 
146

 
 
 
 
 
 
 
Figure 1. Schematic overview of study variables and hypotheses 
 
 
 
 
 
 
147
Individual  differences,  such  as  learning  strategy,  personality  and 
motivational  characteristics,  socio‐economic  demographics  and 
cognitive  ability  have  found  to  influence  the  degree  of  on‐the‐job 
learning taking place (McCall, 1998; Spreitzer et al., 1997; Wouters and 
Buyens,  2006).  Consequently,  and  quite  evidently,  we  posit  that 
employees  with  higher  learning  agility  will  exhibit  a  higher  degree  of 
on‐the‐job learning.  
 
H2.  The  prediction of  being  identified  as a  high  potential or not  by  learning 
agility is mediated by actual on‐the‐job learning.  
 
Regardless of the increasing importance of learning agility to leadership 
practice, amongst all possible criteria for identifying high potentials, an 
above‐average  job  performance  is  unanimously  considered  the  most 
decisive  factor  (Dries  and  Pepermans,  2007,  2008;  Spreitzer  et  al., 
1997). Furthermore, for most companies, evaluating potential based on 
‘hard’  performance  data  is  less  politically  charged  than  assessing 
candidates against a list  of end‐state and learning competencies (Cope, 
1998;  Pepermans,  Vloeberghs  and  Perkisas,  2003).  Nonetheless,  it  is 
crucial that organizations clearly distinguish high performers from high 
potentials.  Although  an  excellent  performance  is  and  should  be  a 
prerequisite for being identified as a high potential, organizations must 
beware  of  halo  effects  in  their  appraisal  processes,  which  cause 
assessors  to  wrongfully  align  potentiality  ratings  with  performance 
ratings (Fields, 2002). We therefore posit that although job performance 
will  be  a  significant  predictor  of  leadership  potential,  learning  agility 
should add significantly to the prediction.  
 
H3.  Learning  agility  predicts  whether  an  employee  is  identified  as  a  high 
potential or not above and beyond performance rating.  
 
Finally, if learning agility is of such crucial importance to high potential 
identification, an important question is if it is an ‘innate’ quality, or if it 
can be developed. Whereas Lombardo and Eichinger (2000) seemed to 
believe that learning agility is a trait‐like characteristic and thus difficult 
to develop, other authors have reported inconsistent findings (Boyatzis 
and  Saatcioglu,  2008;  Boyatzis,  Leonard,  Rhee  and  Wheeler,  1996; 
Boyatzis,  Stubbs  and  Taylor,  2002).  Karaevli  and  Hall  (2006) 
approached  the  matter  differently.  They  focused  on  the  concept  of 
career  variety,  which  they  subdivided  in  functional  variety  (i.e.  the 
diversity a career actor experiences in terms of functional domains) and 
institutional  variety  (i.e.  the  diversity  a  career  actor  experiences  in 
terms of firm, industry and national work contexts). The authors claim 
that  employees  who  have  spent  most  of  their  careers  in  a  single 
148
industry, for example, have a limited knowledge and skill base, and are 
more likely to engage in a limited search for information, compared to a 
person  with  more  varied  experience.  According  to  this  theory,  then, 
career variety is likely to enhance learning agility. In order to assess the 
developability of the learning agility construct, we will also examine the 
effects  of  age,  educational  level  and  work  experience;  however,  since 
previous  research  has  reported  inconsistently  on  these  relationships, 
we will not posit formal hypotheses for these variables.  
 
H4. Learning agility can increase with career variety.  
 
Methodology 
 
Procedure 
 
Seven organizations known to engage in talent management took part in 
the  study.  They  came  from  four  different  sectors:  financial  consulting 
(41%  of  respondents),  distribution  (35%),  ICT  (14%)  and  telecom 
(10%).  We  performed  telephone  interviews  prior  to  survey 
administration to check whether the organizations’ conceptualization of 
‘high  potential’  corresponded  to  the  definition  used  in  our  theoretical 
framework (i.e. that of Cope, 1998).  
 
The ‘raters’ in the current study were managers, each asked to fill out an 
online survey about one subordinate ‘ratee’ (categorized as either high 
potential  or  non‐high  potential).  Ratees  were  assigned  to  raters  by  the 
HR department to ensure a representative balance (in terms of sex, age, 
educational  level,  work  experience  and  functional  level)  between  high 
potentials  and  non‐high  potentials.  Prior  to  filling  out  the  survey,  they 
were  asked  if  they  felt  in  a  position  to  adequately  assess  the  assigned 
ratee’s job performance and learning agility. If the answer was negative, 
they  were  assigned  another  ratee.  We  opted  for  supervisors  as  raters 
rather than use self‐appraisal in order to avoid social desirability effects.  
 
Upon completion, the surveys were sent back to each organization’s HR 
department,  who  then  indicated  whether  the  ratee  in  question  was 
identified  by  the  organization  as  a  high  potential  or  not,  and  removed 
the  rater  and  ratee  identities  to  ensure  respondent  anonymity;  only 
then  were  the  results  sent  to  the  researchers.  The  managers  who 
participated as raters were only informed of the high potential variable 
in  the  study  design  after  completion  of  their  rating  task,  so  that 
knowledge of the study hypotheses could not confound their responses.   
 
 
149
Sample demographics 
 
Raters.  Raters  were  63  managers,  completing  the  online  survey 
about  63  subordinates  reporting  to  them  at  the  time  of  the  data 
collection.  Among  the  raters,  51  (81%)  were  men  and  12  (19%)  were 
women. They belonged mostly to middle management (n = 36; 57%); 10 
were line managers (16%), 9 were senior managers (14%) and 8 were 
executive managers (13%). Their average age was 43.83 (sd = 7.27). On 
average,  raters’  tenure  within  their  current  organization  was  17.77 
years  (sd  =  15.87),  of  which  they  collaborated  with  their  ratees  for  an 
average of 4.84 years (sd = 4.05).  
 
Ratees. As for the ratees, 32 had been identified as high potentials at 
the time of the data collection and 31 had not been identified as such. Of 
the  high  potentials,  22  (69%)  were  men  and  10  (31%)  were  women. 
Their average age was 36.59 (sd = 6.38). The majority of them obtained 
either  a  university  degree  (n  =  16;  50%)  or  a  college  degree  (n  =  14; 
44%). On average, they had 13.53 years of work experience (sd = 6.94). 
Of  the  high  potential  ratees,  6  were  in  functions  at  a  non‐management 
level  (19%);  7  were  line  managers  (22%);  12  were  middle  managers 
(37%); 6 were senior managers (19%); and 1 was an executive manager 
(3%).  
 
Of  the  non‐high  potentials,  18  (58%)  were  men  and  13  (42%)  were 
women. Their average age was 33.45 (sd = 6.44). The majority of them 
obtained a college degree (n = 19; 61%); 6 obtained a university degree 
(19%). On average, they had 11.48 years of work experience (sd = 7.14). 
The majority of the non‐high potential ratees were in functions at a non‐
management level (n = 23; 74%); 4 were line managers (13%); 3 were 
middle managers (10%); and 1 was a senior manager (3%).  
 
Measures 
 
Learning  agility.  In  order  to  assess  learning  agility,  we  used  the 
Choices  instrument  developed  by  Lombardo  and  Eichinger  (2000), 
which is based on an extensive content analysis of interview and survey 
data collected over several years by the Center for Creative Leadership. 
A  validation  study  indicated  that  the  instrument  contains  four  factors 
(Lombardo  and  Eichinger,  2000):  mental  agility  (i.e.  being  attracted  to 
new ideas and complexity and being a quick thinker), people agility (i.e. 
actively  searching  for  feedback  and  being  open  to  diverse  people  and 
ideas),  change  agility  (i.e.  taking  part  in  change  and  optimization 
processes),  and  results  agility  (i.e.  delivering  results  even  under  first‐
150
time  or  difficult  circumstances).  Example  items  are:  ‘Is  able  to  see 
relationships between issues others do not see as related’ and ‘Is able to 
present  ideas  and  concepts  in  a  way  adapted  to  the  target  audience’. 
Each item was rated on a 5‐point Likert scale ranging from ‘1. Not at all 
applicable  to  this  person’  to  ‘5.  Very  clearly  applicable  to  this  person’. 
Possible bias that might occur by having raters rate items of which they 
cannot  give  a  valid  assessment  (for  instance  because  it  deals  with  a 
specific behavior they have not yet been able to observe) is countered in 
the instrument by the extra response category ‘I don not know/cannot 
give a valid appraisal’.  
 
On­the­job  learning.  This  variable  was  measured  using  two  items, 
developed  by  Spreitzer  et  al.  (1997).  The  authors  identified  two 
dimensions of on‐the‐job learning: job content learning (i.e. ‘Relative to 
other managers you have worked with, how effectively has this person 
learnt new technical, functional, service, or customer information?’) and 
behavioral  skill  learning  (i.e.  ‘Relative  to  other  managers  you  have 
worked  with,  how  effectively  has  this  person  learnt  new  behavioral 
skills—that is, new ways of interacting effectively with people in getting 
the job done?’). Both items are scored on a 5‐point Likert scale ranging 
from ‘1. Not at all effectively’ to ‘5. Very effectively’.  
   
Job  performance  rating.  Job  performance  rating  was  measured  by 
one  single  item,  i.e.  ‘At  the  most  recent  performance  appraisal,  this 
person’s performance was rated as…’. Response categories were: ‘Does 
not  reach  the  expected  level’,  ‘Reaches  the  expected  level  to  some 
extent’, ‘Reaches the expected level’ and ‘Exceeds the expected level’ (as 
in Fields, 2002).  
 
Career  variety.  In  order  to  assess  ratees’  career  variety,  the  two 
items  developed  by  Karaevli  and  Hall  (2006)  were  used.  The  autors 
distinguish  between  institutional  career  variety  (i.e.  ‘For  how  many 
different organizations has this person worked during his or her entire 
career?’)  and  functional  career  variety  (i.e.  ‘In  how  many  different  job 
domains has this person worked during his or her entire career?’). 
     
Results 
 
Table  2  provides  an  overview  of  the  means,  standard  deviations  and 
intercorrelations of the study variables.  
Table 2. Means, standard deviations and intercorrelations of study variables (n = 63) 
 
  m  sd  1  2  3  4  5  6  7  8  9 
1. Learning agility (total)  b  3.75  .65  (.98)                 
2. Mental agility b  3.35  .63  .94**  (.96)               
3. People agility b  3.27  .59  .89**  .85**  (.95)             
4. Change agility b  4.16  .73  .96**  .89**  .79**  (.92)           
5. Results agility b  4.21  .85  .93**  .80**  .72**  .88**  (.92)         
6. Age d  35.0 6.55  .00  ‐.06  ‐.17  .00  .16  ‐       
7. Work experience d  12.5 7.06  ‐.039  ‐.11  ‐.19  ‐.02  .11  .95**  ‐     
8. Educational level c  9.05  1.21  .39**  .41**  .30*  .39**  .36**  ‐.19  ‐.25*  ‐   
9. Institutional career variety b  1.90  .84  .25  .15  .19  .24  .29*  .26*  .33**  ‐.14  ‐ 
10. Functional career variety a  2.76  .84  .29*  .19  .18  .31*  .38**  .35**  .32*  .03  .19 
11. Job content on‐the‐job learning b  3.95  .71  .65**  .73**  .51**  .61**  .58**  ‐.16  ‐.24  .44**  .07 
12. Behavioral skill on‐the‐job learning b  3.63  .68  .54**  .52**  .62**  .50**  .40**  ‐.33**  ‐.35**  .26*  .11 
151

13. Performance rating b  3.76  .73  .60**  .54**  .64**  .55**  .53**  ‐.30*  ‐.32*  .30*  ‐.04 
Notes. Coefficient alphas are on the primary diagonal; * p <.005; ** p <.001. 
a 4‐point scale, b 5‐point scale, c 10‐point scale, d open‐ended item; for each scale, we used the original format as proposed by its authors. Although 

this means that the response format across the questionnaire was inconsistent, several authors have argued that such a design would increase 
reliability, as the odds of common method variance decrease (i.e. the odds of respondents responding similarly to all items in a scale in an 
undifferentiating manner; e.g. Schwarz, Schwarz and Rizzuto, 2008). 
Table 2. Continued 
 
10  11  12  13 
       
       
       
       
       
       
       
       
       
‐       
.14  ‐     
.19  .57**  ‐   
152

.22  .60**  .73**  ‐ 


Notes. Coefficient alphas are on the primary diagonal; * p <.005; ** p <.001. 
 
 
 
Table 3. Indirect effect of learning agility on being identified as a high potential or not, mediated by on‐the‐job learning 
 
  Point Estimate  Product of coefficients  Bootstrapping 
  SE Z Percentile 95%  BC 95% CI  BCa 95% CI 
Lower  Upper  Lower  Upper  Lower  Upper 
Job content learning  .146  .074  1.959  .002  .281  .016  .295  .017  .297 
Behavioral skill learning  ‐.034  .053  ‐.633  ‐.160  .080  ‐.160  .080  ‐.154  .086 
Note. 5000 bootstraps used. 
 
153
In order to test H1 and H2, we ran a mediation analysis in SPSS (17.0) to 
discern direct from indirect effects of learning agility on being identified 
as  a  high  potential  or  not.  Job  content  learning  and  behavioral  skill 
learning, the two dimensions of on‐the‐job learning from the Spreizer et 
al.  measure  (1997),  were  included  as  possible  mediators.  Table  3 
displays  estimates  and  95%  bias  corrected  (BC)  and  accelerated  (BCa) 
confidence  intervals  for  the  indirect  effects  of  learning  agility  on  being 
identified  as  a  high  potential  or  not.  When  zero  is  contained  in  the  BC 
and  BCa  confidence  interval  of  a  mediator,  the  indirect  effect  through 
that  particular  mediator  is  considered  not  significantly  different  from 
zero. Results indicated that job content learning, but not behavioral skill 
learning,  significantly  mediated  the  effect  of  learning  agility  on  being 
identified  as  a  high  potential  or  not.  However,  this  indirect  effect  was 
smaller  when  compared  to  the  significant  direct  effect  eminent  in  our 
results (β = .33, t = 3.03, p < .01). The total effect of learning agility also 
proved  significant  (β  =  .44,  t  =  5.47,  p  <  .001).  The  partial  mediation 
model  explained  37.2%  of  total  variance  in  being  identified  as  a  high 
potential or not (F(3,62) = 11.63, p < .001); the adjusted R square value 
corrected  the  amount  of  variance  explained  to  34%.  As  a  direct 
significant effect was found of learning agility onto being identified as a 
high potential or not, we accept H1; H2 can only be accepted in part as 
the prediction model is only partially mediated by job content learning, 
but not by behavioral skill learning.  
 
In order to assess the relative impact of job performance rating and that 
of learning agility in predicting being identified as a high potential or not 
(H3),  we  performed  hierarchical  logistic  regression  analysis. 
Performance rating was added to the prediction model in step one and 
learning  agility  in  step  two.  The  step  one  model  proved  statistically 
significant (χ²(1, n = 63) = 9.60, p < .01), indicating that the model was 
able  to  distinguish  ratees  identified  as  high  potentials  from  employees 
not identified as such based on their job performance ratings. The model 
explained  between  14.1%  (Cox  and  Snell  R  Square)  and  18.9% 
(Nagelkerke  R  Square)  of  variance,  and  correctly  classified  66.7%  of 
ratees. Upon adding learning agility in step two, the model still proved 
significant  (χ²(2,  n  =  63)  =  26.63,  p  <  .001).  Moreover,  this  addition 
significantly improved the model (Δχ²(1, n = 63) = 17.03, p < .001), and 
raised  the  amount  of  explained  variance  to  34.5%  (Cox  and  Snell  R 
Square)  and  46%  (Nagelkerke  R  Square).  The  full  model  correctly 
classified  77.8%  of  ratees.  As  is  shown  in  Table  4,  performance  rating 
made  a  unique  statistically  significant  contribution  to  the  step  one 
model.  However,  its  contribution  became  insignificant  once  learning 
agility was added to the model in step two. Learning agility thus became 
154
the  only  significant  contributor,  recording  an  odds  ratio  of  18.17, 
indicating  that  ratees  scoring  high  on  learning  agility  were  over  18 
times  more  likely  to  be  classified  as  a  high  potential.  These  findings 
cause us to accept H3. 
 
Table 4. Hierarchical logistic regression predicting the likelihood of being identified as a 
high potential or not 
 
  B  S.E.  Wald  p odds ratio 
Step 1. 
Performance rating 1.21 .44 7.63 .006 3.35
 
Step 2.  
Performance rating .21 .54 .15 .699 1.23
Learning agility  2.90 .89 10.70 .001 18.17
 
Multiple  regression  analysis  was  conducted  to  assess  the  degree  to 
which  career  variety  (subdivided  into  two  variables,  institutional  and 
functional  variety,  as  in  Karaevli  and  Hall,  2006)  can  predict  learning 
agility  (H4).  Although  we  did  not  formulate  explicit  hypotheses  about 
age, educational level and work experience, due to inconsistent findings 
in  the  literature,  we  also  added  these  variables  in  our  regression. 
Collinearity statistics indicated that the multiple correlation of age with 
work  experience  (r  =  .95,  p  <  .001)  was  unacceptably  high  (Tolerance: 
.09; Variance inflation factor: 11.40). We opted to eliminate age from the 
equation as the work experience variable relates more explicitly to work 
domain.  The  four  predictors  together  explained  31.2%  of  the  total 
variance  in  learning  agility  (F(4,62)  =  6.58,  p  <  .001).  However,  as  our 
sample size was relatively small (n = 63), it is better to report adjusted R 
square,  which  was  26.5%.  Three  of  the  four  predictors  proved 
significant:  a  higher  educational  level  (β  =  .40,  p  <  .001),  and  more 
institutional  (β  =  .28,  p  <  .05)  and  functional  variety  (β  =  .27,  p  <  .05) 
significantly  increased  learning  agility  scores,  while  work  experience 
did  not  make  a  significant  contribution  (β  =  ‐.12,  p  =  .339).  We  thus 
accept  H4  in  that  learning  agility  does  seem  to  increase  with  career 
variety.  We  found  the  same  for  educational  level;  for  the  work 
experience  variable,  however,  no  significant  relationship  with  learning 
agility was found.  
 
Discussion 
 
In  the  current  paper  we  aimed  to  test  the  assumption  from  the  recent 
talent  management  literature  that  learning  agility  is,  or  should  be,  the 
primary  indicator  of  leadership  potential  (Dries  and  Pepermans,  2007; 
Lombardo and Eichinger, 2000; Peters and Smith, 1996; Spreitzer et al., 
155
1997).  Furthermore,  we  examined  whether  learning  agility,  if  it  is 
indeed such an important ‘metacompetency’ for future leaders, could be 
enhanced with greater career variety.  
 
Our  results  indicated  that,  indeed,  supervisor‐rated  scores  of  learning 
agility  were  predictive  of  ratees  being  identified  as  high  potentials  or 
not.  Learning  agility  even  predicted  leadership  potential  above  and 
beyond  job  performance,  which  is  considered  the  irrefutable  core 
characteristic of all high potentials (Dries and Pepermans, 2008; McCall, 
1998).  Employees  identified  as  high  potentials  by  their  organizations 
were  believed  to  demonstrate  higher  overall  learning  agility  than  non‐
high potentials. Furthermore, this effect was only partially mediated by 
on‐the‐job  learning,  indicating  that  the  effect  could  not  be  fully 
explained  by  actual,  observed,  learning  behavior  on  the  job.  In  this 
respect,  it  appeared  that  demonstrating  quick  learning  in  terms  of 
technical aspects of a job (i.e. job content on‐the‐job learning) was much 
more important in order to receive the high potential label than learning 
to  interact  better  with  others  (i.e.  behavioral  skill  on‐the‐job  learning), 
as the latter type of on‐the‐job learning was not found to be a mediator 
in  our  model.  Despite  the  evidence  of  partial  mediation  by  job  content 
on‐the‐job  learning,  the  direct  effect  of  learning  agility  on  being 
identified  as  a  high  potential  or  not  was  larger,  indicating  that 
supervisors  did  not  base  their  ratings  of  learning  agility  on  actual  on‐
the‐job learning alone, and that learning agility and on‐the‐job learning 
behavior were thus perceived as two related but distinct variables in the 
rating process. As mentioned in the Introduction, we do not believe that 
these  effects  are  attributable  to  organizations  explicitly  assess  learning 
agility  when  identifying  high  potentials.  In  the  current  study,  this 
possibility  was  checked  during  the  telephone  interviews  with  the 
participating  organizations,  who  all  indicated  that  learning  agility  was 
not  explicitly  incorporated,  in  this  form,  in  their  competency  profiles 
used  to  identify  high  potentials.  Thus,  it  appears  that  high  learning 
agility  is  a  (‘covert’)  common  factor  among  those  identified  as  high 
potentials  (Dries  and  Pepermans,  2007).  This  is  an  interesting  finding, 
seeing  as  organizations  today  are  going  to  great  lengths  to  ensure  that 
their competency frameworks are up to date – whereas learning agility 
is potentially an ‘overarching’  metacompetency, of which the relevance 
for high potential identification is expected to be constant and ongoing 
(Briscoe and Hall, 1999).   
 
Apart  from  linking  the  learning  agility  variable  to  leadership  potential, 
we  also  examined  whether  it  can  be  enhanced  with  experience.  Our 
findings appear somewhat inconsistent. As hypothesized, we found that 
156
learning  agility  is  positively  related  to  career  variety.  According  to 
Karaevli  and  Hall  (2006),  such  a  finding  would  indicate  that,  indeed, 
learning  agility  can  be  enhanced  with  experience.  However,  when  we 
put this finding alongside some of our other findings – i.e. that learning 
agility is positively related to educational level, but not to age and work 
experience – a more plausible explanation is one assuming that causality 
is the other way around. This would mean that learning agility is a more 
or  less  stable  trait  (as  was  also  the  conviction  of  Lombardo  and 
Eichinger,  2000),  that  does  not  significantly  increase  with  age  or  work 
experience. In addition, people with high learning agility would tend to 
seek  out  more  learning  opportunities  by  pursuing  a  high  educational 
level  and  a  variety  of  career  experiences.  However,  since  the  current 
study was not set up as longitudinal, the above inferences are tentative.   
 
Limitations 
 
We  identified  four  issues  that  warrant  reflection,  and  should  be  taken 
into  account  by  readers  in  interpreting  and  applying  the  results 
reported  in  this  article.  First  of  all,  the  respondent  sample  upon  which 
the  analyses  were  performed  was  rather  small  (n  =  63),  which  implies 
that  generalization  of  the  findings  should  be  cautious.  Nonetheless,  as 
there  is  so  little  empirical  research  about  talent  management  and  high 
potentials  –  partly  due  to  the  discretion  of  organizations  about  their 
procedures – we do feel that the current study represents a contribution 
for  personnel  researchers  and  practitioners  alike.  Looking  at  the  HR 
practitioner literature, it is clear that there is a need for more research 
on talent management. Organizations and consultants worldwide report 
that there is a need for more empirical evidence (and guidelines) about 
talent  management  and  high  potentials  –  so  far  there  is  mostly 
conjecture (Buckingham and Vosburgh, 2001).  
 
Second,  we  must  consider  the  possibility  that  halo  bias  crept  into  our 
findings. Since the supervisors acting as raters in the study were asked 
to  assess  the  performance,  learning  agility  and  degree  of  on‐the‐job 
learning of their ratees simultaneously, it is feasible that scores on one 
variable  influenced  scores  on  the  other  variables  (be  it  in  an 
unconscious way). However, banning halo bias (and relatedly, common 
method variance) entirely from assessment settings poses a challenge in 
nearly  all  forms  of  empirical  research  (Schwarz,  Rizzuto  and  Schwarz, 
2008).  Furthermore,  halo  bias  is  also  quite  present  in  the  practice  of 
performance  and  potentiality  assessment,  so  that  interrelations 
between the study variables can be both attributable to our study design 
and  to  the  reality  of  high  potential  identification.  Furthermore,  having 
157
supervisors  rate  their  subordinates  without  knowledge  of  the  high 
potential  variable  in  the  study  design  at  the  time  of  survey 
administration  reduced  the  odds  of  the  high  potential  ratees  receiving 
systematically higher scores on the independent variables.  
 
A third limitation of the current paper is that the categorization of ratees 
into  high  potentials  and  non‐high  potentials  was  decided  upon  by  the 
participating organizations, without there being an external criterion by 
which  to  evaluate  the  accuracy  of  this  dichotomization.  However,  the 
concept  of  ‘high  potential’  is  by  definition  organization‐specific  (see 
Cope, 1998). A person considered a high potential in one organization, is 
not necessarily identified as a high potential in another (Boudreau and 
Ramstad,  2005).  Furthermore,  as  there  is  no  such  thing  as  a  universal 
list of criteria an employee must meet in order to be considered a high 
potential, it would be very difficult (if not impossible) for researchers to 
categorize  respondents  in  a  high  potential  versus  a  non‐high  potential 
group based on external criteria. To counter this limitation, as reported 
in the Methodology section, we performed telephone interviews prior to 
survey  administration  to  reduce  potential  heterogeneity  across 
organizations in terms of their definition of what a high potential is.  
 
Finally,  one  could  identify  a  possible  paradox  in  this  paper.  While  we 
argue  for  a  developmental  approach  to  high  potential  identification  in 
the  Introduction,  our  results  imply  that  learning  agility  might  not 
increase  with  age  or  work  experience  (which,  in  turn,  implies  that 
learning agility is something that could be selected for). We propose that 
the  selection  and  the  developmental  perspective  need  not  be  mutually 
exclusive:  why  not  incorporate  learning  agility  as  a  high  potential 
identification  criterion  amongst  other  criteria  deemed  relevant  by 
organizations?  Learning  agility  is  to  be  considered  a  function  of  both 
‘wanting’ and ‘being able’ to learn – and where ‘being able’ may be more 
of a dispositional characteristic, the ‘wanting’ aspect might be enhanced 
by  (individually  or  organizationally  initiated)  interventions  aimed  at 
increasing the motivation to learn. However, more research needs to be 
done  (preferably  longitudinal)  in  order  to  examine  to  what  extent 
learning agility can be developed in diverse samples of employees.  
 
Implications 
 
As  regards  future  research  into  learning  agility,  longitudinal  designs 
especially  offer  an  interesting  way  forward.  By  studying  the  learning 
experiences  of  employees  throughout  time,  causal  inferences  could  be 
made  about  the  stability  of  learning  agility,  and  about  the  processes 
158
linking learning agility to the acquisition of new skills and competencies. 
Examining  populations  who  are  exposed  to  many  learning  experiences 
in  relatively  short  periods  of  time  (for  instance,  management  trainees) 
could  be  a  possible  suggestion.  Furthermore,  longitudinal  research 
might  also  prove  useful  in  order  to  expose  potential  conflict  between 
short  and  long‐term  concerns  in  management  development.  It  appears 
that  there  is  an  inherent  conflict  between  the  need  for  performance  in 
the  short  run  (i.e.  having  the  end‐state  competencies  required  for 
success in today’s business environment) and learning the competencies 
necessary  for  success  in  the  long  run.  Prior  research  has  shown  that 
performance  concerns,  particularly  early  in  skill  acquisition,  can 
override  learning  objectives  (e.g.  Kanfer  and  Ackerman,  1989).  Risk 
taking may impede performance in the short run but facilitate learning 
in the long run (Spreitzer et al., 1997). Finally, it would be interesting to 
research to what extent organizations take learning agility into account 
(explicitly or implicitly) in their high potential identification procedures. 
Although  organizations  do  acknowledge  that  the  ability  to  learn  from 
experience  is  important  (e.g.  Bossuyt  and  Dries,  2008),  the  question 
remains whether and how they take this factor into account in practice. 
A popular theme in the literature on high potential identification is the 
tension between assessment of end‐state competencies, learning agility 
(i.e.  metacompetencies),  situational  variables  and  individual  career 
orientation  (Hollenbeck,  McCall  and  Silzer,  2006).  Studies  separating 
‘generic’  or  ‘universal’  criteria  (if  these  in  fact  exist)  from  aspects  that 
should  be  tailored  to  fit  specific  organizational,  cultural  and  temporal 
contexts,  would  greatly  improve  knowledge  about  the  topic,  and 
contribute  to  assessments  of  leadership  potential  taking  place  in 
practice.  
 
With  respect  to  practical  implications,  one  suggestion  could  be  for 
organizations  to  incorporate  some  form  of  learning  agility  assessment 
into  their  selection  and  development  procedures.  First  of  all, 
assessments of learning agility might prove useful to predict the future 
performance  and  potential  of  employees.  The  literature  indicates  that 
the  predictive  value  of  ‘regular’,  end‐state  competencies  depends 
strongly on the degree in which they will still be relevant several years 
after  assessment,  as  well  as  on  the  degree  to  which  they  can    be 
developed  (and  over  what  period  of  time)  (Briscoe  and  Hall,  1999). 
Secondly,  applying  learning  agility  in  different  HRM  processes  may 
prove  useful  for  assessors  to  more  adequately  separate  performance 
from  potentiality  assessments  (see  the  discussion  of  halo  bias  in  the 
Introduction). One of the main causes for halo  bias to occur is that the 
same  end‐state competencies are often used for both performance and 
159
potentiality  assessments  (which  often  also  take  place  roughly  at  the 
same  time  of  year  and  by  the  same  assessors).  By  using  end‐state 
competencies  as  the  basis  for  performance  assessments  and  learning 
agility as the basis for potentiality assessments, it is conceptually more 
clear for assessors that these are truly two separate scales on which to 
rate  employees.  Thirdly,  by  assessing  learning  agility  in  junior 
employees,  organizations  will  be  encouraged  to  optimize  their  training 
and  development  programs.  A  large  volume  of  research  indicates  that 
training  initiatives  (e.g.  job  rotation,  mentoring,  coursework)  rarely 
generate  the  desired  results  (e.g.  Karaevli  and  Hall,  2006).  Often,  the 
quality  of  the  training  initiatives  in  itself  is  put  into  question  as  a 
response  to  its  failure  –  where  another  possible  explanation  could  be 
that  the  initiatives  were  not  matched  adequately  with  the  learning 
agility  of  the  participants  (i.e.  not  all  participants  possess  an  equal 
amount  of  ‘wanting’  and  ‘being  able’  to  learn).  Furthermore, 
organizations  need  to  be  aware  of  the  fact  that  they  themselves  can 
create  barriers  to  learning,  even  in  their  high  potential  programs.    The 
following  issues  have  been  mentioned  in  the  literature  as  barriers 
hindering  the  transfer  from  learning  agility  to  on‐the‐job  learning 
behavior:  being  forced  into  a  development  track  that  moves  either  too 
fast  or  too  slow;  a  lack  of  developmental  opportunities,  structures  or 
support in the organization; and constraints in terms of time, space and 
budget (e.g. Karaevli and Hall, 2006). 
 
We  wish  to  conclude  this  paper  with  a  call  for  further  research  and 
theory  development  about  talent  management  in  general,  and  high 
potential programs specifically. From the current paper, as well as from 
some  previous  publications  (e.g.  Briscoe  &  Hall,  1999;  Dries  & 
Pepermans,  2007;  Lombardo  &  Eichinger,  2000),  we  can  conclude  that 
assessments of learning agility may prove as a useful tool for identifying 
future  leaders  in  organizations,  as  certain  issues  with  using  end‐state 
competencies  or  performance  data  are  bypassed.  However,  since  it  is 
not  yet  clear  to  which  extent  learning  agility  (or,  at  least,  the 
operationalization  of  learning  agility  applied  in  the  current  paper)  is  a 
trait‐like  characteristic,  as  opposed  to  a  competency  that  can  be 
developed, we must take care not to return to the so‐called ‘right stuff’ 
thinking  so  prominent  in  the  selection  perspective  on  high  potential 
identification  (McCall,  1998).  The  current  war  for  talent  demands  a 
developmental approach, as ‘talent’ in its traditional, elitist construction 
is  becoming  increasingly  scarce.  There  is  an  urgent  need  for  a 
broadening  of  the  concepts  of  talent  and  talent  management  (Bossuyt 
and Dries, 2008; Dyer and Humphries, 2002; Iles, 1997; Sullivan, 1999). 
We  believe  that  by  incorporating  the  concept  of  learning  agility  (i.e. 
160
wanting  to  learn  and  being  able  to  learn)  more  centrally  into  talent 
management  discourse,  a  less  elitist,  more  inclusive  view  on  talent 
management will be achievable. 
 
References 
 
Bar‐On, R. (1997), The Bar­On Emotional Quotient inventory (EQ­i): a test  of emotional intelligence, 
Multi‐Health Systems, Toronto, Canada.   
Bossuyt,  T.  and  Dries,  N.  (2008),  “Talent  management  en  flexibele  loopbaanpaden  voor  de 
werknemers  van  morgen  [Talent  management  and  flexible  career  tracks  for  tomorrow's 
workforce]”, in Calmeyn, H., De Witte, K. and Weverbergh, J. (Eds.), Licht op leren 2008: Leren 
en  ontwikkelen  in  een  talentgerichte  maatschappij  [Spotlight  on  learning  2008:  Learning  and 
developing in a talent­oriented society], LannooCampus, Leuven, Belgium, pp. 55‐90.  
Boudreau,  J.W.  and  Ramstad,  P.  (2005),  “Talentship  and  the  evolution  of  human  resource 
management:  from  professional  practices  to  strategic  talent  decision  science”,  Human 
Resource Planning Journal, Vol. 28 No. 2, pp. 17‐26.  
Boyatzis, R.E., Leonard, D., Rhee, K. and Wheeler, J. (1996), “Competencies can be developed but not 
in the way we thought”, Capability, Vol. 2 No. 2, pp. 25‐41. 
Boyatzis, R.E. and Saatcioglu, A. (2008), “A twenty year view of trying to develop emotional, social 
and  cognitive  intelligence  competencies  in  graduate  management  education”,  Journal  of 
Management Development, Vol. 27 No. 3, pp. 92‐108.  
Boyatzis,  R.E.,  Stubbs,  E.C.  and  Taylor,  S.N.  (2002),  “Learning  cognitive  and  emotional  intelligence 
competencies through graduate management education”, Academy of Management Journal on 
Learning and Education, Vol. 1 No. 2, pp. 150‐62. 
Briscoe, J.P. and Hall, D.T. (1999), “Grooming and picking leaders using competency frameworks: do 
they  work?  An  alternative  approach  and  new  guidelines  for  practice”,  Organizational 
Dynamics, Vol. 28 No. 4, pp. 37‐52.  
Buckingham,  V.  and  Vosburgh,  R.  (2001),  "The  21st  century  human  resources  function:  it's  the 
talent, stupid!", Human Resource Planning, Vol. 24 No. 4, pp. 17‐23. 
Cope, F. (1998), “Current issues in selecting high potentials”, Human Resource Planning, Vol. 21 No. 
3, pp. 15‐7. 
Dries,  N.  and  Pepermans,  R.  (2007),  “Using  emotional  intelligence  to  identify  high  potentials:  a 
metacompetency perspective”, Leadership and Organization Development Journal, Vol. 28 No. 
8, pp. 749‐70. 
Dries,  N.  and  Pepermans,  R.  (2008),  "Real  high  potential  careers:  An  empirical  study  into  the 
perspectives  of  organizations  and  high  potentials”,  Personnel  Review,  Vol.  37  No.  1,  pp.  85‐
108. 
Dyer, S., and Humphries, M. (2002), “Normalising workplace change through contemporary career 
discourse”,  in  Morrow,  L.,  Verins,  I.  and  Willis,  E.  (Eds.),  Mental  health  and  work:  issues  and 
perspectives, Auseinet, Bedford Park, Australia, pp. 48‐62.  
Fields,  G.S.  (2002),  “Predicting  potential  for  promotion:  how  the  data  in  human  resource 
information systems can be used to help organizations gain competitive advantage”, Working 
paper No. 02­14, Cornell University, Ithaca, NY. 
Frank, F.D. and Taylor, C.R. (2004), “Talent management: trends that will shape the future”, Human 
Resource Planning, Vol. 27 No. 1, pp. 33‐41.  
Fugate,  M.,  Kinicki,  A.J.,  and  Ashforth,  B.E.  (2004),  “Employability:  a  psycho‐social  construct,  its 
dimensions, and applications”, Journal of Vocational Behavior, Vol. 65 No. 1, pp. 14‐38. 
Hall, D.T. (2002), Careers in and out of organizations, Sage Publications, Thousand Oaks, CA. 
Hollenbeck,  G.P.,  McCall,  M.W.  and  Silzer,  R.F.  (2006),  “Theoretical  and  practitioner  letters: 
leadership competency models”, The Leadership Quarterly, Vol. 17 No. 4, pp. 398‐413.  
Iles,  P.  (1997),  “Sustainable  high  potential  career  development:  a  resource‐based  view”,  Career 
Development International, Vol. 2 No. 7, pp. 347‐53.  
Jerusalim,  R.S.  and  Hausdorf,  P.A.    (2007),  “Managers'  justice  perceptions  of  high  potential 
identification practices”, Journal of Management Development, Vol. 26 No. 10, pp. 933‐50. 
Kaiser,  R.B.,  Lindberg,  J.  T.,  and  Craig,  S.B.  (2007),  “Assessing  the  flexibility  of  managers:  a 
comparison of methods”, International Journal of Selection and Assessment, Vol. 15 No. 1, pp. 
40‐55. 
161
Kanfer, R., and Ackerman, P.L. (1989), “Motivation and cognitive abilities: an integrative/aptitude‐
treatment  interaction  approach  to  skill  acquisition”,  Journal  of  Applied  Psychology­
Monograph, Vol. 74, pp. 657‐90.  
Karaevli,  A.  and  Hall,  D.T.  (2006),  “How  career  variety  promotes  the  adaptability  of  managers:  a 
theoretical model”, Journal of Vocational Behavior, Vol. 69 No. 3, pp. 359‐73. 
Kolb,  D.A.  (1984),  Experiential  learning:  experience  as  the  source  of  learning  and  development, 
Prentice Hall, New Jersey, NY.  
Lewis,  R.E.  and  Heckman,  R.J.  (2006),  “Talent  management:  a  critical  review”,  Human  Resources 
Management Review, Vol. 16 No. 2, pp. 139‐54. 
Lombardo,  M.M.  and  Eichinger,  R.W.  (2000),  “High  potentials  as  high  learners”,  Human  Resource 
Management, Vol. 39 No. 4, pp. 321‐30.  
London,  M.  and  Smither,  J.W.  (1999),  "Empowered  self‐development  and  continuous  learning", 
Human Resource Management, Vol. 38 No. 1, pp. 3‐15. 
McCall, M.W. (1998), High flyers: developing the next generation of leaders, Harvard Business School 
Press, Boston, MA.  
Murphy,  P.  and  Jackson,  S.  (1999),  "Managing  work  role  performance:  challenges  for  twenty‐first 
century  organizations  and  their  employees",  in  Ilgen,  D.,  Pulakos,  E.  (Eds.),  The  changing 
nature of performance: implications for staffing motivation and development, Jossey‐Bass, San 
Francisco, CA, pp. 325‐65.  
Pepermans,  R.,  Vloeberghs,  D.  and  Perkisas,  B.  (2003),  “High  potential  identification  policies:  an 
empirical study among Belgian  companies”, Journal of Management Development, Vol. 22 No. 
8, pp. 660‐78. 
Peters,  J.  and  Smith,  P.A.C.  (1996),  “Action  learning  and  the  leadership  development  challenge”, 
Journal of Workplace Learning, Vol. 10 No. 6, pp. 284‐91.  
Porter,  G.  and  Tansky,  J.    (1999),  “Expatriate  success  may  depend  on  a  learning  orientation: 
considerations  for  selection  and  training”,  Human  Resource  Management,  Vol.  38  No.  1,  pp. 
47‐60.  
Pulakos, E.D., Arad, S., Donovan, M.D. and Plamondon, K.E. (2000), “Adaptability in the workplace: 
development of a taxonomy of adaptive performance”, Journal of Applied Psychology, Vol. 85, 
No. 4, pp. 612‐24.  
Pulakos, E.D., Schmitt, N., Dorsey, D.W., Arad, S., Hedge, J.W. and Borman, W.C. (2002), "Predicting 
adaptive performance: further tests  of  a model of adaptability", Human Performance, Vol. 15 
No. 4, pp. 299 ‐ 323.  
Schwarz,  A.,  Rizzuto,  T.,  and  Schwarz,  C.  (2008).  "Urban  legend  of  common  method  bias:  nine 
common errors and their impacts”, paper presented at the Hawaii International Conference 
on System Sciences, 7‐10 January, Manoa, HI. 
Sloan, E.B. (1994), "Assessing and developing versatility: executive  survival  skill for the brave new 
world", Consulting Psychology Journal, Vol. 46 No. 1, pp. 24 ‐ 31.  
Spreitzer,  G.M.,  McCall,  M.W.  and  Mahoney,  J.D.  (1997),  “Early  identification  of  international 
executive potential”, Journal of Applied Psychology, Vol. 82 No. 1, pp. 6‐29. 
Sullivan, S.R., Carden, W.A. and Martin, D.F. (1998), “Careers in the next millennium: directions for 
future research”, Human Resource Management Review, Vol. 8 No. 2, pp. 165‐85. 
Taylor, T.R. (1997), Administrator’s manual for APIL Battery, Jetline, Parktown, South Africa.   
Wagner, R.K. and Sternberg, R.J. (1991), Tacit knowledge inventory for managers, The Psychological 
Corporation, San Antonio, TX.  
Wechsler, D. (1944), The measurement of adult intelligence, Williams and Wilkins, Baltimore, MD. 
Wouters,  K.  and  Buyens,  D.  (2006),  “Managerial  learning  from  on‐the‐job  experiences:  an 
integrative  framework  to  guide  future  research”,  paper  presented  at  the  Academy  of 
Management Annual Meeting, 11‐16 August, Atlanta, GA. 
  
162
163
 
 
 
 
Chapter III. 
Career outcomes in high potential careers 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Paper 5. 
Dries, N. , Pepermans, R. & Van Roy, E. (in review). The role of employability and firm‐
specific  capital  in  shaping  the  psychological  contract:  Do  high  potentials  and  long‐
tenured employees get the better deal? 
 
Paper 6. 
Dries, N., Van Acker, F. & Pepermans, R. (in review). Effects of the high potential label on 
performance, career success and commitment: A matter of communication? 
164
 
165
Paper 5. 
 
The role of employability and firm­specific 
capital in shaping the psychological contract: 
Do high potentials and long­tenured 
employees get the better deal? 
 
 
This  study  empirically  tested  the  assumption  that  individual  employability  is  directly 
related  to  perceived  psychological  contract  obligations.  Talent  category  (i.e.  being 
identified as a high potential or not) and organizational tenure were incorporated into 
the  prediction  model  as  moderators.  They  were  used  as  indicators  of  firm‐specific 
capital, which is believed to enhance employees’ bargaining position with respect to the 
terms of their psychological contracts. An online survey (n = 103) was set up. We found 
no  consistent  support  for  the  thesis  that  more  employable  employees  would  perceive 
weaker  psychological  contract  obligations  in  terms  of  job  security  and  organizational 
loyalty. Hence, the idea that the downfall of the organizational career and decreases in 
job  security  could  somehow  be  counterbalanced  by  developing  individuals’ 
employability  seems  overly  simplistic.  Furthermore,  high  potentials  and  long‐tenured 
employees perceived stronger obligations from the side of the employer to provide job 
security, than from their side to stay with their employer, particularly under conditions 
of  high  employability.  This  warrants  attention  towards  the  central  role  of  employees’ 
bargaining position in shaping the psychological contract. 
 
 
 
 

 
166

Figure VI. Topicality of Paper 5.
167
Paper 5. 
 
The role of employability and firm­specific capital in shaping the 
psychological contract: Do high potentials and long­tenured 
employees get the better deal? 
 
The recent literature on careers, calling attention to dramatic changes in 
the  nature  of  employment  relationships  due  to  organizational 
restructuring,  downsizing,  outsourcing,  and  delayering,  reveals  a 
renewed  interest  in  the  concept  of  the  psychological  contract  (Coyle‐
Shapiro & Neuman, 2004; Hess & Jepsen, 2009). The general consensus 
seems  to  be  that  psychological  contracts  have  shifted  from  providing 
employment  security  within  a  single  organization  to  employability 
security across relevant labor markets (Clarke & Patrickson, 2008; Van 
Der Heijde & Van Der Heijden, 2006).  
 
Although  some  authors  advocate  the  credo  that  the  desire  for  job 
security is going out of fashion (Tulgan, 2001), several empirical studies 
have  found  that  claims  about  the  current‐day  prevalence  of 
‘boundaryless  careers’  (i.e.  careers  that  defy  traditional  employment 
assumptions;  Arthur  &  Rousseau,  1996)  are  strongly  in  need  of  more 
nuance  (e.g.  Baruch,  2006;  Briscoe  &  Finkelstein,  2009;  Granrose  & 
Baccili,  2006;  Guest  &  MacKenzie  Davey,  1996;  Sullivan,  1999; 
Verbruggen,  Sels  &  Forrier,  2007).  Furthermore,  a  number  of  recent 
thought pieces have  asserted that postmodern career discourse is only 
beneficial for employee groups in strong bargaining positions, i.e. highly 
skilled,  well‐educated  professionals  (Guest,  2001).  In  addition,  they 
claim  that  it  is  potentially  harmful  for  employees  who  do  not  exhibit 
unique  or  particularly  valuable  skills  (e.g.  Dyer  &  Humphries,  2002; 
Lucas,  Liu  &  Buzzanell,  2006;  Van  Buren,  2003).  Van  Buren  (2003) 
states  that  the  downfall  of  job  security  in  general  should  not  be  an 
excuse  for  organizations  to  provide  less  career  support  to  employees. 
He  argues  that  organizations  today,  in  order  to  counterbalance  the 
negative  consequences  of  the  decrease  in  job  security,  need  to  take  up 
their  responsibilities  in  terms  of  providing  more  transparent  and 
realistic  information  about  career  prospects,  as  well  as  greater 
opportunities for enhancing individual employability.  
 
In  this  paper  we  will  empirically  test  the  assumption  that  individual 
employability is directly related to perceptions about the psychological 
contract.  In  addition,  we  will  study  possible  moderation  effects.  We 
expect  that  the  firm‐specific  capital  of  employees  (in  our  study, 
organizational tenure and being identified as a high potential) will act as 
a moderator the model, in that more  firm‐specific capital enhances the 
168
bargaining  position  of  employees,  which  in  turn  shapes  their 
psychological contracts. 
 
Theoretical model 
 
Figure 1 presents an overview of the hypothesized relationships in our 
theoretical model.  
 
Employability as ‘safety net’ in a boundaryless career climate 
 
The recent literature on the psychological contract has, for a large part, 
focused on changes in the psychological contract relating to the ‘death’ 
of the traditional‐organizational career (Hess & Jepsen, 2009; Rousseau, 
2000). However, due to the fact that there are many different definitions 
and operationalizations of the psychological contract, the accumulation 
of  research  findings  into  theory  is  proving  a  difficult  process  (Arnold, 
1996;  Guest,  1998).  In  the  current  study  we  use  Denise  Rousseau’s 
original definition of the psychological contract (1989): “an individual’s 
beliefs  regarding  the  terms  and  conditions  of  a  reciprocal  exchange 
agreement between the focal person and another party.  Key issues here 
include  the  belief  that  a  promise  has  been  made  and  a  consideration 
offered in exchange for it, binding the parties to some set of reciprocal 
obligations” (p. 123).  
 
Initially,  Rousseau  (1989)  identified  two  core  elements  within  the 
psychological  contract,  i.e.  ‘transactional’  obligations  and  ‘relational’ 
obligations.  In  a  transactional  exchange,  employers  offer  specific 
economic rewards in exchange for employees performing limited tasks 
within  a  specific  timeframe.  In  a  relational  exchange,  the  employment 
relationship is open‐ended and trust, security and loyalty are the focus 
of  the  exchange.  Although  the  distinction  between  relational  and 
transactional obligations suggests that employees experience one or the 
other,  most  researchers  agree  that  employees’  psychological  contracts 
contain  both  elements  concurrently  (Arnold,  1996;  Conway  &  Briner, 
2005).  More  recently,  Rousseau  (2000)  added  a  ‘balanced’  element  to 
her  model  of  the  psychological  contract,  in  an  attempt  to  grasp  the 
drastic  changes  reported  in  the  careers  literature.  The  weight  of  the 
relational  element  within  the  psychological  contract  is  claimed  to  have 
reduced  over  time  (Hess  &  Jepsen,  2009).  A  balanced  exchange,  then, 
involves dynamic and open‐ended employment arrangements.  
 
 
 
 
 
 
 
 
 
 
 
 
 
169

 
 
 
 
 
Figure 1. Theoretical model 
 
 
170
In  a  balanced  exchange,  employees  are  required  to  continuously 
contribute to the employer’s competitive advantage, in return receiving 
career  development  advantages  relevant  both  inside  the  organization 
and out (Rousseau, 2000).   
 
The  balanced  element  of  the  psychological  contract  fits  well  with  the 
notion  of  employability  (e.g.  Fugate,  2006;  Van  Der  Heijde  &  Van  Der 
Heijden, 2006). Fugate, Kinicki & Ashfort (2004) define employability as 
“a  host  of  person‐centered  constructs  that  combine  synergistically  to 
help  workers  effectively  adapt  to  the  myriad  of  work‐related  changes 
occurring  in  today’s  economy”  (p.  15).  Proponents  of  employability 
argue  that  employability  security  (i.e.  the  provision  of  development 
opportunities  that  encourage  the  development  of  human  capital  and 
enhance future job prospects) is slowly replacing employment security 
in the new boundaryless career climate (Clarke & Patrickson, 2008; Van 
Buren,  2003).  Accordingly,  the  literature  suggests  the  possibility  (or 
even probability) that more employable employees will be less focused 
on  having  job  security  and  a  long‐term  future  with  their  current 
employer  (Briscoe  &  Finkelstein,  2009).  Consequently,  we  hypothesize 
that  higher  employability  leads  to  weaker  perceived  relational 
obligations and stronger balanced obligations. As regards transactional 
obligations,  we  also  expect  a  negative  relationship  as  the  transactional 
element of the psychological contract describes ‘merely doing one’s job’, 
which  does  not  at  all  correspond  with  the  concept  of  employability 
(Fugate, 2006; Van Der Heijde & Van Der Heijden, 2006).  
 
H1.1.  Employees  who  see  themselves  as  more  employable  perceive  weaker 
relational  psychological  contract  obligations,  both  from  the  side  of  the 
employee and the employer.  
H1.2. Employees who see themselves as more employable perceive stronger 
balanced  psychological  contract  obligations,  both  from  the  side  of  the 
employee and the employer. 
H1.3.  Employees  who  see  themselves  as  more  employable  perceive  weaker 
transactional  psychological  contract  obligations,  both  from  the  side  of  the 
employee and the employer. 
 
The role of firm­specific capital in shaping the psychological contract 
 
The literature on the psychological contract has suggested that different 
types  of  employees  can  be  distinguished  within  contemporary 
organizations,  based  on  the  career  opportunities  available  to  them 
(Martin  &  Hetrick,  2006;  Rousseau  &  McLean  Parks,  1993).  Employees 
with a large amount of firm‐specific capital, who manage to become part 
of  an  organization’s  core,  are  still  receiving  the  promises  and  rewards 
associated  with  the  traditional‐organizational  career;  those  in  the 
171
periphery  of  the  organization  are  seen  as  more  ‘expendable’  and  thus 
need  to  rely  more  on  their  external  employability  in  order  to  achieve 
security in the labor market (Rousseau & McLean Parks, 1993).  
 
Our  first  set  of  moderation  hypotheses  relates  to  the  effects  of  talent 
category (operationalized as being identified as a high potential or not) 
on  the  different  elements  of  the  psychological  contract.  As  high 
potentials are defined as “those individuals within an organization who 
are recognized, at that point in time, as the  organization’s likely  future 
leaders” (Cope, 1998, p. 15), this variable is particularly well suited for 
operationalizing  core  employees  with  high  firm‐specific  capital.  Dries 
and Pepermans (2008), in their interview study of employees identified 
as  high  potentials,  found  that  they  not  only  received  the  majority  of 
advancement  opportunities  within  their  organizations,  but  also  the 
highest‐quality training. The non‐high potentials, consequently, seemed 
to  receive  nor  job  security,  nor  the  development  needed  to  be  highly 
employable. Other research has uncovered that employees’ evaluations 
of promises about internal career opportunities are the best predictors 
of their intention to quit and of their job search behaviors, as well as of 
their  organizational  loyalty  (e.g.  De  Vos  &  Meganck,  2009;  Rousseau, 
1995).  
 
Accordingly, we posit that being identified as a high potential moderates 
the  relationships  between  employability  and  the  different  elements  of 
the psychological contract. First of all, we expect that being identified as 
a  high  potential  decreases  the  negative  effect  of  employability  on 
relational  obligations,  i.e.  that  high  potentials  with  high  employability 
will  perceive  stronger  relational  obligations  than  non‐high  potentials 
with  high  employability.  Second,  as  high  potentials  are  believed  to 
contribute strongly to the competitive advantage of their organizations 
and  receive  specific  career  benefits  in  exchange  (Dries  &  Pepermans, 
2008),  we  expect  that  being  identified  as  a  high  potential  further 
strengthens the positive effect of employability on balanced obligations. 
Third,  we  expect  that  being  identified  as  a  high  potential  further 
strengthens  the  negative  effect  of  employability  on  transactional 
obligations.  High  potentials  are  selected,  in  part,  for  their  professional 
drive  and  their  ability  and  willingness  to  do  more  than  just  their  jobs 
(Dries & Pepermans, 2008).  
 
H2.1.  The  (negative)  effect  of  employability  on  perceived  relational 
psychological  contract  obligations  is  moderated  by  talent  category;  it  is 
weaker for  employees identified as  high potentials (i.e.,  we expect  a negative 
interaction effect).  
H2.2.  The  (positive)  effect  of  employability  on  perceived  balanced 
psychological  contract  obligations  is  moderated  by  talent  category;  it  is 
172
stronger for employees identified as high potentials  (i.e., we expect a positive 
interaction effect). 
H2.3.  The  (negative)  effect  of  employability  on  perceived  transactional 
psychological  contract  obligations  is  moderated  by  talent  category;  it  is 
stronger for employees identified as high potentials (i.e., we expect a positive 
interaction effect). 
 
Our  second  set  of  moderation  hypotheses  focuses  on  organizational 
tenure  as  a  moderator  between  employability  and  psychological 
contract obligations. Human capital theory suggests that working for an 
organization  longer  increases  an  employee’s  firm‐specific  capital. 
Individuals  are  believed  to  enlarge  the  value  of  their  firm‐specific 
human  capital  through  their  on‐the‐job  experiences  over  time, 
subsequently enhancing their productivity to their firms (Parent, 2002).  
 
Psychological  contract  research  has  demonstrated  that  employees  are 
likely  to  strive  for  balance  in  their  exchange  relationships  over  time 
(Robinson,  Kraatz  &  Rousseau,  1994;  Rousseau,  2001).  Furthermore, 
individuals  expect  that  promises  made  in  the  past  will  endure  in  the 
future.  For  example,  if  an  organization  used  to  offer  relatively  secure 
employment,  internal  career  opportunities  and  ongoing  training  and 
development,  then  individuals  are  likely  to  expect  that  they  will 
continue to be rewarded in these ways as long as they continue to meet 
their end of the bargain. This is particularly the case for mid‐ and late‐ 
career  employees  whose  past  experiences  have  been  grounded  in 
relational  psychological  contract  obligations  (Clarke  &  Patrickson, 
2008). Several authors have asserted that job and organizational tenure 
will  be  positively  correlated  with  relational  obligations  and  negatively 
correlated  with  transactional  obligations  (Mathieu  &  Zajac,  1990; 
Millward & Hopkins, 1998; Rousseau, 1989). 
 
As  for  the  moderating  effects  of  organizational  tenure,  we  expect  that 
longer  organizational  tenure  decreases  the  negative  effect  of 
employability  on  relational  obligations,  i.e.  that  employees  with  longer 
tenure  who  are  highly  employable  will  perceive  stronger  relational 
obligations  than  shorter‐tenured  employees  with  high  employability. 
Furthermore,  we  expect  longer  organizational  tenure  to  weaken  the 
positive  effect  of  employability  on  balanced  obligations.  Long‐tenured 
employees have been found to stagnate in terms of self‐initiated as well 
as  organization‐initiated  career  development  activities  (Hess  &  Jepsen, 
2009). Finally, we expect that longer organizational tenure strengthens 
the  negative  effect  of  employability  on  transactional  obligations.  The 
transactional  element  of  the  psychological  contract  implies  short‐term 
obligations between employees and employers, whereas longer‐tenured 
173
employees  tend  to  perceive  their  relationship  with  their  employer  as 
long‐term (Clarke & Patrickson, 2008). 
 
H3.1.  The  (negative)  effect  of  employability  on  perceived  relational 
psychological  contract  obligations  is  moderated  by  internal  tenure;  it  is 
weaker  for  employees  with  longer  organizational  tenure  (i.e.,  we  expect  a 
negative interaction effect). 
H3.2.  The  (positive)  effect  of  employability  on  perceived  balanced 
psychological  contract  obligations  is  moderated  by  internal  tenure;  it  is 
weaker  for  employees  with  longer  organizational  tenure  (i.e.,  we  expect  a 
negative interaction effect). 
H3.3.  The  (negative)  effect  of  employability  on  perceived  transactional 
psychological contract obligations is moderated by internal tenure; the effect 
is stronger for employees with longer organizational tenure (i.e., we expect a 
positive interaction effect). 
 
This study aims to contribute to the work and organizational psychology 
literature in general, and the postmodern careers literature specifically, 
in  a  fourfold  manner.  First  of  all,  by  empirically  examining  the 
relationship  between  employability  and  specific  elements  of  the 
psychological  contract.  Although  there  is  much  discourse  about  such  a 
relationship  (e.g.  (Clarke  &  Patrickson,  2008;  Van  Buren,  2003),  there 
has  not  been  much  empirical  work  looking  into  these  assumptions. 
Second,  by  applying  nuance  in  the  prediction  model  through  the 
incorporation  of  two  indicators  of  firm‐specific  capital  (i.e. 
organizational  tenure  and  being  identified  as  a  high  potential)  as 
moderators.  We  strongly  believe  that  the  bargaining  position  of 
individual employees (determined, in part, by their firm‐specific capital) 
is  of  pivotal  importance  in  the  process  of  constructing  psychological 
contracts (Guest, 2001). Third, by critically examining some popular, but 
under‐examined,  assumptions  about  modern‐day  careers.  One  specific 
assumption  that  will  be  tackled  is  the  assumption  that  employees  are 
less  and  less  concerned  with  job  security  and  intra‐organizational 
career opportunities (Tulgan, 2001). Fourth, by building on data coming 
from employees identified as high potentials (who are then compared to 
a  sample  of  non‐high  potentials).  Although  identified  as  one  of  the  key 
issues  facing  21st  century  HRM  practice  (Buckingham  &  Vosburgh, 
2001), talent management and high potential career management have 
hardly  been  researched.  Literature  on  the  topic  is  based  mostly  on 
conjecture and experiences from practice (Dries & Pepermans, 2008).  
 
 
 
 
 
 
174
Method 
 
Research setting and procedure 
 
The  current  study  was  part  of  a  larger  Belgian  research  project 
concerned  with  current‐day  talent  management  practice.  A  large 
number  of  organizations  known  to  engage  in  talent  management  were 
asked  for  their  participation  in  the  project.  Composing  the  sample 
proved  an  extensive  process.  First  of  all,  many  organizations  were 
hesitant to participate in  the research due to their discrete procedures 
about  talent  management  (for  instance,  they  did  not  communicate  to 
employees  who  was  identified  as  a  high  potential  and  who  was  not). 
These  organizations  were  typically  concerned  that  participating  in 
studies  about  talent  management  would  cause  some  sort  of 
confidentiality  breach.  Second,  of  the  organizations  who  did  want  to 
participate  in  the  study,  a  large  number  were  in  an  immature  stage  of 
talent  management  implementation,  meaning  that  they  had  not  yet 
‘officially’ identified employees as high potentials, which excluded them 
from  participation  in  the  study.  Finally,  in  order  to  reduce  possible 
heterogeneity  in  our  study  sample,  the  definition  on  which 
organizations’ high potential identification was based had to correspond 
to  the  one  used  in  our  research  (i.e.  the  definition  by  Cope,  1998). 
Earlier  work  on  talent  management  has  demonstrated  that 
organizations  use  many  different  terms  and  definitions  in  order  to 
operationalize  what  ‘talent’  means  (Dries  &  Pepermans,  2008).    In  the 
end,  five  large  Belgian  organizations  participated  in  this  particular 
study. They were each active in a different sector and there was a strong 
variation  in  organization  size:  distribution  (approx.  20000  employees), 
telecom  (approx.  15000  employees),  HR  consulting  (approx.  3500 
employees),  food  (approx.  2500  employees)  and  advertising  (approx. 
1000 employees).  
 
Each  organization  assigned  an  employee  from  the  HR  department  as 
coordinator  for  the  research  project.  These  coordinators  were 
responsible  for  distributing  the  survey  among  targeted  potential 
respondents  within  their  organization.  Each  of  them  was  asked  to 
compose a sample of employees identified as high potentials and a list of 
average  performers  who  were  considered  solid  contributors,  but  not 
high potentials (we added this clarification so that organizations would 
not  select  poor  performers  for  the  non‐high  potential  sample). 
Furthermore,  we  asked  them  to  balance  the  sample  in  terms  of  longer 
and  shorter  organizational  tenure.  Participation  in  the  survey  was 
voluntary,  although  employees  were  encouraged  to  participate.  Each 
175
coordinator  sent  out    three  follow‐up  e‐mails  requesting  respondent 
participation over a period of, on average, six weeks.  
 
An  important  issue  for  the  participating  organizations  was  that  the 
survey itself would not disclose whether respondents were identified as 
high  potentials  or  not  (in  all  five  organizations,  high  potentials  were 
informally  notified  of  their  status,  but  non‐high  potentials  were  not). 
Furthermore,  we  were  asked  to  ensure  respondent  anonymity.  We 
designed  the  study  accordingly,  in  that  two  identical  online  surveys 
were  developed,  placed  on  two  (minimally)  different  URLs.  The  HR 
coordinators  then  made  up  two  sets  of  e‐mails  for  potential 
respondents. High potentials received an e‐mail with one URL; non‐high 
potentials  with  the  other.  Apart  from  the  URL,  all  instructions  to 
respondents  were  identical.  The  survey  did  not  ask  for  respondents’ 
names  or  any  other  information  that  might  compromise  their 
anonymity.  In  the  analyses,  we  were  able  to  discern  from  the  URL 
respondents  had  used  whether  they  belonged  to  the  high  potential  or 
the non‐high potential category.  
 
An  important  contextual  factor  of  the  study  was  its  timeframe.  The 
study  data  were  collected  between  October  and  December  of  2008,  a 
time during which the global economic crises was strongly mediatized. 
It  is  very  likely  that  the  specific  economic  climate  within  which  the 
study  took  place  influenced  some  of  its  results  (for  instance,  the 
importance  attached  to  job  security).  We  must  take  this  issue  into 
account in interpreting our findings.  
 
Sample demographics 
 
A  total  of  103  respondents  filled  out  the  online  survey;  49  (48%)  had 
been  identified  as  high  potentials  at  the  time  of  data  collection  and  54 
(52%) had not. Of the high potentials, 25 (51%) were men and 24 (49%) 
were women. Their average age was 34.70 (sd = 6.43). Most of them had 
gone through either a university (n = 17, 35%) or a college education (n 
=  21,  43%).  As  regards  functional  level,  19  of  the  high  potential 
respondents  were  at  the  non‐managerial  level  (39%),  10  were  line 
managers  (20%),  17  were  middle  managers  (35%),  and  3  were  at  the 
executive  level  (6%).  Their  average  organizational  tenure  was  6.49 
years (sd = 5.73).  
 
Of  the  non‐high  potentials,  41  (76%)  were  men  and  13  (24%)  were 
women. Their average age was 42.11 (sd = 8.07). Most of them had gone 
through either a university (n = 13, 24%) or a college education (n = 20, 
176
37%).  As  regards  functional  level,  23  of  the  non‐high  potential 
respondents  were  at  the  non‐managerial  level  (43%),  11  were  line 
managers (20%), 20 were middle managers (37%); none of them were 
at the executive level. The average organizational tenure in years for the 
non‐high potentials was 15.38 (sd = 8.97).  
 
Measures 
 
The  original  versions  of  the  scales  in  the  survey  were  translated  to 
Dutch through translation ‐ back translation. Scale scores were created 
by averaging item scores per scale. Coefficient alphas were added on the 
main diagonal of the correlation matrix in Table 1.  
 
Employability. Employability was measured using the 25‐item scale 
developed  by  Fugate  and  Kinicki  (2008).  A  sample  item  is  ‘I  am 
optimistic  about  my  future  career  opportunities’.  Respondents  were 
instructed to indicate the extent to which they agreed to the statements 
in  the  items  on  a  seven‐point  Likert  scale  ranging  from  ‘strongly 
disagree’ to ‘strongly agree’.  
 
Psychological  contract  obligations.  In  order  to  assess  respondents’ 
perceptions  about  psychological  contract  obligations,  we  used  the 
Psychological  Contract  Inventory  (PCI;  Rousseau,  2000).  We 
incorporated three of the four scales of the PCI in our online survey, i.e. 
relational, balanced and transactional (we did not use the ‘transitional’ 
scale as it was less relevant for our research goals). In the PCI, each scale 
is  conceptualized  as  comprising  several  subscales,  each  of  them 
applicable  to  both  employee  and  employer  obligations.  First  of  all,  the 
relational scale contains two subscales, i.e. stability and loyalty. Stability, 
in terms of employee obligations, refers to remaining with the firm and 
to  doing  what  is  required  to  keep  one’s  job;  in  terms  of  employer 
obligations, it means offering stable wages and long‐term employment. 
Loyalty,  from  an  employee  obligations  perspective  denotes  supporting 
the  firm,  manifesting  loyalty  and  commitment  to  the  organization’s 
needs and interests. From the employer’s side, it implies supporting the 
well‐being  and  interests  of  employees  and  their  families.  Second,  the 
balanced  scale  contains  three  subscales,  i.e.  external  employability, 
internal  advancement  and  dynamic  performance.  As  for  employee 
obligations, external employability implies developing marketable skills; 
as  regards  employer  obligations,  it  refers  to  enhancing  worker’s  long‐
term employability outside the organization as well as within it. Internal 
advancement  from  the  employee  obligations  side  is  defined  as 
developing  skills  valued  by  one’s  current  employer.  Employer 
177
obligations  in  this  respect  relate  to  creating  employee  career 
development  opportunities  within  the  firm.  Dynamic  performance 
obligations  refer  to  successfully  achieving  new  and  increasingly 
demanding  goals,  to  help  the  firm  become  and  remain  competitive  for 
employees;  for  employers,  it  denotes  promoting  continuous  learning 
and  helping  employees  successfully  execute  escalating  performance 
requirements. Third, the transactional scale contains two subscales, i.e. 
narrow  and  short‐term.  Narrow  obligations,  from  the  side  of  the 
employee, denote performing only a fixed or limited set of duties; doing 
only what one is paid to do. From the employer’s side, it implies offering 
employees  only  limited  involvement  in  the  organization,  little  or  no 
training or other employee development. Finally, short‐term obligations 
suggest  employees  not  being  obligated  to  remain  with  the  firm  and 
commit to work only for a limited amount of time. From the employer’s 
side,  it  means  offering  employment  for  only  a  specific  or  limited  time, 
and not being obligated to future commitments.  
 
Each subscale of the PCI is composed out of four items about employee 
obligations  and  four  items  about  employer  obligations,  each 
accompanied  by  the  following  instruction:  “Consider  your  relationship 
with  your  current  employer.  To  what  extent  has  your  employer  made 
the  following  commitment  or  obligation  to  you?”  or  “To  what  extent 
have  you  made  the  following  commitment  or  obligation  to  your 
employer?”  Items  were  scored  on  a  seven‐point  Likert  scale  ranging 
from ‘not at all’ to ‘to a great extent’. Sample items are ‘Concern for my 
long‐term  well‐being’  (classified  as  a  loyalty  employer  obligation)  and 
‘Build  skills  to  increase  my  value  in  this  organization’  (classified  as  an 
internal  advancement  employee  obligation).  Analyses  were  carried  out 
on  the  subscale  level,  and  not  the  scale  level,  as  we  felt  that  averaging 
scores  over  the  subscales  to  obtain  an  aggregate  scale  score  would 
jeopardize  the  measure’s  validity  and  lead  to  a  lack  of  nuance  in 
interpreting the findings. For instance, it is conceivable that respondents 
perceive internal advancement obligations made by their employer to a 
large  extent,  but  not  external  employability  obligations.  Averaging  out 
these  responses  would  then  lead  to  inadequate  interpretations  of  the 
data. Other researchers (e.g. Porter, Pearce, Tripoli & Lewis, 1998) have 
reported  similar  concerns,  causing  them  to  work  with  individual  items 
rather than scales. Coefficient alphas (see Table 1) were .60 or above for 
most  subscales,  indicating  satisfactory  internal  consistency  (Hair, 
Anderson,  Tatham  &  Black,  1998).  However,  for  the  two  transactional 
employer obligations subscales, alpha values were suboptimal. Previous 
studies  have  already  discussed  the  issue  of  the  PCI  possibly  being 
178
unstable  across  research  samples  (Arnold,  1996;  Conway  &  Briner, 
2005).  
 
Organizational  tenure.  Organizational  tenure  was  measured  with 
one open‐ended item, i.e. “for how long have you been working for your 
current  employer?”  Respondents  were  instructed  to  reply  in  terms  of 
years and months.   
 
Control variables 
 
As the sample demographics for employees identified as high potentials 
versus  non‐high  potentials  revealed  possible  significant  differences 
between  the  two  groups  in  terms  of  background  variables,  we 
performed  independent  sample  t‐tests  (for  the  continuous  variables) 
and  χ²  difference  tests  (for  the  categorical  variables)  to  check  for 
unwanted variance in the data. We found that, on average, the non‐high 
potential  respondents  contained  a  significantly  larger  proportion  of 
males,  compared  to  the  high  potential  respondents  (χ²(1)  =  5.88,  p  < 
.05),  and  that  the  high  potentials  were  significantly  younger  (t(99)  = 
5.06,  p  <  .001).  Furthermore,  the  high  potentials  reported  significantly 
lower  organizational  tenure  (t(100)  =  5.91,  p  <  .001).  Organizational 
tenure could, of course, not be incorporated as a control variable since it 
is one of the independent variables in the model. Furthermore, because 
of the high correlation between organizational tenure and age (r = .77, p 
<  .001),  we  could  also  not  incorporate  age  as  a  control  due  to  ensuing 
multicollinearity  issues  in  our  model.  Consequently,  only  gender  was 
added as a control variable in the analyses.  
 
Analysis 
 
Hierarchical regression analysis was applied to test for the hypothesized 
effects.  The  independent  research  variables  were  standardized  into  z‐
scores  in  order  to  facilitate  interpretation  of  the  interaction  effects 
(Cohen,  Cohen,  West,  &  Aiken,  2003).  Separate  hierarchical  multiple 
regression  analyses  for  each  psychological  contract  type  were 
conducted  to  test  the  moderator  influence  of  talent  category  and 
organizational  tenure  in  the  relationship  between  employability  and 
perceived  psychological  contract  obligations.  In  the  first  step,  gender 
(i.e. the control variable) was entered into the regression equation as a 
predictor. In the second  step, we entered employability. The third step 
added talent category and tenure to the regression equation. Finally, the 
interaction  terms  between  employability  and  talent  category  and 
between employability and organizational tenure were entered into the 
179
model.  Analyses  were  conducted  with  the  SPSS  17.0  program.  In  the 
Results section, we report only significant β coefficients for independent 
variables (and interaction terms) from models (i.e. model steps) with a 
significant ∆ R² (Pallant, 2007).   
 
Results 
 
Table  1  presents  the  means,  standard  deviations,  and  correlations 
among  the  study  variables.  Some  interesting  points  emerge  from  these 
preliminary analyses. To begin with, our sample reported a rather high 
average  organizational  tenure  (m  =  16.00,  sd  =  9.19),  which  is  an 
important  contextual  side  note  considering  our  research  questions. 
Furthermore,  looking  at  the  psychological  contract  obligations  scores, 
we  see  that  both  for  employee  and  employer  obligations  the  lowest 
scores  were  given  to  external  employability  obligations  and  to 
transactional  obligations.  It  thus  appears  that  respondents,  in  general, 
perceive  stronger  obligations  relating  to  intra‐organizational  dynamics 
compared  to  obligations  relating  to  inter‐organizational  dynamics.  We 
go  further  into  these  preliminary  findings  later  on  in  the  Discussion. 
Another  interesting  finding  is  that  correlations  between  employee  and 
employer  obligations  for  each  psychological  contract  element, 
respectively,  are  consistently  positive  and  significant,  which  can  be 
interpreted  as  a  confirmation  of  the  notion  that  employees  strive  for 
balance in their psychological contracts (Robinson et al., 1994). Finally, 
with  respect  to  the  question  of  relational  and  transactional  contracts 
being  two  ends  of  the  same  spectrum  or  two  separate  constructs 
(Arnold, 1996; Conway & Briner, 2005; Guest, 1998), we did find several 
negative correlations between the different relational and transactional 
employee  and  employer  obligations.  However,  the  magnitude  of  the 
coefficients (around .20) implies related but separate constructs.   
 
Tables 2 through 7 give an overview of the main and interaction effects 
of  employability,  talent  category  and  tenure  on  the  different  types  of 
perceived  psychological  contract  obligations.  Looking  at  the  R²  values 
for the different regression models, we see that effect sizes range from 
.07  to  .32  for  the  significant  models,  with  the  vast  majority  of  models 
demonstrating  an  R²  of  over  .14.  Only  for  short‐term  (transactional) 
employee obligations, none of the prediction models was significant.  
 
 
 
 
 
Table 1. Means, standard deviations and intercorrelations of the study variables (n = 103) 
 
  m  sd 1  2  3  4  5  6  7  8  9 
Employee obligations       
1. Relational (Stability)  4.92  1.31 (.81)    
2. Relational (Loyalty)  5.31  1.03 .41** (.80) a    
3. Balanced (External employability)  3.41  1.35 ‐.24* ‐.04 (.88)    
4. Balanced (Internal advancement)  5.07  .87 .20* .56** .27** (.72)    
5. Balanced (Dynamic performance)  5.14  .89 .21* .53** .34** .71** (.75)   
6. Transactional (Narrow)  2.25  .87 .04 ‐.21* .23* ‐.22* ‐.22*  (.78) 
7. Transactional (Short term)  3.41  1.34 ‐.24* .07 .40** .12 .10  .23*  (.77)
Employer obligations    .19    
8. Relational (Stability)  4.36  1.08 .45** .32** ‐.14 .18 .02  .00  ‐.09 (.75)
9. Relational (Loyalty)  5.40  .92 .19 .35** ‐.10 .31** .04  .02  ‐.08 .45** (.84)
10. Balanced (External employability)  3.72  1.31 .04 .21* .51** .38** .17  .10  .02 .21* .43**
11. Balanced (Internal advancement)  4.68  1.37 .20* .31** ‐.04 .43** .16  .07  ‐.22* .32** .56**
180

12. Balanced (Dynamic performance)  4.72  1.00 .22* .48** ‐.04 .44** .21*  .00  ‐.16 .37** .61**
13. Transactional (Narrow)  3.39  .93 ‐.05 ‐.17 .19 ‐.25* ‐.07  .45**  .15 ‐.19 ‐.22*
14. Transactional (Short term)  2.90  1.09 ‐.37** ‐.22* .20* ‐.12 .01  .23*  .40** ‐.37** ‐.19*
15. Employability  5.22  .58 .09 .38** .22* .54** .40**  ‐.21*  .05 .13 .26**
16. Tenure  16.00  9.19 .20* .14 ‐.35** ‐.07 ‐.17  ‐.42**  ‐.20* .27** .17
Notes. Cronbach’s alphas were added on the main diagonal. 
a One item was removed from the scale (the original α‐value was .57). 
b α‐values could not be improved by removing items from the scales. 

* p < .05; ** p < .01. 
Table 1. Continued 
 
10  11  12  13  14  15  16 
         
         
         
         
         
         
         
         
         
         
         
(.85)         
.54*  (.93)       
181

.59**  .82**  (.80)     


‐.04  ‐.42**  ‐.30**  (.53) b   
‐.02  ‐.17  ‐.26**  .26**  (.57) b 
.51**  .43**  .52**  .01  ‐.09  ‐‐
‐.09  ‐.03  .05  ‐.26**  ‐.27**  .00 ‐‐
Notes. Cronbach’s alphas were added on the main diagonal. 
a One item was removed from the scale (the original α‐value was .57). 
b α‐values could not be improved by removing items from the scales. 

* p < .05; ** p < .01. 
 
 
182
Main effects of employability 
 
As  for  the  relational  psychological  contract  obligations,  the  results  are 
presented  in  Tables  2  and  3.  We  see  that  the  main  effects  that  were 
found run in the opposite direction as hypothesized in H1.1. According 
to  our  findings,  employees  who  see  themselves  as  more  employable 
perceive stronger stability and loyalty employee obligations, and loyalty 
employer obligations (where we hypothesized a negative relationship). 
No main effects were found for stability employer obligations. We must 
thus reject H1.1. 
 
Tables 4 and 5 present the effects of employability, talent category and 
tenure  on  the  balanced  obligations.  Both  for  employee  and  employer 
obligations, we found positive main effects of employability on all three 
balanced  obligation  subscales:  external  employability,  internal 
advancement and dynamic performance. Hence, H1.2 is accepted.  
 
Finally,  the  results  that  were  found  in  terms  of  the  transactional 
psychological contract obligations are exhibited in Tables 6 and 7. Only 
one  main  effect  was  found  (corresponding  to  the  hypothesized 
direction),  i.e.  a  negative  effect  of  employability  on  narrow  employee 
obligations. Consequently, we can only partially accept H1.3. 
 
Interaction effects of employability and talent category 
 
Only  one  significant  interaction  effect  of  employability  and  talent 
category  was  found  (Table  2):  a  negative  effect  on  stability  employee 
obligations.  This  finding  indicates  that,  under  conditions  of  low 
employability, high potentials are more likely to perceive an obligation 
to  stay  with  their  organizations  than  non‐high  potentials;  and  under 
conditions of high employability, high potentials are less likely to do so. 
Although  this  is  an  intriguing  finding,  it  does  not  correspond  to  H2.1. 
Since  no  other  interaction  effects  were  found  with  respect  to  talent 
category, we also reject H2.2 and H2.3.  
 
Interestingly,  although  not  explicitly  hypothesized,  we  did  find  several 
main  effects  of  talent  category  on  psychological  contract  obligations. 
High potentials were found to perceive stronger stability employer (but 
not  employee)  obligations,  weaker  external  employability  employee 
obligations and weaker narrow employer obligations.  
 
 
 
Table 2. Hierarchical regression analyses examining main and interaction effects of employability, talent category and organizational tenure on 
perceived relational employee psychological contract obligations  
 
  Perceived employee obligations: Relational 
  Stability  Loyalty 
  Model  Coefficients  Model  Coefficients 
  R2(F)  Δ R2(Δ F)  β  SE  R2(F)  Δ R2(Δ F)  β  SE 
Step 1.  .02  .02      .00  .00     
Gender  (2.39)  (2.39)  .32  .21  (.00)  (.00)  .00  .21 
Step 2.  .04  .02      .14  .14     
Gender  (1.87)  (1.34)  .35  .21  (16.71**)  (8.36**)  .11  .20 
Employability       .12  .10      .38**  .09 
Step 3.  .07  .03      .15  .01     
Gender  (1.82)  (1.75)  .34  .21  (.13)  (4.17**)  .10  .20 
183

Employability       .12  .10      .38**  .10 


Talent category       .17  .23      ‐.10  .22 
Tenure      .21  .11      ‐.05  .11 
Step 4.  .14  .07      .18  .03     
Gender  (2.50*)  (3.66*)  .38  .21  (1.96)  (3.49**)  .11  .20 
Employability       .31*  .14      .45**  .14 
Talent category (TC)      .26  .23      ‐.03  .22 
Tenure      .28**  .11      .01  .11 
Employability x TC      ‐.41*  .20      ‐.17  .20 
Employability x Tenure      ‐.24*  .11      ‐.20*  .10 
Notes. Prior to being entered into the regression analysis, all variables were standardized into z‐scores; Talent category: high potentials = 1, non‐
high potentials = 0; Gender: men = 1, women = 0; * p < .05; ** p < .01. 
Table 3. Hierarchical regression analyses examining main and interaction effects of employability, talent category and organizational tenure on 
perceived relational employer psychological contract obligations 
 
  Perceived employer obligations: Relational 
  Stability  Loyalty 
  Model  Coefficients  Model  Coefficients 
  R2(F)  Δ R2(Δ F)  β  SE  R2(F)  Δ R2(Δ F)  β  SE 
Step 1.  .00  .00      .00  .00     
Gender  (.10)  (.10)  .06  .21  (.17)  (.17)  .09  .21 
Step 2.  .02  .02      .07  .07     
Gender  (1.03)  (1.97)  .10  .21  (3.93*)  (7.67**)  .16  .20 
Employability       .14  .10      .27**  .10 
Step 3.  .18  .16      .11  .04     
Gender  (5.21**)  (9.21**)  .13  .20  (3.11**)  (2.20)  .22  .21 
184

Employability       .16  .09      .28**  .10 


Talent category      .55**  .22      .42*  .23 
Tenure      .46**  .11      .20*  .11 
Step 4.  .19  .01      .14  .03     
Gender  (3.81**)  (1.02)  .16  .20  (2.66*)  (1.67)  .23  .21 
Employability       .28*  .13      .28*  .14 
Talent category (TC)      .59**  .22      .36  .23 
Tenure      .48**  .11      .15  .11 
Employability x TC      ‐.26  .19      .03  .20 
Employability x Tenure      ‐.08  .10      .19  .11 
Notes. Prior to being entered into the regression analysis, all variables were standardized into z‐scores; Talent category: high potentials = 1, non‐
high potentials = 0; Gender: men = 1, women = 0; * p < .05; ** p < .01. 
Table 4. Hierarchical regression analyses examining main and interaction effects of employability, talent category and organizational tenure on 
perceived balanced employee psychological contract obligations 
 
  Perceived employee obligations: Balanced 
  External employability  Internal advancement  Dynamic performance 
  Model  Coefficients  Model  Coefficients  Model  Coefficients 
  R2(F)  Δ R2(Δ F)  β  SE  R2 (F)  Δ R2(Δ F)  β  SE  R2 (F)  Δ R2(Δ F)  β  SE 
Step 1.  .05  .05      .00  .00      .01  .01     
Gender  (5.18*)  (5.18*)  ‐.46*  .20  (.40)  (.40)  ‐.13  .21  (1.36)  (1.36)  ‐.24  .21 
Step 2.  .09  .04      .29  .29      .16  .15     
Gender  (4.63**)  (3.93*)  ‐.41*  .20  (20.61**)  (40.65**)  .02  .18  (9.69**)  (17.79**)  ‐.13  .19 
Employability       .19*  .10      .54**  .09      .39**  .09 
Step 3.  .28  .19      .31  .02      .22  .06     
Gender  (9.56**)  (13.34**)  ‐.42*  .19  (11.03**)  (1.32)  .04  .18  (6.79**)  (3.41*)  ‐.11  .20 
185

Employability       .18*  .09      .54**  .09      .38**  .09 


Talent category      ‐.54**  .20      ‐.09  .20      ‐.18  .21 
Tenure      ‐.52**  .10      ‐.15  .10      ‐.27**  .10 
Step 4.  .29  .01      .35  .04      .23  .01     
Gender  (6.51**)  (.58)  ‐.42*  .19  (8.43**)  (2.53)  .02  .18  (4.62**)  (.44)  ‐.10  .20 
Employability       .19  .13      .42**  .12      .46**  .13 
Talent category (TC)      ‐.51**  .21      ‐.16  .20      ‐.15  .22 
Tenure      ‐.49**  .10      ‐.21*  .10      ‐.25*  .11 
Employability x TC      ‐.03  .18      .27  .17      ‐.15  .19 
Employability x Tenure      ‐.10  .10      .18*  .09      ‐.07  .10 
Notes. Prior to being entered into the regression analysis, all variables were standardized into z‐scores; Talent category: high potentials = 1, non‐
high potentials = 0; Gender: men = 1, women = 0; * p < .05; ** p < .01. 
Table 5. Hierarchical regression analyses examining main and interaction effects of employability, talent category and organizational tenure on 
perceived balanced employer psychological contract obligations 
 
  Perceived employer obligations: Balanced 
  External employability  Internal advancement  Dynamic performance 
  Model  Coefficients  Model  Coefficients  Model  Coefficients 
  R2(F)  Δ R2(Δ  β  SE  R2(F)  Δ R2(Δ  β  SE  R2(F)  Δ R2(Δ F)  β  SE 
Step 1.  .06  .06      .00  .00      .01  .01     
Gender  (6.03**)  (6.03**)  ‐.50*  .20  (.42)  (.42)  ‐.13  .21  (.51)  (.51)  ‐.15  .21 
Step 2.  .29  .23      .19  .19      .28  .27     
Gender  (20.55**)  (33.12**)  ‐.35*  .18  (11.44**)  (22.37* ‐.01  .19  (18.84* (36.98**)  .00  .18 
Employability       .49**  .09      .43**  .09      .53**  .09 
Step 3.  .31  .02      .22  .03      .28  .00     
Gender  (10.70**)  (.89)  ‐.33*  .18  (6.76**)  (1.88)  .08  .20  (9.47**)  (.35)  .01  .19 
186

Employability       .49**  .09      .44**  .09      .53**  .09 


Talent category      ‐.04  .20      .41*  .21      .10  .21 
Tenure      ‐.12  .10      .09  .10      .08  .10 
Step 4.  .31  .00      .29  .07      .32  .04     
Gender  (7.11**)  (.26)  ‐.32  .19  (6.45**)  (4.78**)  .11  .19  (7.48**)  (2.79)  .04  .19 
Employability       .54**  .12      .53**  .13      .63**  .12 
Talent category (TC)      ‐.02  .21      .35  .21      .06  .20 
Tenure      ‐.11  .10      .03  .10      .04  .10 
Employability x TC      ‐.12  .18      ‐.14  .18      ‐.17  .18 
Employability x Tenure      ‐.04  .09      .25**  .10      .17*  .09 
Notes. Prior to being entered into the regression analysis, all variables were standardized into z‐scores; Talent category: high potentials = 1, non‐
high potentials = 0; Gender: men = 1, women = 0; * p < .05; ** p < .01.
Table 6. Hierarchical regression analyses examining main and interaction effects of employability, talent category and organizational tenure on 
perceived transactional employee psychological contract obligations 
 
  Perceived employee obligations: Transactional 
  Narrow  Short­term 
  Model  Coefficients  Model  Coefficients 
  R2(F)  Δ R2(Δ F)  β  SE  R2(F)  Δ R(Δ F)  β  SE 
Step 1.  .00  .00      .00  .00     
Gender  (.06)  (.06)  ‐.05  .21  (.01)  (.01)  ‐.02  .21 
Step 2.  .05  .05      .00  .00     
Gender  (2.51)  (4.95*)  ‐.12  .21  (.14)  (.27)  ‐.01  .21 
Employability       ‐.22*  .10      .05  .10 
Step 3.  .17  .12      .05  .05     
Gender  (4.92**)  (7.04**)  ‐.08  .20  (1.31)  (2.48)  ‐.01  .21 
187

Employability       ‐.23**  .09      .05  .10 


Talent category      ‐.27  .22      ‐.25  .24 
Tenure      ‐.40**  .11      ‐.26*  .12 
Step 4.  .17  .00      .07  .02     
Gender  (3.26**)  (.10)  ‐.09  .20  (1.26)  (1.16)  ‐.04  .22 
Employability       ‐.25*  .14      ‐.09  .14 
Talent category (TC)      ‐.26  .23      ‐.25  .24 
Tenure      ‐.39**  .11      ‐.25*  .12 
Employability x TC      .04  .20      .27  .21 
Employability x Tenure      ‐.03  .10      ‐.04  .11 
Notes. Prior to being entered into the regression analysis, all variables were standardized into z‐scores; Talent category: high potentials = 1, non‐
high potentials = 0; Gender: men = 1, women = 0; * p < .05; ** p < .01. 
Table 7. Hierarchical regression analyses examining main and interaction effects of employability, talent category and organizational tenure on 
perceived transactional employer psychological contract obligations 
 
  Perceived employer obligations: Transactional 
  Narrow  Short­term 
  Model  Coefficients  Model  Coefficients 
  R2(F)  Δ R2(Δ F)  β  SE  R2(F)  Δ R2(Δ F)  β  SE 
Step 1.  .01  .01      .00  .00     
Gender  (.56)  (.56)  .16  .21  (.33)  (.33)  .12  .21 
Step 2.  .01  .00      .01  .01     
Gender  (.30)  (.04)  .16  .21  (.52)  (.71)  .09  .21 
Employability       .02  .10      ‐.09  .10 
Step 3.  .10  .09      .09  .08     
Gender  (2.54*)  (4.75**)  .10  .21  (2.28*)  (4.00*)  .18  .21 
188

Employability       .01  .10      ‐.09  .10 


Talent category      ‐.53*  .23      .08  .23 
Tenure      ‐.33**  .11      ‐.26*  .11 
Step 4.  .17  .07      .09  .00     
Gender  (3.34**)  (4.57**)  .11  .20  (1.59)  (.28)  .16  .21 
Employability       .06  .14      ‐.16  .14 
Talent category (TC)      ‐.43*  .22      .07  .24 
Tenure      ‐.24*  .11      ‐.26*  .12 
Employability x TC      ‐.15  .20      .15  .21 
Employability x Tenure      ‐.31**  .10      .00  .11 
Notes. Prior to being entered into the regression analysis, all variables were standardized into z‐scores; Talent category: high potentials = 1, non‐
high potentials = 0; Gender: men = 1, women = 0; * p < .05; ** p < .01. 
 
189
Interaction effects of employability and organizational tenure 
 
As  concerns  interaction  effects  with  organizational  tenure,  we  found 
several  significant  effects,  albeit  not  in  the  hypothesized  direction. 
Consequently, we must reject H3.1, H3.2 and H3.3. Results indicate that 
higher  employability  weakens  the  positive  effect  of  tenure  on  stability 
employee  obligations;  strengthens  the  positive  effect  of  employability 
on  internal  advancement  employer  obligations;  and  strengthens  the 
positive  effect  of  employability  on  dynamic  performance  employer 
obligations.    
 
For  organizational  tenure,  as  well,  we  found  several  interesting  main 
effects.  We  found  positive  effects  of  tenure  on  stability  employee  and 
employer  psychological  contract  obligations.  Negative  effects  of  tenure 
were  found  for  external  employability  employee  obligations,  dynamic 
performance  employee  obligations,  narrow  employee  and  employer 
obligations, and short‐term employer obligations. 
 
Discussion 
 
The  survey  study  reported  in  this  paper  investigated  the  relationships 
between  employability,  perceived  psychological  contract  obligations, 
talent  category  (i.e.  being  identified  as  a  high  potential  or  not)  and 
organizational tenure. Some interesting results were found, although for 
the  most  part  inconsistent  with  our  hypothesized  moderation  model. 
Overall,  it  appears  that  the  relationships  between  employability  and 
perceived  psychological  contract  obligations  are  less  straightforward 
and more complex than we would expect based on the literature.  
 
To  begin  with,  we  found  that  higher  employability  leads  to  stronger 
perceived  relational  psychological  contract  obligations,  where  we 
expected the opposite. This finding indicates that employees with higher 
employability  do  not  necessarily  have  a  more  inter‐organizational 
mobility  focus, or in other words, aspire to more boundaryless careers 
in  terms  of  crossing  organizational  boundaries.  Furthermore, 
employability  also  does  not  decrease  the  beliefs  of  employees  with 
respect  to  long‐term  employment  security  offered  by  their 
organizations. The thesis put forward in the literature that the downfall 
of  the  traditional‐organizational  career  might  be  compensated  by 
developing individuals’ employability thus seems to be overly simplistic 
(Van Buren, 2003). Furthermore, the data showed that our respondents, 
overall,  perceive  stronger  obligations  relating  to  intra‐organizational 
psychological  contract  terms  (i.e.  relational  obligations,  internal 
190
advancement  and  dynamic  performance  obligations)  than  obligations 
relating  to  inter‐organizational  psychological  contract  terms  (i.e. 
external  employability  obligations,  transactional  obligations).  In 
general,  the  importance  people  attach  to  job  security,  even  in  today’s 
‘boundaryless career world’, is underestimated in the careers literature 
when  we  consider  the  empirical  evidence.  Despite  the  fact  that 
traditional careers have been declared dead repeatedly over the last few 
decades,  a  volume  of  research  indicates  that  accounts  about  the  speed 
and  inevitability  of  the  shift  from  traditional‐organizational  to  more 
boundaryless  career  types  have  to  be  put  into  perspective  (Baruch, 
2006;  Briscoe  &  Finkelstein,  2009;  Guest  &  MacKenzie  Davey,  1996; 
Granrose  &  Baccili,  2006;  Sullivan,  1999;  Verbruggen  et    al.,  2007). 
Especially  in  an  economic  climate  characterized  by  a  worldwide 
financial  crisis,  job  security  is  expected  to  be  important  to  employees 
(Barrington,  2002;  Collings  &  Mellahi,  2009).  Consequently,  an 
alternative  explanation  for  the  fact  that  higher  employability  leads  to 
stronger  perceived  relational  obligations  is  that,  like  employees  with 
high  firm‐specific  capital,  individuals  with  stronger  career  self‐
management  skills  and  higher  employability  are  in  a  stronger 
bargaining  position  in  terms  of  dictating  their  exchange  relationships 
with their employers. As a result, they might be more able to secure the 
finite  amount  of  much‐desired  relational  career  benefits  within  their 
organizations (Guest, 2001; Van Buren, 2003).  
 
As  for  perceived  balanced  psychological  contract  obligations,  our 
findings  were  mostly  in  line  with  expectations.  People  with  high 
employability  perceive  a  stronger  obligation  of  themselves  and  their 
organizations  to  co‐invest  in  external  employability,  internal  career 
advancement opportunities and continuous performance improvement. 
This  finding  makes  obvious  sense  since  the  notion  of  employability  is 
explicitly  part  of  the  balanced  psychological  contract  concept  as 
developed  by  Rousseau  (2000).  It  is  important  to  note,  however,  that 
high  employability  also  leads  to  higher  perceived  balanced  obligations 
from  the  side  of  the  employer.  Again,  this  implies  that  individual 
employability  is  by  no  means  a  substitute  for  organizational  career 
support.  
 
Furthermore,  we  found  no  relationship  between  employability  and 
short‐term  employment  obligations  from  the  side  of  the  employer, 
which  implies  that  the  expected  duration  of  the  employment 
relationship  is  nor  shorter  nor  longer  for  people  with  higher 
employability.  We  did  find  effects  of  employability  on  the  other 
transactional psychological contract obligations subscale, which relates 
191
to  the  ‘narrowness’  of  employees’  job  tasks  and  responsibilities.  As 
expected,  people  with  higher  employability  express  that  they  feel 
obliged to take up  broad responsibility in their jobs as opposed to just 
doing their jobs.  
 
Although  talent  category  and  organizational  tenure  were  originally 
incorporated  in  our  research  model  as  moderators,  we  also  wish  to 
discuss some interesting main effects of both variables on the different 
psychological  contract  obligations.  As  for  the  talent  category  variable, 
we found that high potentials expect to receive more job security from 
their employers, although they do not perceive stronger obligations for 
staying with their current employer themselves. Moreover, we found an 
interaction effect of employability and talent category implying that the 
higher  the  employability  of  high  potentials,  the  less  they  perceive 
themselves  as  obligated  to  stay  with  their  current  employer. 
Furthermore,  they  perceive  a  higher  obligation  from  their  employer  to 
offer  them  job  content  with  broad  responsibilities.  The  above  findings 
imply  that  high  potentials  are,  indeed,  well  aware  of  their  strong 
bargaining  position  relative  to  their  employing  organizations  (Guest, 
2001).  The  importance  of  retention  initiatives  and  career  planning  for 
this  particularly  valuable  group  is  thus  again  highlighted.  On  the  other 
hand,  the  high  potentials  in  our  sample  reported  weaker  perceived 
obligations  in  terms  of  maintaining  their  external  employability 
perspectives.  It  appears  that  high  potentials  feel  relatively  comfortable 
in their position in the organization, so much so that they  expect to be 
able to stay there until possibly something better comes along – but that 
they  do  not  actively  pursue  opportunities  outside  of  the  organization 
themselves.  This  is  a  possible  indication  of  high  potentials’  supposed 
awareness  of  their  organization‐specific  status  and  seems  to  confirm 
earlier  studies  of  high  potential  employees.  Dries  and  Pepermans 
(2008),  in  their  interview  study  of  employees  identified  as  high 
potentials, found that they were unlikely to pursue career opportunities 
outside  of  their  organizations  as  they  believed  their  chances  for  rapid 
advancement  were  higher  within  their  own  organization  due  to  the 
inherent career ‘promises’ encapsulated in the high potential label.  
  
As expected, organizational tenure was found to be positively related to 
perceived obligations of job security and loyalty (both from the side of 
the  employee  and  the  employer).  Similarly,  longer‐tenured  employees 
perceive  weaker  short‐term  employment  obligations  from  the  side  of 
the  employer.  Furthermore,  as  concerns  the  balanced  psychological 
contract  obligations,  long‐tenured  employees  were  found  to  perceive 
less  obligations  to  engage  in  external  employability  maintenance  and 
192
continuous  performance  improvement.  However,  they  do  perceive 
stronger  obligations  in  terms  of  having  broad  responsibilities  in  their 
jobs  (both  from  the  side  of  the  employee  and  the  employer).  These 
findings  confirm  the  expectations  raised  in  the  Introduction  about  the 
relationship between tenure and the desire for career stability (Clarke & 
Patrickson,  2008;  Mathieu  &  Zajac,  1990;  Millward  &  Hopkins,  1998; 
Rousseau,  1989).  They  also  seem  to  confirm  some  of  the  stereotypical 
conceptions  about  long‐tenured  employees,  i.e.  that  they  tend  to 
‘plateau’  in  their  careers  as  a  results  of  their  diminished  interest  in 
learning  and  development  (Hess  &  Jepsen,  2009).  However,  some 
nuance  must  be  applied.  First  of  all,  in  analogy  with  the  interaction 
effect we found for talent category, the higher the employability of long‐
tenured  employees,  the  less  they  perceive  themselves  as  obligated  to 
stay  with  their  current  employer.  Second,  longer‐tenured  employees 
with  high  employability  perceive  stronger  internal  advancement  and 
dynamic performance obligations from the side of the employer. 
 
In  sum,  we  found  no  consistent  support  for  the  thesis  that  more 
employable  employees  perceive  weaker  psychological  contract 
obligations  in  terms  of  job  security  and  organizational  loyalty. 
Furthermore, although higher employability coincides with a perceived 
obligation for more career self‐development (both inside and outside of 
the current employing organization) and the obligation to take up broad 
responsibilities  in  the  job,  highly  employable  employees  also  perceive 
more  obligations  in  terms  of  career  support  from  their  organizations. 
Consequently,  the  proposition  that  the  demise  of  the  organizational 
career  and  decreases  in  job  security  could  somehow  be 
‘counterbalanced’ by developing individuals’ employability seems to be 
overly simplistic (Van Buren, 2003). Finally, although the findings of our 
study  were  inconsistent  with  our  moderation  hypotheses,  we  did  find 
support for the notion of firm‐specific capital shaping the psychological 
contracts of employees. Both for employees identified as high potentials 
and for long‐tenured employees we found that they perceived stronger 
obligations  from  the  side  of  the  employer  to  provide  job  security,  than 
from  their  side  to  stay  with  their  employer,  particularly  under 
conditions  of  high  employability.  These  findings  warrant  attention 
towards the pivotal importance of employees’ bargaining position in the 
organization in shaping the psychological contract (Guest, 2001). 
 
Limitations  
 
The  current  study  faced  a  number  of  limitations.  First  of  all,  the 
respondents  in  the  study  came  from  five  organizations  from  different 
193
sectors and backgrounds. As our sample size was not especially large (n 
= 103), we were not able to take cultural or other contextual factors into 
account  in  our  analyses.  Furthermore,  the  survey  was  administered 
during  a  time  when  the  global  financial  crisis  was  heavily  discussed  in 
the  media,  which  in  all  probability  influenced  the  importance  attached 
by  respondents  to  employment  security.  For  all  of  these  reasons,  we 
must  exert  caution  in  generalizing  the  findings  of  our  study  across 
timeframes and settings.  
 
Furthermore,  the  measures  used  in  this  study  to  operationalize  our 
research model have received some critique in the literature. As regards 
the  Psychological  Contract  Inventory  (PCI)  developed  by  Rousseau 
(2000),  heterogeneities  have  been  found  across  studies  suggesting 
scaling problems when items are aggregated into scale scores (Conway 
&  Briner,  2005;  Porter  et  al.,  1998).  Also,  the  PCI  stresses  obligations, 
whereas  there  is  still  discussion  about  whether  the  psychological 
contract  is  about  obligations,  expectations,  beliefs,  promises,  or 
perceptions  (Arnold,  1996;  Guest,  1998).  In  addition,  the  PCI  departs 
exclusively  from  the  perspective  of  the  employee,  who  is  instructed  to 
indicate  both  his  or  her  own  obligations,  and  the  obligations  made  by 
the  employer.  Several  authors  have  questioned  this  approach,  arguing 
that  it  ‘anthropomorphizes’  the  organization,  and  does  not  provide 
insight  into  which  parties  exactly  are  considered  by  employees  as 
enacting these obligations (Arnold, 1996; Guest, 1998). However, there 
is  no  consensus  to  date  about  what  type  of  psychological  contract 
measurement would be most reliable or valid across studies (Conway & 
Briner,  2005).  We  opted  to  use  the  PCI  as  the  contents  of  its  subscales 
were  particularly  well  suited  for  answering  our  research  questions.  As 
for employability, Fugate and Kinicki (2008) themselves recognize that 
different  conceptualizations  and  operationalizations  exist,  each  with 
their own merit. For instance, the measure developed by  Van Der Heijde 
and Van Der Heijden (2006) adopts a more competency‐based approach 
to  employability,  whereas  the  Fugate  and  Kinicki  (2008)  measure 
focuses on person‐centered employability. Further research needs to be 
done  in  order  to  determine  the  concurrent  validity  of  the  different 
measures, and expose in which research designs which measure is more 
appropriate  and  why.  Another  issue  is  the  fact  that  Fugate  and  Kinicki 
(2008)  describe  five  subscales  in  their  measure  (i.e.  openness  to 
changes  at  work,  work  and  career  resilience,  work  and  career 
proactivity, career motivation, and work identity) whereas we used the 
overall total score of the scale as our employability measure. We chose 
to  use  one  overall  employability  score  as  regression  analyses  were 
already  quite  elaborate  (seeing  as  we  analyzed  the  PCI  at  the  subscale 
194
level),  and  we  were  not  convinced  that  using  the  employability 
subscales would add to the interpretation of our research model.  
 
Another  limitation  lies  in  the  fact  that  our  study  was  cross‐sectional, 
which  may  result  in  common  method  variance.  However,  two  of  the 
variables in our research model (i.e. talent category and organizational 
tenure)  were  factual  instead  of  attitudinal.  Furthermore,  the 
instructions  and  scale  anchors  for  the  PCI  items  and  the  employability 
items  were  quite  different,  which  somewhat  decreases  the  chance  of 
response  bias  (Schwarz,  Rizzuto  &  Schwarz,  2008).  In  order  to  avoid 
common method variance, longitudinal designs or designs with different 
measurement  points  are  more  valid.  However,  as  the  data  collection 
process  for  this  study  was  already  quite  demanding,  such  a  design 
proved unattainable.  
 
Finally,  it  was  impossible  to  fully  debrief  the  respondents  since  they 
were not aware of the talent category variable in the survey. However, 
this  intervention  was  specifically  requested  by  the  participating 
organizations,  who  wanted  to  ensure  discretion  of  their  talent 
management  procedures.  Nonetheless,  it  remains  unclear  to  which 
extent  the  withholding  of  information  about  internal  career 
opportunities by organizations is to be considered unethical (Van Buren, 
2003).   
 
Implications for practice 
 
There appears to be a gap between theory and practice as concerns the 
supposed  changes  in  the  psychological  contract.  Several  empirical 
studies  found  that  while  employers  are  reluctant  to  develop  external 
employability  in  their  employees,  employees’  expectations  with  regard 
to career management and job security continue to be dominated by the 
relational  element  of  the  psychological  contract,  despite  the  fact  that 
organizations  and  industries  have  changed  (Atkinson,  2002;  Baruch, 
2001;  Clarke  &  Patrickson,  2008).  It  appears  that  employees  may  have 
unrealistic  expectations  about  organizations,  and  that  organizations 
from  their  side  are  failing  to  live  up  to  their  employees’  expectations 
(Granrose  &  Baccili,  2006).  Employees  may  need  some  help  in 
developing  more  realistic  perceptions  of  what  type  of  promises 
organizations  are  willing  and  able  to  make  in  today’s  strongly 
competitive  global  economy  (Granrose  &  Bacilli,  2006).  Organizations, 
from their side, should consider the consequences of possibly failing to 
live  up  to  their  employees’  expectations  about  internal  career 
opportunities  and  job  security  (Granrose  &  Bacilli,  2006).  Van  Buren 
195
(2003)  argues  that  organizations  should,  at  least,  provide  more 
transparent information about employment prospects as well as invest 
more  in  the  enhancement  of  employees’  external  employability.  One 
question  raised  in  the  literature,  however,  is  whether  psychological 
contracts cease to be psychological contracts when they are made open 
and explicit (Guest, 1998).  
 
One  particular  issue  that  arose  from  this  study  is  the  importance  of 
expectations  management  in  retaining  ‘core’  employees  such  as  high 
potentials and long‐tenured employees. The motives as to why they do 
not appear to actively pursue opportunities outside of the organization 
seem to be grounded mostly in perceived career promises rather than in 
feelings  of  loyalty  per  se.  Frustrating  these  perceived  promises  might 
then be detrimental for their retention (Dries & Pepermans, 2008).  
 
Implications for research 
 
Several  authors  have  argued  that  careers  researchers,  from  their  side, 
need  to  take  up  their  responsibility,  and  scrutinize  actual  facts  about 
modern‐day careers rather than develop new dogmas to replace the old 
ones  (Dyer  &  Humphries,  2002).  The  discourse  on  the  demise  of  the 
traditional  career  fits  all  too  well  with  what  has  been  called  a  ruthless 
economy (referring to the postmodern rhetoric and the segmentation of 
employees  in  ‘core’  and  ‘expendable’  employees,  described  in  the 
Introduction)  while,  at  the  same  time,  the  role  of  this  economy  in 
dictating  the  shape  of  the  new  career  is  either  obscured  or  glorified 
(Dyer  &  Humphries,  2002;  Lucas  et  al.,  2006).  The  rhetoric  about 
boundaryless  careers  intensifies  effects  of  having  a  two‐tiered 
workforce,  comprising  a  small  top  tier  of  highly  skilled  employees 
whose  bargaining  position  enables  them  to  demand  and  receive  fair 
treatment from employers (‘winner‐take‐all’), and a large second tier of 
employees  whose  skills  are  replaceable  (Van  Buren,  2003).  More 
empirical  work  needs  to  be  done  to  discriminate  conjectural  from 
empirically  supported  ideas  about  postmodern  careers  (Briscoe  & 
Finkelstein, 2008; Guest, 1998). 
 
References 
 
Atkinson,  C.  (2002).  Career  management  and  the  changing  psychological  contract.  Career 
Development International, 7 (1), 14‐23.  
Arnold,  J.F.  (1996).  The  psychological  contract:  A  concept  in  need  of  greater  scrutiny.  European 
Journal of Work and Organizational Psychology, 5 (4), 511‐520.  
Arthur,  M.B.  &  Rousseau,  D.M.  (1996).  The  boundaryless  career:  A  new  employment  principle  for  a 
new organizational era. New York, NY: Oxford University Press.  
Barrington, L. (2002). Despite hard times companies view leadership development
as a priority. Executive Action, 34, The Conference Board, New York.
196
Baruch,  Y.  (2001).  Employability:  A  substitute  for  loyalty?  Human  Resource  Development 
International, 4 (4), 543‐566.  
Baruch,  Y.  (2006).  Career  development  in  organizations  and  beyond:  Balancing  traditional  and 
contemporary viewpoints. Human Resource Management Review, 16, 125‐138.  
Bossuyt, T. & Dries, N. (2008). Talent management en flexibele loopbaanpaden voor de werknemers 
van morgen [Talent management and flexible career tracks for tomorrow's workforce]. In H. 
Calmeyn, K. De Witte & J. Weverbergh (Eds.), Licht op leren 2008: Leren en ontwikkelen in een 
talentgerichte maatschappij [Spotlight on learning 2008: Learning and developing in a talent­
oriented society]. Leuven, Belgium: LannooCampus.  
Briscoe,  J.P.  &  Finkelstein,  L.M.  (2009).  The  “new  career”  and  organizational  commitment:  Do 
boundaryless and protean attitudes make a difference? Career Development  International, 14 
(3), 242‐260.  
Buckingham,  M.  &  Vosburgh,  R.M.  (2001).  The  21st  century  Human  Resources  function:  It's  the 
talent, stupid! Human Resource Planning, 24 (4), 17‐23. 
Clarke, M. & Patrickson, M. (2008). The new covenant of employability. Employee Relations, 30 (2), 
121‐141.  
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis 
for the behavioral sciences. Hillsdale, NJ: Erlbaum. 
Conway, N. & Briner, R. (2005). Understanding psychological contracts at work. Oxford, UK: Oxford 
University Press.  
Cope, F. (1998). Current issues in selecting high potentials. Human Resource Planning, 21 (3), 15‐17. 
Coyle‐Shapiro,  J.A.  &  Neuman,  J.H.  (2004).  The  psychological  contract  and  individual  differences: 
The role of exchange and creditor ideologies. Journal of Vocational Behavior, 64, 150‐164.  
De  Vos,  A.  &  Meganck,  A.  (2009).  What  HR  managers  do  versus  what  employees  value:  Exploring 
both  parties’  views  on  retention  management  from  a  psychological  contract  perspective. 
Personnel Review, 38 (1), 45‐60.  
Dries,  N.  &  Pepermans,  R.  (2008).  "Real"  high  potential  careers:  An  empirical  study  into  the 
perspectives of organizations and high potentials. Personnel Review, 37 (1), 85‐108. 
Dyer,  S.  &  Humphries,  M.  (2002).  Normalising  work‐place  change  through  contemporary  career 
discourse.  In  L.  Morrow,  I.  Verins,  &  E.  Wills  (Eds.),  Mental  Health  and  Work:  Issues  and 
perspectives. Australia: Common Wealth of Australia. 
Fugate,  M.  (2006).  New  perspectives  on  employability.  In  J.  Greenhaus  &  G.  Callanan  (Eds.), 
Encyclopedia of Career Development. Thousand Oaks, CA: Sage.  
Fugate,  M.,  &  Kinicki,  A.J.  (2008).  A  dispositional  approach  to  employability:    Development  of  a 
measure and test of its implications for employee reactions to organizational change. Journal 
of Occupational and Organizational Psychology, 81, 503‐527. 
Granrose,  C.S.  &  Baccili,  P.A.  (2006).  Do  psychological  contracts  include  boundaryless  or  protean 
careers? Career Development International, 11 (2), 163‐182.  
Guest,  D.  (1998).  Is  the  psychological  contract  worth  taking  seriously?  Journal  of  Organizational 
Behavior, 19, 649‐664.  
Guest,  D.  (2001).  Industrial  relations  and  human  resource  management.  In  J.  Storey  (Ed.),  Human 
resource management: A critical text. London, UK: Thompson Learning.  
Guest, D. & MacKenzie Davey, K. (1996). Don’t write off the traditional career. People Management, 
February, 22‐23.  
Hair, J.F., Anderson, R.E., Tatham, R.L. & Black, W.C. (1998). Multivariate data analysis with readings. 
Englewood Cliffs, NJ: Prentice Hall.  
Hess,  D.M.   &  Jepsen,  N.  (2009).  Career  stage  and  generational  differences  in  psychological 
contracts. Career Development International, 14 (3), 261‐283.  
Lucas,  K.,  Liu,  M.,  &  Buzzanell,  P.  M.  (2006).  No  limits  careers:  A  critical  examination  of  career 
discourse in the U.S. and  China. In M. Orbe, B. J. Allen, & L. A. Flores (Eds.), International and 
intercultural communication annual 28. Thousand Oaks, CA: Sage.  
Martin, G. & Hetrick, S. (2006). Corporate reputations, branding and people management: a strategic 
approach to HR. Oxford, UK: Elsevier Ltd.  
Mathieu,  J.E.  &  Zajac,  D.M.  (1990).  A  review  and  meta‐analysis  of  the  antecedents,  correlates,  and 
consequences of organizational commitment. Psychological Bulletin, 108, 171‐194.  
Millward,  L.J.  &  Hopkins,  L.J.  (1998).  Psychological  contracts,  organizational  and  job  commitment. 
Journal of Applied Social Psychology, 28, 1530‐1556.  
Pallant, J. (2001). SPSS Survival Manual. Philadelphia, PA: Open University Press.  
Parent,  D.  (2002).  Matching,  human  capital,  and  the  covariance  structure  of  earnings.  Labour 
Economics, 9, 375‐404. 
197
Porter,  L.W.,  Pearce,  J.L.,  Tripoli,  A.  &  Lewis,  K.  (1998).  Differential  perceptions  of  employers' 
inducements: Implications for psychological contracts. Journal of Organizational Behavior, 19, 
769‐782.  
Robinson,  S.L.,  Kraatz,  M.S.  &  Rousseau,  D.M.  (1994).  Changing  obligations  and  the  psychological 
contract : A longitudinal study. Academy of Management Journal, 37, 137‐152.  
Rousseau,  D.M.  (1989).  Psychological  and  implied  contracts  in  organizations.  Employee 
Responsibilities and Rights Journal, 2, 121‐139.  
Rousseau,  D.M.  (1995).  Psychological  contracts  in  organizations :  Understanding  written  and 
unwritten agreements. London, UK: Sage.   
Rousseau,  D.M.  (2000).  Psychological  Contract  Inventory  Technical  Report.  Carnegie  Mellon 
University, Pittsburgh, PA.  
Rousseau,  D.M.  (2001).  Schema,  promise  and  mutuality:  The  building  blocks  of  the  psychological 
contract. Journal of Occupational and Organizational Psychology, 74, 511‐541.  
Rousseau,  D.M.  &  McLean  Parks,  J.M.  (1993).  The  contracts  of  individuals  and  organizations. 
Research in Organizational Behavior, 15, 1‐43.  
Schwarz, A., Rizzuto, T., & Schwarz, C. (2008). Urban legend of common method bias: nine common 
errors  and  their  impacts.  Paper  presented  at  the  Hawaii  International  Conference  on  System 
Sciences, 7‐10 January, Manoa, HI. 
Sullivan,  S.  (1999).  The  changing  nature  of  careers:  A  review  and  research  agenda.  Journal  of 
Management, 25 (3), 45‐84.  
Tulgan, B. (2001). Winning the talent wars. Employment Relations Today, 23 (1), 104.  
Van  Buren,  H.J.  III  (2003).  Boundaryless  careers  and  employability  obligations.  Business  Ethics 
Quarterly, 13 (2), 131‐149.  
Van Der Heijde, C.M. & Van Der Heijden B.I.J.M. (2006). A competence‐based and multidimensional 
operationalisation  and  measurement  of  employability.  Human  Resource  Management,  45, 
449‐476.   
Verbruggen,  M.,  Sels,  L.  &  Forrier,  A.  (2007).  Unraveling  the  relationship  between  organizational 
career  management  and  the  need  for  external  career  counselling.  Journal  of  Vocational 
Behavior, 71 (1), 69 ‐ 83. 
 
 

 
198
199
Paper 6. 
 

Effects of the high potential label on 
performance, career success and 
commitment: A matter of communication? 4 
 
 
This  large‐scale  quantitative  study  examined  the  effects  of  being  identified  as  a  high 
potential (or not) on a number of career variables, taking into account the openness of 
communication  that  is  adopted  by  organizations  with  respect  to  their  talent 
management  procedures.  A  total  of  711  respondents  (250  high  potentials  versus  461 
non‐high potentials) from 12 organizations participated in the online survey. Half of the 
participating  organizations  had  implemented,  at  the  time  of  survey  administration,  a 
policy  of  openly  disclosing  which  employees  are  on  their  high  potential  list;  the  other 
half  had  a  policy  of  keeping  this  information  hidden  from  their  employees.  We  found 
positive  main  effects  of  being  identified  as  a  high  potential,  and  of  organizational 
communication  openness,  on  job  performance,  career  success,  and  commitment. 
Furthermore,  we  found  interaction  effects  of  being  identified  as  a  high  potential  with 
communication openness on salary, satisfaction with performance, and satisfaction with 
job  security.  Overall,  results  indicate  that  adopting  an  open  communication  strategy 
about  talent  management  generates  the  most  desirable  outcomes.  However,  nuance 
must  be  applied  in  interpreting  and  generalizing  these  findings.  Implications  for 
research and practice are spelled out.   

                                                            
4 An earlier version of this paper received the Arnon E. Reichers Best Student Paper 

Award at the 2009 Academy Of Management conference.  
 
 

 
200

Figure VII. Topicality of Paper 6.
201
Paper 6. 
 
Effects of the high potential label on performance, career success 
and commitment: A matter of communication? 
 
Despite over a decade of debate and hype about the ‘war for talent’ as a 
strategic  business  challenge  and  a  critical  factor  of  influence  for 
corporate  performance  (Collings  &  Mellahi,  2009;  Michaels,  Handfield‐
Jones  &  Axelrod,  2001),  the  subject  of  talent  management  is  grossly 
under‐researched  (Boudreau  &  Ramstad,  2006;  Lewis  &  Heckman, 
2006). This is particularly surprising considering the general consensus 
among HR (and general business) practitioners about the importance of 
talent  management,  even  more  so  in  the  context  of  the  current 
recessionary climate (Collings & Mellahi, 2009).  
 
To  date,  there  is  not  even  a  widespread  or  consensual  definition  of 
talent management (Aston & Morton, 2005). Lewis and Heckman (2006) 
concluded from their review of the literature that there is an unsettling 
degree  of  ambiguity  regarding  the  definition,  scope  and  goals  of  talent 
management. They identified three distinct streams of discourse: talent 
management  as  a  collection  of  human  resource  practices  (under  a 
different label); talent management as processes supporting the flow of 
employees  through  positions  within  an  organization;  and  talent 
management  as  focusing  on  talent  generically,  regardless  of 
organizational  or  functional  boundaries.  Collings  and  Mellahi  (2009) 
added  a  fourth  perspective,  stating  that  identifying  key  positions  in 
organizations should be the starting point of talent management, rather 
than talented individuals per se.  
 
Nonetheless, even the latter authors recognize the pivotal importance of 
identifying  and  developing  so‐called  ‘high  potentials’  for  any  talent 
management program. High potentials are “those individuals within an 
organization  who  are  recognized,  at  that  point  in  time,  as  the 
organization’s  likely  future  leaders”  (Cope,  1998,  p.  15).  Although,  in 
practice,  talent  management  is  all  too  often  equated  with  (or  rather, 
limited  to)  the  attraction,  selection,  development  and  retention  of  high 
potentials  (Bossuyt  &  Dries,  2008;  Jerusalim  &  Hausdorf,  2007), 
management  scholars  do  seem  to  agree  that  investing  in  this  group  is 
crucial  to  ensure  continuity  of  management  resources  (Hall  &  Seibert, 
1992)  and  boost  the  retention,  motivation  and  performance  of  this 
specific  group  (Jerusalim  &  Hausdorf,  2007).  Collings  and  Mellahi 
(2009)  even  argue  that  the  focus  of  talent  management  should  be 
mostly on high potentials and not on all employees in the organization, 
202
as  such  an  approach  “would  facilitate  a  more  deliberate  utilization  of 
organizational resources” (p. 3).  
 
Although  several  authors  have  published  on  the  topic  of  talent 
management  (e.g.  Baruch  &  Peiperl,  1997;  Buckingham  &  Vosburgh, 
2001;  Fields,  2002;  McCall,  1998;  Segalla,  Rouziès  &  Flory,  2001; 
Spreitzer,  McCall  &  Mahoney,  1997),  only  a  small  number  of  empirical 
studies have come out using data collected from employees identified as 
high potentials themselves (e.g. Dries & Pepermans, 2007, 2008; Viney, 
Adamson  &  Doherty,  1997).  As  a  consequence,  the  literature  on  high 
potentials is largely based on conjecture, rather than empirical evidence 
(Pepermans,  Vloeberghs  &  Perkisas,  2003).  HRM  practitioners  and 
scholars alike indicate that there is an urgent need for more research on 
talent management and high potentials for validation and benchmarking 
purposes (Bossuyt & Dries, 2008; Dries & Pepermans, 2008). 
 
Research model 
 
The current paper seeks to respond to the research gap outlined above. 
Specifically,  it  will  examine  the  effects  of  being  identified  as  a  high 
potential  (or  not)  on  a  number  of  career  variables,  taking  into  account 
the  openness  of  communication  that  is  adopted  by  organizations  with 
respect  to  their  talent  management  procedures.  The  question  of 
whether  or  not  to  disclose  high  potential  status  –  and  relatedly,  non‐
high  potential  status  –  to  employees  is  seen  as  pivotal  to  the 
implementation  of  any  talent  management  program,  and  a  quite 
complex  question  at  that  (e.g.  Burke,  1997;  Cope,  1998).  Figure  1 
provides an overview of the hypothesized relationships in the study. As 
is  shown  in  the  figure,  we  will  examine  both  main  effects  of  talent 
category  on  career  variables  (e.g.  “are  high  potentials,  on  average, 
intrinsically  more  committed  to  their  careers?”)  and  moderated  effects 
(e.g. “are high potentials, on average, more committed to their careers, 
only  when  or  especially  when  their  high  potential  status  is 
communicated  to  them?”).  Note  that  Figure  1  does  not  imply  strict 
causality  for  the  hypothesized  effects,  since  theoretical  arguments  can 
often  be  raised  for  different  or  opposite  causal  inferences  (e.g.  “high 
performance  is  a  prerequisite  for  being  identified  as  a  high  potential” 
and  “being  awarded  the  high  potential  label  motivates  employees  to 
such a degree, that it affects their performance in a positive way”).  
 
 
 
 
 
 
 
 
 
 
 
 
 
203

 
 
 
 
 
 
 
Figure 1. Theoretical model 
 
 
 
204
Individual career variables related to high potential identification 
 
As for main effects, we expect being identified as a high potential or not 
to  be  significantly  related  to  three  career  variables:  job  performance, 
career success, and commitment.  
 
First  of  all,  we  expect  to  find  significant  differences  between  high 
potentials  and  non‐high  potentials  in  terms  of  job  performance. 
Although authors in the talent management domain seem to agree that 
identification  of  high  potentials  should  be  based  on  both  person‐
centered  and  situation‐centered  variables  as  well  as  their  interactions 
(Hollenbeck,  McCall  &  Silzer,  2006),  amongst  all  possible  criteria  for 
identifying  high  potentials,  an  above‐average  job  performance  is 
unanimously  considered  the  most  decisive  factor  (Dries  &  Pepermans, 
2008).  An  employee  who  does  not  consistently  receive  ‘exceeds 
expectations’ performance ratings is very unlikely to be considered for 
entering  any  high  potential  program  (Cope,  1998;  Pepermans  et  al., 
2003). Furthermore, for most companies, evaluating potential based on 
performance  ratings  is  less  politically  charged  (and  less  controversial) 
than  assessing  candidates  against  a  list  of  competencies  (Derr,  Jones  & 
Toomey, 1988).  
 
A  note  of  caution  must  be  applied  however,  since  it  is  crucial  that 
organizations  clearly  distinguish  high  performers  from  high  potentials. 
Although  an  excellent  performance  is  and  should  be  a  prerequisite  for 
being  identified  as  a  high  potential,  organizations  must  beware  of  halo 
effects in their appraisal processes, which cause assessors to wrongfully 
align potentiality ratings with performance ratings (Fields, 2002).  
 
H1.  Employees  identified  as  high  potentials  score  significantly  higher  on 
indicators of job performance than employees who are not identified as high 
potentials.  
 
Second,  we  expect  high  potentials  and  non‐high  potentials  to  differ  in 
terms  of  career  success  as  well.  As  high  potential  programs  are 
especially  designed  to  ensure  management  continuity  by  developing 
and retaining employees with leadership potential (Dries & Pepermans, 
2008;  Jerusalim  &  Hausdorf,  2007),  we  expect  that  ‘traditional’  career 
opportunities,  such  as  upward  promotions  and  salary  increases,  are 
awarded  more  to  these  employees  than  to  others.  Put  differently,  we 
expect  high  potentials  to  achieve  more  objective  career  success. 
Furthermore,  successes  achieved  early  in  the  career  tend  to  be  self‐
enforcing in that supervisors tend to expect more of, and consequently, 
are more supportive of, employees that have produced more in the past 
205
(Eden,  1984).  Therefore,  it  is  sometimes  said  that  being  awarded  the 
high  potential  label  engages  an  employee  in  a  self‐fulfilling  cycle  of 
success (Dries & Pepermans, 2008; Hall & Chandler, 2005).  
 
The  current  study  will  also  study  possible  effects  on  subjective  career 
success.  Research  has  indicated  that  objective  and  subjective  career 
success measures tend to intercorrelate, but not very highly so (Arthur, 
Khapova & Wilderom, 2005). All too often, it is assumed that subjective 
career  success  is  a  direct  function  of  objective  career  success  (e.g. 
“people  at  top  positions  who  earn  high  salaries  must  be  very  satisfied 
with their careers”), where the reality is most likely much more complex 
than that (Hall & Chandler, 2005). For instance, ex‐post rationalizations 
may take place causing  subjective  evaluations of one’s career to be (or 
remain)  positive  even  when  it  is  not  successful  in  terms  of  objective 
indicators (Nicholson & de Waal‐Andrews, 2005). Such mechanisms can 
be seen as a form of ‘cognitive dissonance reduction’ where individuals 
adjust  their  beliefs  or  convictions  in  order  to  maintain  a  positive  self‐
concept  (Festinger,  1957).  In  the  current  study,  we  expect  high 
potentials  to  score  higher  than  non‐high  potentials  in  terms  of 
subjective career success based on both of the above arguments. First of 
all,  as  subjective  career  success  tends  to  be  at  least  in  part  related  to 
objective  career  success,  we  expect  that  high  potentials  will  be  more 
satisfied with their careers based on the fact that they, on average, are 
more  likely  to  achieve  objective  success.  Second,  as  interview  research 
in high potential populations has demonstrated (e.g. Dries & Pepermans, 
2008;  Viney  et  al.,  1997),  most  employees  identified  as  high  potentials 
consider  their  career  a  priority  in  their  lives,  and  are  willing  to  make 
sacrifices (mostly in terms of work‐life balance) in order to succeed. One 
could argue then, that in order to ‘keep going’, they need to feel satisfied 
with what they are doing, so as to rationalize all the personal effort they 
have put in (Nicholson & de Waal‐Andrews, 2005; Sturges, 1999).  
 
H2.  Employees  identified  as  high  potentials  score  significantly  higher  on 
indicators  of  career  success  than  employees  who  are  not  identified  as  high 
potentials. 
 
Finally,  we  also  expect  to  see  significant  differences  in  terms  of 
commitment. Earlier research has indicated that high potentials tend to 
be  very  committed  to  their  careers  –  sometimes  so  much  so  that  they 
end  up  having  trouble  relating  to  colleagues  on  a  personal  level  –  and 
that they are prone to overachieving and burn‐out (Dries & Pepermans, 
2007;  Dubouloy,  2004).  Furthermore,  organizations  are  generally  very 
concerned  about  the  return  on  investment  of  their  talent  management 
procedures.  Therefore,  it  is  not  unlikely  that  they  would  probe  the 
206
organizational  commitment  of  ‘potential  high  potential’  candidates 
during the identification process (Dries & Pepermans, 2008).  
 
H3.  Employees  identified  as  high  potentials  score  significantly  higher  on 
indicators  of  commitment  than  employees  who  are  not  identified  as  high 
potentials. 
 
Open communication or strategic ambiguity? 
 
As  mentioned  earlier,  a  major  concern  for  organizations  engaging  in 
talent  management  is  whether  or  not  to  communicate  openly  about 
their procedures regarding the identification of potential (Burke, 1997; 
Cope,  1998).  Most  firms  report  that  they  prefer  to  sustain  a  secretive 
status for high potentials (Burke, 1997). An older study (Fresina, 1987) 
found that 78 percent of the firms participating in their interview study 
did  not  inform  high  potentials  of  their  status.  The  same  interviewees, 
however,  estimated  that  high  potential  status  became  apparent 
informally 90 percent of the time. Recent publications have pleaded for 
open  organizational  communication  (Ng  &  Butts,  2009;  Van  Buren, 
2003;  Vandenberg,  Richardson  &  Eastman,  1999).  Furthermore,  some 
literature states that many contemporary organizations are shifting to a 
more  open  communication  strategy  about  talent  management  (e.g. 
Göbel‐Kobialka,  1998;  Snipes,  2005).  Nonetheless,  recent  empirical 
studies  have  found  that  closed  communication  (i.e.  not  disclosing  high 
potential  status  to  employees  and  only  sharing  information  about 
organizational talent management procedures with ‘insiders’ in HR and 
management)  is  still  the  norm  (e.g.  Dries  &  Pepermans,  2007,  2008). 
The  possible  advantages  of  ‘strategic  ambiguity’  in  terms  of 
communication  were  already  spelled  out  over  two  decades  ago  by 
Eisenberg  (1984),  who  stated  that  although  open  communication  is 
generally  the  most  ethical  approach,  it  is  not  always  strategically 
interesting,  especially  when  privileged  positions  in  the  organizations 
need to be preserved.  
 
Organizations  typically  are  afraid  to  communicate  openly,  as  they  fear 
that  non‐high  potentials  will  get  demotivated  and  leave  (whereas  they 
can be valuable contributors), while simultaneously, the high potentials 
will  develop  the  so‐called  ‘crown  prince  syndrome’  and  rest  on  their 
laurels  (Göbel‐Kobialka,  1998).  Furthermore,  openly  disclosing  high 
potential status may cause self‐fulfilling prophecy effects, so that it is no 
longer clear whether the high potential identification process is valid, as 
positive  results  could  be  attributable  both  to  the  high  potentials’ 
intrinsic talent, and to the preferential treatment they are given (Dries & 
Pepermans, 2008; Burke, 1997). 
207
However,  the  effects  of  open  or  closed  communication  about  high 
potential status have  never been empirically examined. Our second set 
of  hypotheses  relates  to  possible  moderation  effects  of  open  versus 
closed  communication.  Whereas  our  first  set  of  hypotheses  examines 
main effects of being a high potential or not, the moderation hypotheses 
allow us to address the following question: “to what extent does an open 
communication further intensify these main effects?”.  
 
For  instance,  it  is  conceivable  that  high  potentials  who  are  aware  of 
their status achieve a higher performance level as a result of the positive 
feedback  encapsulated  in  the  label  itself.  Research  on  the  Pygmalion 
effect has repeatedly demonstrated that high expectations conveyed by 
a credible, authoritative source motivate employees to do even better in 
the future. This effect is expected to be self‐perpetuating; once set upon 
a high performance track by the positive leadership of a supervisor with 
high  expectations,  subordinates  have  been  found  to  sustain  high 
performance on their own (Eden, 1984; Goddard, 1985; Kierein & Gold, 
2000).  We  thus  expect  that  differences  in  performance  between  high 
potentials and non‐high potentials will be more pronounced when they 
are aware of their respective status.  
 
H4.1.  Under  conditions  of  open  communication  about  organizational  talent 
categorization policies, the differences between high potentials and non‐high 
potentials in terms of job performance are more pronounced.  
 
Moderation  effects  can  also  be  hypothesized  for  career  success  and 
commitment. As for career success, one could argue that differentiating 
between  employees  in  terms  of  promotion  opportunities  and  salary 
would be easier to implement under conditions of open communication. 
Under conditions of closed communication, such differentiation is more 
likely  to  be  perceived  as  intransparent  and  unjust  by  employees  and 
employee  unions  (Burke,  1997).  Furthermore,  knowledge  of  being 
awarded the high potential label may enhance subjective career success 
as  positive  input  about  career  prospects  tends  to  reinforce  employees’ 
career  identity  and  satisfaction  through  a  so‐called  ‘psychological 
success cycle’ (Hall & Chandler, 2005).  
 
H4.2.  Under  conditions  of  open  communication  about  organizational  talent 
categorization policies, the differences between high potentials and non‐high 
potentials in terms of career success are more pronounced. 
 
As for career and organizational commitment, social exchange processes 
might  be  more  pronounced  under  conditions  of  open  communication. 
By  investing  heavily  in  the  organization‐specific  capital  of  their  most 
talented  employees,  organizations  render  them  ‘imperfectly  mobile’,  in 
208
that  the  transaction  costs  (i.e.  the  costs  associated  with  an  inter‐
organizational  career  move)  for  both  employer  and  employee  increase 
(Peteraf,  1993;  Zuckerman,  Kim,  Ukanwa  &  von  Rittmann,  2003).  For 
organizations,  losing  an  employee  in  which  they  have  invested  a  great 
deal  denotes  suboptimal  return  on  investment  (Peteraf,  1993).  For  the 
invested‐in  employees,  changing  employers  is  not  desirable  as  their 
talent  status  is  organization‐specific,  and  consequently,  the  odds  of 
having  a  successful  career  are  higher  if  they  stay  (Zuckerman  et  al., 
2003). A qualitative study by Dries and Pepermans (2008) revealed that 
high  potentials  who  are  informed  of  their  status,  are  well  aware  that 
their high potential label is organization‐specific, and that receiving the 
same  career  prospects  elsewhere  is  very  uncertain.  Social  exchange 
processes  further  enhance  these  dynamics:  organizations  offer  long‐
term  career  perspectives  to  employees  identified  as  high  potentials  in 
turn for their efforts and loyalty (Shore & Barksdale, 1998). 
 
H4.3.  Under  conditions  of  open  communication  about  organizational  talent 
categorization policies, the differences between high potentials and non‐high 
potentials in terms of commitment are more pronounced. 
 
Method 
 
Research context 
 
Data for this study were collected between January and April 2008 (i.e. 
before  the  start  of  the  global  financial  crisis,  which  is  an  important 
contextual  side  note),  from  the  Belgian  divisions  of  twelve  large, 
internationally  active  organizations,  as  part  of  a  broader  research 
initiative  addressing  current  issues  in  talent  management.  A  large 
number  of  organizations  known  to  engage  in  talent  management  were 
initially addressed with the question to participate. However, in order to 
reduce  heterogeneity  of  the  sample,  participating  organizations  had  to 
meet certain inclusion standards. First of all, participating organizations 
had to be able to produce two separate lists of employees according to 
the  study’s  definitions  of  high  potentials  (i.e.  employees  identified  as 
potential  future  leaders  within  the  organization),  and  non‐high 
potentials,  which  we  will  from  this  point  on  refer  to  as  ‘average 
performers’  (i.e.  employees  who  are  currently  not  identified  as  high 
potentials,  but  considered  adequate  performers  and  contributors).  We 
deliberately  chose  ‘average  performers’  for  the  non‐high  potential 
sample,  as  we  wanted  to  filter  out  the  possibility  of  organizations 
selecting  their  poorest  performers  for  this  subsample,  which  would 
exaggerate  the  differences  between  high  potentials  and  non‐high 
potentials.  
209
 
Although we did allow for variation in terminology (some organizations, 
for  instance,  preferred  the  term  ‘future  leaders’  over  ‘high  potentials’), 
differences  in  definition  were  not  allowed.  The  resulting  sample  of 
organizations  was  thus  homogeneous  in  terms  of  their  conceptions  of 
high  potentials  and  average  performers.  However,  there  were  some 
heterogeneities  in  terms  of  organizational  and  high  potential  program 
characteristics, which are shown in Table 1.  
 
Some  observations  can  be  made.  Most  of  the  organizations  included  in 
the  sample  are  hierarchically  structured  and  turnover  rates  are  highly 
variable, with numbers ranging from 1 per cent to 17 per cent on a year 
basis. The same goes for the percentage of employees identified as high 
potentials  (between  less  than  1  and  20  per  cent  –  interestingly, 
organizations  who  do  not  communicate  openly  about  their  high 
potential  programs  tend  to  identify  a  larger  proportion  of  their 
workforce as high potentials). Since the process of finding and screening 
organizations  willing  and  apt  to  participate  in  the  study  was  already 
quite rigorous, excluding this type of variability proved impossible.  
 
Procedure 
 
Each  participating  organization  assigned  a  contact  person  within  the 
human  resource  (HR)  department,  who  would  be  responsible  for 
distributing information about the online survey to employees from the 
different talent categories. The HR contact persons and the researchers 
worked  closely  together  in  putting  together  the  samples  per 
organization,  thus  ensuring  that  the  high  potential  and  average 
performer subsamples per organization were counterbalanced as much 
as  possible  in  terms  of  age,  gender,  type  of  work  and  functional  level. 
Potential  respondents  received  an  e‐mail  containing  a  URL  to  the 
survey. However, the URL was different for high potentials and average 
performers, so that data from the different talent categories received a 
different code in the database. The content of the surveys, however, was 
identical for both talent categories. This procedure ensured respondent 
anonymity  since  no  names  or  contact  data  were  disclosed  outside  of 
their  organizations.  Furthermore,  as  all  respondents  received  identical 
instructions and surveys, no information about talent categorization by 
the  organization  was  revealed  to  employees.  Such  an  intervention 
proved  crucial  in  order  to  convince  organizations  to  participate  in  the 
study,  considering  the  fact  that  the  half  of  the  participating 
organizations  did  not  communicate  openly  to  employees  about  their 
talent management procedures (see Table 1).  
210
 
Participation  in  the  survey  was  voluntary.  Potential  respondents  were 
stimulated to partake through the inclusion of automatically generated 
feedback reports in the survey. The potential utility of this feedback for 
personal  and  career  development  was  briefly  addressed.  Each  HR 
contact  person  sent  out  at  least  two  follow‐up  e‐mails  to  potential 
respondents  within  their  organization,  encouraging  them  to  partake  in 
the survey. Overall, as indicated in Table 1, response rates were rather 
high (between 27% and 100% in the different organizations).  
 
The  research  design  explicitly  aimed  to  minimize  possible  common 
method variance. First of all, several of the variables in the study were 
factual  rather  than  attitudinal  (i.e.  talent  category,  organizational 
communication  openness,  and  objective  career  success).  Second,  the 
online  survey  was  administered  in  two  blocks,  separated  by  a  few 
weeks. Performance and subjective career success were surveyed in one 
block, commitment and objective career success in the other. The order 
in  which  the  blocks  were  to  be  filled  out  was  random  for  each 
respondent.  
 
Sample demographics 
 
A  total  of  711  respondents  (250  high  potentials  and  461  average 
performers)  completed  the  online  survey.  Of  the  high  potentials,  73 
percent were men; of the average performers, 68 percent. 47 percent of 
high  potentials  were  between  36  and  45  years  old  (compared  to  35 
percent  of  average  performers);  43  percent  were  between  26  and  35 
(compared  to  40  percent  of  average  performers);  7  percent  were  over 
45 (compared to 18 percent of average performers) and 3 percent were 
under  26  (compared  to  7  percent  of  high  performers).  The  majority  of 
high  potentials  had  completed  a  university  education  (52%)  or  a  post‐
university  education  (e.g.  an  MBA  or  a  PhD)  (26%),  whereas  of  the 
average  performers,  33%  had  a  university  degree  and  15%  a  post‐
university  degree.  Most  respondents  held  a  managerial  position  at  the 
time  of  survey  administration  (74%  of  the  high  potentials  and  58%  of 
the  average  performers).  As  for  work  experience,  high  potentials 
reported  an  average  of  12.56  years  (sd  =  6.23),  of  which  7.67  years 
within the current organization (sd = 5.35); for average performers this 
was 14.48 years (sd = 8.25) and 8.97 years (sd = 7.67). 
Table 1. Descriptives for the 12 participating organizations  
 
Organizational characteristics  1   2  3  4  5  6  7 
Sector  financial  financial  energy  consulting  ICT  production  production 
Organization size (global)    85,000    65,000    15,000    37,000      82,000  15,000  124,000 
Organizational structure  hierarchic  hierarchic  hierarchic  hierarchic  network  network  hierarchic 
Turnover per year (national, in %)  5  2           3  ‐  17  4  ‐ 
High potential program characteristics               
Communication   open  open  open  open  open  open  closed 
High potential ratio (in %)  <1  3  3  5     6  18  3 
Response rates (in %)               
High potentials  39  63  49  52  45  43  57 
Average performers  68  65  46  40  35  35  39 
 
Organizational characteristics  8  9  10  11  12  Note. Organizations 4 
211

and  7  did  not  provide 


 
Sector  public consulting ICT consulting ICT 
Organization size (global)          600        120    80,000    16,000       1,500  numbers  for  national 
Organizational structure  hierarchic flat flat flat hierarchic  turnover per year. 
Turnover per year (national, in %)  1 15 12 5   9 
High potential program characteristics   
Communication   closed closed closed closed closed 
High potential ratio (in %)  10 10 20 20 20 
Response rates (in %)   
High potentials  63 100 27 40 60 
Average performers  86 53 31 47 87 
212
Measures 
 
English  versions of  the scales in the  survey were translated into Dutch 
through  translation  ‐  back  translation.  Unless  otherwise  indicated,  all 
measures used seven‐point Likert response scales ranging from strongly 
disagree  (1)  to  strongly  agree  (7).  Scale  scores  were  created  by 
averaging  item  responses  per  scale.  Coefficient  alphas  (with  respect  to 
internal  consistency)  were  added  on  the  main  diagonal  of  the 
correlation matrix in Table 2.  
 
Job  performance.  We  operationalized  this  variable  as  both 
supervisor‐rated  and  self‐rated  performance.  Supervisor‐rated 
performance  was  assessed  by  the  item  “Last  time  I  received  a 
performance rating, the overall appraisal of my performance was…”. In 
addition,  we  also  measured  self‐rated  performance  using  the  item  “My 
own personal appraisal of my performance during that period is…”. Both 
items  required  a  response  on  a  five‐point  scale  ranging  from  ‘not 
meeting  expectations’  (1)  to  ‘substantially  exceeding  expectations’  (5) 
(Fields, 2002). 
 
Career  success.  Both  objective  and  subjective  career  success  were 
measured.  Indicators  of  objective  career  success  measured  were  the 
number  of  promotions  received  within  the  current  organization;  the 
relative  speed  in  which  these  promotions  occurred  (i.e.  the  number  of 
promotions  received  divided  by  the  number  of  years  working  for  the 
organization); current net month salary; and net month salary increase 
over the course of working for the current organization. All items were 
open‐ended.  We  chose  these  variables  as  they  are  most  often  used  as 
operationalizations of objective career success in the literature (Ng, Eby, 
Sorensen  &  Feldman  2005)  –  however,  we  did  add  an  ‘intra‐
organizational’  aspect  to  these  variables  as  the  high  potential  label  is 
also organization‐specific.  
 
Although  the  career  satisfaction  scale  by  Greenhaus,  Parasuraman  and 
Wormley  (1990)  is  frequently  used  to  measure  subjective  career 
success,  several  authors  have  raised  questions  about  its 
unidimensionality, as subjective career success is believed to encompass 
multiple  dimensions  (e.g.  Heslin,  2005).  Consequently,  we  used  the 
multidimensional scale of subjective  career success recently developed 
by  Dries,  Van  Loo  &  Pepermans  (2009).  The  scale  contains  ten 
subscales:  performance  (i.e.  success  in  terms  of  attaining  verifiable 
results  and  meeting  set  goals);  advancement  (i.e.  success  in  terms  of 
progressing  and  growing,  both  in  terms  of  level  and  experience); 
213
personal  development  (i.e.  success  in  terms  of  reaching  one’s  full 
potential  through  self‐management  of  challenges  and  learning 
experiences); creative outlet (i.e. success in terms of creating something 
innovative and extraordinary); financial security (i.e. success in terms of 
being  able  to  meet  one’s  financial  needs);  job  security  (i.e.  success  in 
terms  of  being  able  to  meet  one’s  employment  needs);  work‐life 
dynamics  (i.e.  success  in  terms  of  achieving  personal  satisfaction  and 
happiness, both in the family and in the work domain); recognition (i.e. 
success  in  terms  of  being  adequately  rewarded  and  appreciated  for 
one’s  efforts  and  talents);  inter‐personal  cooperation  (i.e.  success  in 
terms of working well together with peers, superiors, subordinates and 
clients);  and  contribution  (i.e.  success  in  terms  of  contributing  to  the 
collective, such as the organization or general society, as an individual). 
A  sample  item  is  “I  am  able  to  express  myself  through  work  and 
accomplishment”  (self‐development).  Coefficient  alphas  (see  Table  3) 
were  .60  or  above  for  all  subscales,  indicating  satisfactory  internal 
consistency (Hair, Anderson, Tatham & Black, 1998). In order to further 
assess  the  internal  structure  of  the  subjective  career  success  scale,  we 
performed  a  confirmatory  factor  analysis  (CFA)  in  LISREL  8.80  using 
robust  maximum  likelihood  estimation  due  to  violations  of  the 
multivariate normality assumption. Goodness of fit was evaluated using 
a  combinatorial  rule  basing  on  the  Satorra‐Bentler  χ²  statistic,  the 
comparative fit index (CFI) of which values of .95 and less indicate good 
fit,  the  root  mean  square  error  of  approximation  (RMSEA)  of  which 
values  of  .06  to  .08  indicate  a  good  fit  and  the  standardized  root  mean 
square residual (SRMR) of which values of .08 and less indicate a good 
fit (as suggested by Hu & Bentler, 1999). The fit indices suggest that the 
hypothesized  measurement  model  showed  an  acceptable  fit  with  the 
empirical  data  (χ²SB  (751)  =  4879.41,  p  <  .01;  CFI  =.95;  RMSEA  =.08; 
SRMR =.08). 
 
Commitment. In order to operationalize commitment, we used three 
separate  scales:  career  commitment,  organizational  commitment  and 
inter‐organizational  mobility  preference.  Career  commitment  was 
measured using the 21‐item scale developed by Day and Allen (2004). A 
sample item is “I have volunteered for important assignments with the 
intent  of  helping  to  further  my  advancement  possibilities”.  In  order  to 
assess  organizational  commitment,  we  used  the  scale  by  Rottinghaus, 
Day  and  Borgen  (2005).  The  scale  contains  15  items,  e.g.  “I  am 
extremely  glad  that  I  chose  this  organization  to  work  for  over  others  I 
was  considering  at  the  time  I  joined”.  Finally,  inter‐organizational 
mobility  preference  (i.e.  the  degree  to  which  inter‐organizational 
mobility  is  considered  desirable)  was  measured  using  one  of  two 
214
subscales,  suggested  by  Briscoe,  Hall  &  Frautschy  DeMuth  (2006)  in 
their measure for measuring boundaryless career attitudes (we did not 
use  the  other  subscale,  boundaryless  mindset,  as  it  is  less  relevant  for 
our  research  goals  in  this  study).  The  inter‐organizational  mobility 
preference scale is made up of five items, one example being “I like the 
predictability  that  comes  with  working  continuously  for  the  same 
organization”.  Note  that,  although  the  name  of  the  scale  implies  that 
higher  scores  indicate  higher  inter‐organizational  mobility  preference, 
all  five  items  are  worded  in  such  a  way  (also  in  the  original  version  of 
the  scale),  that  they  are  to  be  interpreted  analogously  to  the 
commitment  scales,  i.e.  a  higher  score  means lower  intention  to  search 
employment elsewhere.  
 
Analysis 
 
Differences  between  high  potentials  and  average  performers  were 
analyzed by means of three MANOVA models, i.e. one model with the job 
performance  variables  as  the  dependents,  one  model  with  the  career 
success  variables  as  the  dependents,  and  one  model  with  the 
commitment  variables  as  the  dependents.  In  all  three  models,  talent 
category  and  communication  about  talent  categorization  were  the 
independent  variables.  We  tested  all  socio‐demographic  variables  (i.e. 
gender, age, educational level, functional level and work experience) for 
significant differences between high potentials and average performers, 
in  order  to  assess  whether  they  should  be  incorporated  as  control 
variables.  We  included  gender,  age  and  educational  level  as  control 
variables on theoretical as well as data‐based grounds. First of all, these 
variables  are  likely  to  impact  on  each  of  the  dependents  (e.g.  Ng  et  al., 
2005).  Second,  χ²  difference  tests  revealed  significant  differences 
between  high  potentials  and  average  performers  with  regard  to  these 
variables.  
 
In  interpreting  the  MANOVA  analyses,  a  particularly  important  remark 
is that the interpretation of the partial Eta squared effect sizes (η²p), is 
not identical to that of other types of effect sizes (e.g. R², Cohen’s d). The 
‘rule‐of‐thumb’ for interpreting η²p values is: small = η²p < .06; medium 
=  .06  ≤  η²p  <  .14;  large  =  η²p  ≥  .14  (Cohen,  1988).  Observed  statistical 
power (π) is considered acceptable at .80 or above. 
 
Results 
 
Table  2  presents  the  means,  standard  deviations,  and  correlations 
among the study variables. Interestingly, correlations between objective 
215
and subjective measures of similar constructs (i.e. performance ratings 
with  satisfaction  with  performance,  promotion  variables  with 
satisfaction  with  advancement,  and  salary  measures  with  satisfaction 
with  degree  of  financial  security)  tend  to  be  small  –  although,  for  the 
most part, significant, which is in line with the literature (e.g. Arthur et 
al., 2005).  
 
Effects of talent category and communication on job performance 
 
An  omnibus  F‐test  of  between‐subjects  effects  revealed  that  the  model 
that  was  tested  (including  all  independent  and  control  variables)  was 
significant, both for self‐rated (F(12,535) =  3.69, p = .00, η²p  = .08, π = 
.99) and supervisor‐rated (F(12,535) = 6.66, p = .00, η²p = .13, π = 1.00) 
job performance. Figure 2 illustrates the direction of the effects of talent 
category  on  self‐rated  (a)  and  supervisor‐rated  (b)  job  performance 
under  conditions  of  closed  and  open  communication  about 
organizational talent categorization policies.  
 
 
 
 

 
Figure 2. Means plots for ‘job performance’ variables 
 
Table 2. Means, standard deviations and intercorrelations of the study variables (n = 711) 
 
  m sd 1  2  3  4  5  6 
job performance   
1. self‐rated performance  3.61 .76 ‐‐  
2. supervisor‐rated performance  3.68 .75 .51** ‐‐  
career success:   
objective    
3. intra‐organizational promotions received  2.01 2.04 .17** .24** ‐‐ 
4. speed of intra‐organizational promotions received 0.26 .29 .12** .26** .45**  ‐‐
5. net month salary  2653.24 1776.16 .03 .05 .26**  .10* ‐‐
6. intra‐organizational net month salary increase  909.43 1379.15 .07 .07 .44**  .08 .61** ‐‐
subjective   
7. satisfaction with performance  5.28 .75 .27** .29** .19**  .15** .02 .05
8. satisfaction with advancement  5.22 .96 .22** .30** .19**  .25** .02 .02
9. satisfaction with personal development  5.20 .79 .18** .25** .21**  .17** .02 .06
216

10. satisfaction with creative outlet  4.82 1.15 .18** .10* .01  .03 .04 .01


11. satisfaction with degree of financial security  4.66 1.44 .05 .22** .20**  .10* .06 .10*
12. satisfaction with degree of job security  5.16 1.32 .01 .12** .13**  .10* .02 .01
13. satisfaction with work‐life dynamics  5.06 .77 .14** .19** .13**  .11** .02 .04
14. satisfaction with recognition  5.13 .90 .10* .32** .30**  .18** .09* .13**
15. satisfaction with inter‐personal cooperation  5.48 .81 .18** .31** .26**  .17** ‐.01 .06
16. satisfaction with contribution  5.18 .76 .19** .27** .27**  .16** .12** .15**
commitment   
17. career commitment   5.10 .59 .30** .23** .15**  .16** .16** .04
18. organizational commitment  4.80 .89 .08 .16** .25**  .11 .14** .11*
19. inter‐organizational mobility preference  3.37 1.19 ‐.03 .01 .11*  ‐.09* .00 .11*
Notes. * p < .05; ** p < .01; coefficient alphas for all multiple‐item scales are inserted on the main diagonal.  
Table 2. Continued 
 
7  8  9  10  11  12  13  14  15  16  17  18  19 
             
             
             
             
             
             
             
             
             
             
(.79)             
.55**  (.60)           
.78**  .58**  (.86)         
217

.45**  .26**  .35**  ‐‐       


.29**  .22**  .31**  ‐.02  ‐‐     
.23**  .21**  .28**  ‐.05  .17**  ‐‐    
.63**  .45**  .70**  .25**  .35**  .33** (.77)    
.49**  .45**  .56**  .07  .44**  .40** .50** (.60)    
.60**  .46**  .65**  .24**  .33**  .24** .55** .58** (.60)    
.73**  .52**  .69**  .38**  .28**  .26** .56** .61** .62** (.77)    
             
.54**  .52**  .53**  .34**  .10  .10 .38** .27** .36** .44** (.81)   
.43**  .45**  .54**  .09  .18**  .32** .46** .55** .52** .45** .36**  (.89) 
.01  ‐.01  .03  ‐.06  .07  .02 .12** .16** .05 .08 ‐.06  .36**  (.80)
Notes. * p < .05; ** p < .01; coefficient alphas for all multiple‐item scales are inserted on the main diagonal.  
 
 
218
Examining the parameter estimates, we conclude that we can accept H1, 
in  that  high  potentials  score  significantly  higher  in  terms  of  job 
performance, both self‐rated (β = .23, p = .01) and supervisor‐rated (β = 
.47,  p  =  .00).  Furthermore,  Figure  2  shows  that  the  difference  between 
high  potentials  and  average  performers  is  larger  for  supervisor‐rated 
performance  than  for  self‐rated  performance.  A  paired  samples  t‐test 
found no significant difference between self‐rated and supervisor‐rated 
for  average  performers  (t(340)  =  ‐1.13,  p  =  .26),  as  opposed  to  a 
significant  difference  for  high  potentials  (t(207)  =  4.60,  p  =  .00).  It 
appears  that  high  potentials  tend  to  rate  their  own  job  performance 
lower  than  do  their  supervisors,  both  under  conditions  of  closed  and 
open communication about organizational talent categorization policies. 
As  concerns  closed  versus  open  communication,  we  found  a  main 
significant effect on self‐rated performance (β = .38, p = .00), but not on 
supervisor‐rated  performance  (β  =  .20,  p  =  .07).  No  interaction  effects 
were  found,  nor  effects  of  the  control  variables.  Since  no  interaction 
effects were found, we must reject H4.1. 
 
Effects of talent category and communication on career success 
 
First,  we  examined  the  hypothesized  relationships  for  objective  career 
success. The hypothesized model proved significant for all four objective 
career  success  variables,  i.e.  (a)  intra‐organizational  promotions 
received  (F(12,501)  =  12.30,  p  =  .00,  η²p  =  .23,  π  =  1.00);  (b)  speed  of 
intra‐organizational promotions received: (F(12,501) = 8.07, p = .00, η²p 
= .16, π =  1.00); (c)  net  month salary: (F(12,501) = 7.79, p  = .00, η²p  = 
.16,  π  =  1.00);  and  (d)  intra‐organizational  net  month  salary  increase: 
(F(12,501) = 4.68, p = .00, η²p = .10, π = 1.00). Figure 3 illustrates these 
effects  under  conditions  of  closed  and  open  communication  about 
organizational talent categorization policies. 
 
As  is  illustrated  in  Figure  3,  high  potentials  score  higher  than  average 
performers  on  all  indicators  of  objective  career  success.  However, 
although  this  effect  is  significant  for  intra‐organizational  promotions 
received  (β  =  .90,  p  =  .00),  speed  of  intra‐organizational  promotions 
received (β = .19, p = .00), and net month salary (β = 428.73, p = .03), it is 
not  for  intra‐organizational  net  month  salary  increase  (β  =  172.75,  p  = 
.27). Furthermore, we found main effects of communication about talent 
categorization for intra‐organizational promotions received (β = 1.16, p 
=  .00),  speed  of  intra‐organizational  promotions  received  (β  =  .09,  p  = 
.04)  and  net  month  salary  (β  =  ‐489.32,  p  =  .05).  In  addition,  an 
interaction effect was found for net month salary (β = 636.57, p = .05). 
However,  this  effect  runs  in  an  unexpected  direction,  in  that  high 
219
potentials  (but  not  average  performers)  receive  significantly  lower 
salaries under conditions of open communication than under conditions 
of  closed  communication  (although  they  do  earn  more  than  average 
performers in all conditions).  
 
Also, for the objective career success model, the control variables play a 
significant  role.  Gender  (β  =  .52,  p  =  .00),  age  (β  =  1.97,  p  =  .00)  and 
educational  level  (β  =  2.08,  p  =  .03)  had  a  significant  effect  on  intra‐
organizational  promotions  received;  age  (β  =  ‐.09,  p  =  .03)  and 
educational  level  (β  =  .17,  p  =  .00)  on  speed  of  intra‐organizational 
promotions received; gender (β = 491.44, p = .00) and age (β = 1501.90, 
p  =  .00)  on  net  month  salary;  and  age  (β  =  1097.05,  p  =  .00)  on  intra‐
organizational net month salary increase.  
 
Second,  we  examined  the  model  for  subjective  career  success. 
Significant  effects  were  found  for  all  ten  subjective  career  success 
variables,  i.e.  (a)  satisfaction  with  performance  (F(12,578)  =  4.11,  p  = 
.00, η²p = .08, π = 1.00); (b) satisfaction with advancement (F(12,578)  = 
9.26,  p  =  .00,  η²p  =  .16,  π  =  1.00);  (c)  satisfaction  with  personal 
development  (F(12,578)  =  3.47,  p  =  .00,  η²p  =  .07,  π  =  1.00);  (d) 
satisfaction with creative outlet (F(12,578) = 2.56, p = .00, η²p = .05, π = 
.98); (e) satisfaction with degree of financial security (F(12,578) = 3.60, 
p = .00, η²p = .07, π = 1.00); (f) satisfaction with degree of job security 
(F(12,578) = 2.84, p = .00, η²p = .06, π = .99); (g) satisfaction with work‐
life  dynamics  (F(12,578)  =  2.40,  p  =  .01,  η²p  =  .05,  π  =  .97);  (h) 
satisfaction  with  recognition  (F(12,578)  =  5.78,  p  =  .00,  η²p  =  .11,  π  = 
1.00); (i) satisfaction with inter‐personal cooperation (F(12,578) = 3.58, 
p  =  .00,  η²p  =  .07,  π  =  1.00);  and  (j)  satisfaction  with  contribution 
(F(12,578)  =  5.16,  p  =  .00,  η²p  =  .10,  π  =  1.00).  Figure  4  illustrates  the 
effects of talent category on subjective career success under conditions 
of  closed  and  open  communication  about  organizational  talent 
categorization policies. 
 
As  hypothesized,  high  potentials  score  higher  on  all  subjective  career 
success  variables  than  do  average  performers  (satisfaction  with 
performance: β = .35, p = .00; satisfaction with advancement: β = .58, p = 
.00; satisfaction with self‐development: β = .29, p = .00; satisfaction with 
creative outlet: β = .23, p = .05; satisfaction with degree of job security: β 
= .56, p = .00; satisfaction with recognition: β = .35, p = .00; satisfaction 
with  inter‐personal  cooperation:  β  =  .25,  p  =  .00;  and  satisfaction  with 
contribution:  β  =  .28,  p  =  .00),  although  this  effect  is  not  significant  for 
satisfaction with financial security (β = .13, p = .40) and satisfaction with 
work‐life dynamics (β = .14, p = .08).  
220
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 3. Means plots for ‘objective career success’ variables 
221
Combined  with  the  findings  for  objective  career  success,  we  can 
therefore  only partially  accept H2, as  significant effects were found for 
the  majority  of  career  success  variables,  but  not  for  all.  Also,  the 
majority  of  subjective  career  success  indicators  demonstrated  main 
effects  of  communication  about  talent  categorization:  satisfaction  with 
performance (β = .44, p = .00), satisfaction with advancement (β = .50, p 
=  .00),  satisfaction  with  self‐development  (β  =  .32,  p  =  .00),  degree  of 
financial security (β = .42, p = .04), satisfaction with work‐life dynamics 
(β  =  .22,  p  =  .04),  satisfaction  with  recognition  (β  =  .37,  p  =  .00),  and 
satisfaction  with  inter‐personal  cooperation  (β  =  .35,  p  =  .00). 
Furthermore,  we  found  interaction  effects  for  two  of  the  subjective 
career success variables.  
 
First, the interaction effect on satisfaction with performance (β = ‐.27, p 
= .04) implies that high potentials (but not average performers) report 
more  satisfaction  with  performance  under  conditions  of  open 
communication  about  talent  categorization  than  under  conditions  of 
closed  communication.  Second,  average  performers  (but  not  high 
potentials) report lower levels of satisfaction with degree of job security 
under  conditions  of  open  communication  about  talent  categorization 
than  under  conditions  of  closed  communication  (β  =  ‐.48,  p  =  .05). 
Putting  these  findings  alongside  our  findings  for  objective  career 
success, we can only partially accept H4.2 as only few interaction effects 
with communication climate were found.  
 
For subjective career success as well, the control variables entered into 
the model were found to contribute. The effect of gender was significant 
for  satisfaction  with  advancement  (β  =  .26,  p  =  .00),  satisfaction  with 
creative outlet (β = .44, p = .00), and satisfaction with contribution (β = 
.12,  p  =  .05).  The  effect  of  age  was  significant  for  satisfaction  with 
advancement (β = ‐1.00, p = .00) and satisfaction with contribution (β = 
.37,  p  =  .05).  Educational  level  was  significant  for  satisfaction  with 
degree  of  job  security  (β  =  1.89,  p  =  .00),  satisfaction  with  work‐life 
dynamics  (β  =  ‐.28,  p  =  .01),  and  satisfaction  with  inter‐personal 
cooperation (β = ‐.55, p = .00).  
 
 
 
 
 
 
 
 
 
 
 
 
 
222

 
 
 
 
 
 
 
 
 
 
 
 
Figure 4. Means plots for ‘subjective career success’ variables  
 
 
 
 
 
 
 
 
 
 
 
 
 
223

 
 
 
 
 
 
 
 
 
 
 
 
Figure 4. Continued  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
224

Figure 4. Continued  
 
225
Effects of talent category and communication on commitment 
 
An  omnibus  F‐test  demonstrated  that  the  model  for  commitment  was 
significant  for  all  three  commitment‐related  variables:  (a)  career 
commitment  (F(12,  524)  =  3.01,  p  =  .00,  η²p  =  .10,  π  =  .99);  (b) 
organizational  commitment  (F(12,  524)  =  3.26,  p  =  .00,  η²p  =  .10,  π  = 
.99); and (c) inter‐organizational mobility preference (F(12, 524) = 2.17, 
p = .02, η²p = .07, π = .93). Figure 5 illustrates the direction of the effects 
of  talent  category  on  the  commitment  variables  under  conditions  of 
closed  and  open  communication  about  organizational  talent 
categorization policies. 
 
For  all  three  commitment  variables,  we  found  that  high  potentials 
scored  significantly  higher  (career  commitment:  β  =  .25,  p  =  .00; 
organizational  commitment:  β  =  .33,  p  =  .01;  and  inter‐organizational 
mobility preference: β = .37, p = .04). We thus accept H3. No main effects 
were found for communication about talent categorization. 
 
Although Figure 5 implies that there is a crossover interaction between 
talent  category  and  communication  climate  for  inter‐organizational 
mobility  preference  (implying,  that  high  potentials  are  less  likely,  and 
average performers more likely, to desire inter‐organizational mobility 
under  conditions  of  open  communication),  this  effect  was  found  to  be 
only marginally significant (β = .52, p = .09). As no significant interaction 
effects were found, we reject H4.3.. 
 
Finally, educational level was found to have a significant effect on career 
commitment  (β  =  .19,  p  =  .04)  and  inter‐organizational  mobility 
preference (β = .63, p = .01), and age on organizational commitment (β = 
‐.60, p = .02).  
 
Discussion 
 
This study responds to calls in the literature for more research on talent 
management  and  high  potentials  (Collings  &  Mellahi,  2002;  Lewis  & 
Heckman,  2006).  We  empirically  investigated  the  effects  of  being 
identified  as  a  high  potential  (or  not)  on  a  number  of  career  variables 
(i.e.  job  performance,  career  success,  and  commitment),  and  the 
potential moderating effect of the degree of openness that is applied by 
organizations in communicating about their high potential programs.  
226
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 5. Means plots for ‘commitment’ variables 
 
 
 
227
Specifically,  we  wanted  to  examine  to  what  extent  disclosing  to 
employees  whether or  not they  are considered high potentials impacts 
job  performance,  career  success  and  commitment.  Based  on  the 
literature,  we  hypothesized  that  open  communication  would  be  a 
positive moderator for those who are identified as high potentials, but a 
negative  moderator  for  those  who  are  not  (i.e.  ‘average  performers’) 
(Burke, 1997).   
 
For  the  most  part,  our  hypothesized  main  effects  were  confirmed, 
indicating  that  employees  identified  as  high  potentials  are  better 
performers,  achieve  more  successful  careers,  and  are  more  committed 
to their career and organization. Only for salary increase over the course 
of  working  for  the  current  organization,  no  differences  between  high 
potentials  and  average  performers  were  found.  This  finding  is  perhaps 
not entirely surprising; earlier research has indicated that organizations 
are  finding  it  very  difficult  to  link  pay  to  potential.  The  ‘intangible’ 
aspect  of  potential  may  cause  issues  in  terms  of  justice  perceptions  by 
employees,  as  well  as  raise  objection  in  employee  unions  (Bossuyt  & 
Dries, 2008).  
 
We  found  positive  main  effects  of  open  communication  about  talent 
management  procedures  on  almost  all  dependent  variables,  implying 
that  adopting  an  open  communication  strategy  generates  overall 
positive  effects  on  job  performance  and  career  success.  Considering 
these  results,  one  could  wonder  whether  open  communication  about 
who is on the high potential list is indicative of an open communication 
climate  in  general,  or  even  of  a  specific  type  of  organizational  culture 
that  impacts  positively  on  worker  attitudes  and  performance.  Further 
research  needs  to  be  done  in  order  to  examine  exactly  how  open 
communication  about  talent  management  generates  these  positive 
effects.  
 
With regard to our hypotheses about potential moderation effects of the 
openness  of  organizational  communication,  the  evidence  was  less 
striking.  Only  a  few  interaction  effects  were  found.  First,  as  for  net 
month salary (which was one of the measures we used to operationalize 
objective career success), we found, contrary to expectations, that high 
potentials  (but  not  average  performers)  receive  significantly  lower 
salaries under conditions of open communication than under conditions 
of  closed  communication  –  although  they  do  earn  more  than  average 
performers  in  all  conditions.  The  same  trend  is  observable  for  net 
month  salary  increase  over  the  period  of  working  for  the  current 
organization, but in this case the interaction effect was non‐significant. 
228
It thus appears that the salary ‘gap’ between high potentials and average 
performers  is  larger  in  organizations  that  do  not  openly  communicate 
about  high  potential  status  than  in  organizations  that  do.  In  the 
Introduction,  we  hypothesized  that  under  conditions  of  open 
communication  about  talent  management  procedures,  high  potentials 
would  receive  higher  salaries  as  differentiation  between  talent 
categories  in  terms  of  salary  and  career  opportunities  would  be  easier 
to  implement.  We  thus  made  two  assumptions:  first,  that  since  one  of 
the  goals  of  talent  management  is  to  retain  employees  considered  as 
high  potentials,  organizations  may  want  to  offer  them  a 
disproportionate  amount  of  career  ‘rewards’  in  order  to  enhance  the 
social  exchange  relationship;  second,  that  differentiation  between 
employees  is  less  easy  to  implement  (especially  in  terms  of  employee 
attitudes)  when  the  grounds  on  which  the  differentiation  is 
implemented are intransparant. However, an alternative explanation is 
possible.  An  interview  study  conducted  in  a  sample  of  high  potentials 
who were aware of their status found that many of the high  potentials 
felt  rewarded  for  their  hard  work  simply  by  receiving  the  label  itself. 
The authors concluded that the high potential label can be considered as 
a  type  of  psychological  reward  (Dries  &  Pepermans,  2008).  One  could 
argue then, that organizations that adopt an open communication policy 
do  not  necessarily  have  to  use  differential  pay  to  achieve  retention  of 
their high potentials. The question remains, however, how organizations 
that  adopt  a  closed  communication  strategy  about  their  talent 
management procedures succeed in keeping their ‘list’ of high potentials 
a secret when they offer differential pay to high potentials and non‐high 
potentials.  Often,  employees  working  for  such  organizations  can  infer 
from the differential treatment given to certain coworkers who is on the 
list and who is not (Burke, 1997; Dries & Pepermans, 2008).  
 
A second interaction effect that was found was for the satisfaction with 
performance scale, one of the subscales of the subjective career success 
variable.  Results  imply  that  high  potentials  (but  not  average 
performers) report more satisfaction with their own performance under 
conditions  of  open  communication  than  under  conditions  of  closed 
communication.  This  finding  seems  to  fit  in  with  our  hypothesis  that 
openly  disclosing  high  potential  status  could  provoke  self‐fulfilling 
prophecy effects, where employees who receive positive feedback about 
their  performance  consequently  feel  motivated  to  perform  even  better 
(Eden,  1984;  Goddard,  1985;  Kierein  &  Gold,  2000).  However,  we  did 
not find any significant interaction effects between open communication 
and actual self‐rated or supervisor‐rated performance. Furthermore, we 
found  that  high  potentials  (but  not  average  performers)  tend  to  rate 
229
their  own  performance  significantly  lower  than  their  supervisors,  both 
under  conditions  of  open  and  closed  communication.  A  possible 
explanation  for  these  findings  might  be  that  employees  identified  as 
high  potentials  are  typically  ‘overachievers’,  continuously  raising  the 
bar of what it means to be an excellent performer (Dries & Pepermans, 
2008;  Dubouloy,  2004).  As  a  consequence,  they  may  feel  that  they  are 
able  to  achieve  more  than  their  supervisors  expect  of  them,  causing 
them  to  rate  their  own  performance  lower  than  their  supervisors  do. 
Another  important  remark  is  that  the  satisfaction  with  performance 
scale  refers  to  the  performance  throughout  the  entire  career,  whereas 
the job performance measures relate to the performance demonstrated 
throughout the most recent performance appraisal cycle. It thus appears 
that openly communicating to high potentials about their status causes 
them to evaluate their performance throughout their entire career more 
positively, but not necessarily their current job performance. An earlier 
qualitative  study  (Dries  &  Pepermans,  2008)  revealed  that  the  high 
potential  label  is  perceived  by  high  potentials  as  carrying  long‐term 
promises  about  career  advancement,  which  may  explain  why  it  causes 
employees who know they have been identified as high potentials to feel 
more  satisfied  about  the  performance  in  their  careers.  Some  authors, 
however,  advise  organizations  to  engage  in  expectations  management 
when  it  comes  to  the  career  ‘guarantees’  employees  perceive  to  be 
attached to the high potential label. First of all, the high potential label is 
not  often  granted  to  employees  for  an  indefinite  amount  of  time;  in 
order  to  keep  the  label,  high  potentials  have  to  prove  their  worth  year 
after  year  (and  appraisal  cycle  after  appraisal  cycle).  Second,  the 
responsibility  for  career  management  lies  increasingly  in  the  hands  of 
the individual employee (Burke, 1997; Cope, 1998; Dries & Pepermans, 
2008).  Therefore,  it  is  very  important  that  organizations  communicate 
about  the  amount  of  support  versus  self‐management  high  potentials 
can expect.  
 
A  third,  particularly  interesting,  interaction  effect  that  was  found  was 
the effect on satisfaction with the degree of job security, another of the 
subjective  career  success  subscales.  Although  we  found  positive  main 
effects of open communication about talent management procedures on 
almost  all  dependent  variables  –  implying  that  openly  communicating 
about who is on the high potential list and who is not, overall, generates 
the most desirable results – this finding seems to confirm organizations’ 
fears  about  open  communication.  Specifically,  we  found  that  average 
performers (but not high potentials) report lower levels of job security 
when they are aware of the fact that there is a high potential list in the 
organization,  and  they  are  not  on  it.  A  related  finding  (although  only 
230
marginally  significant)  was  the  interaction  effect  for  inter‐
organizational  mobility  preference:  under  conditions  of  open 
communication,  high  potentials  are  less  likely  to  report  an  interest  in 
exploring  employment  opportunities  outside  of  their  current 
organization,  where  average  performers  are  more  likely  to  do  so. 
Although  these  findings  do  imply  that  adopting  open  communication 
policies about talent management might encourage non‐high potentials 
to  be  more  open  to  employment  opportunities  elsewhere,  we  must 
stress that, overall, we found more evidence of potential positive effects 
of open communication than negative effects.  
 
Limitations and implications 
 
Despite  its  potential  contributions  to  the  literature  on  talent 
management and high potentials, the current study did have a number 
of  limitations.  First  of  all,  as  was  mentioned  in  the  Introduction,  the 
concept  of  talent  management  encapsulates  a  much  broader  range  of 
HRM initiatives than merely the categorization of employees into pools 
of  high  potentials  and  non‐high  potentials  –  or  so  it  should  (Bossuyt  & 
Dries, 2008; Lewis & Heckman, 2006). We wish to stress that, although 
the focus of this paper is on talent categorization, we believe that talent 
management  is  in  fact  much  broader.  However,  we  believe  that 
qualitative  approaches  may  be  more  suited  for  studying  talent 
management  from  a  broader  perspective,  as  identifying  and 
operationalizing  all  relevant  aspects  across  organizations  may  be 
difficult  to  achieve  with  a  survey  design.  More  qualitative  and 
quantitative  studies  (the  former  more  exploratory,  the  latter  more 
narrowly  focused)  on  talent  management  and  the  issues  surrounding 
high  potential  identification,  development  and  retention  should  be 
conducted  in  the  future  in  order  to  build  much‐needed  theory  about 
these ‘hot topics’ (e.g. Buckingham & Vosburgh, 2001).  
 
Second,  although  we  attempted  to  homogenize  our  sample  as  much  as 
possible by imposing strict study inclusion criteria, some heterogeneity 
did  creep  into  our  sample.  In  order  to  adequately  capture  these 
heterogeneities,  we  included  a  table  (Table  1)  in  the  Method  section 
summarizing  relevant  organizational  and  high  potential  program 
characteristics  for  the  twelve  participating  organizations.  Most 
noteworthy are the differences in terms of the percentage of employees 
categorized  as  high  potentials.  Furthermore,  the  majority  of 
organizations  in  our  sample  were  large  and  hierarchically  structured, 
probably  due  to  the  fact  that  our  strict  inclusion  criteria  demand 
formalized  high  potential  programs.  It  is  highly  likely  that  smaller  or 
231
flatter  organizations  implement  different  approaches  to  talent 
management,  as  they  are  characterized  by  vastly  different  career 
opportunity  structures.  Therefore,  organizational  size,  culture 
(especially in terms of communication climate) and specific elements of 
the  organization’s  talent  management  policies  should  be  taken  into 
consideration  when  generalizing  the  findings  reported  in  the  current 
study.  Furthermore,  the  non‐longitudinal  nature  of  our  study  design 
implies that causal inferences should be regarded as largely speculative.  
 
Third,  although  we  operationalized  open  versus  closed  communication 
about  talent  management  procedures  simply  as  the  disclosure  or  non‐
disclosure  of  high  potential  status  to  employees,  we  do  acknowledge 
that there are still other types of information an organization can choose 
to  communicate  about  with  different  levels  of  openness:  the  level  and 
type of job employees are targeted for in terms of succession planning; 
the  size  of  the  high  potential  pool  organization‐wide;  what  role  the 
organization will play in employees’ career management; and what role 
employees  are  expected  to  play  in  their  own  career  planning  (Burke, 
1997).  We  opted  for  a  simple  operationalization  of  open  versus  closed 
communication,  however,  in  order  to  be  able  to  make  a  clear‐cut 
distinction between organizations in terms of communication openness. 
Furthermore,  being  informed  of  one’s  high  potential  status  or  not  was, 
for our research goals, the most relevant variable to assess.  
 
A final limitation lies in the fact that we were unable to fully debrief the 
study  respondents  as  the  survey  incorporated  a  variable  that  was  not 
communicated  to  respondents  (i.e.  whether  they  are  identified  as  high 
potentials or not),  although they did receive extensive  feedback on the 
other  individual‐level  variables.  This  issue  was  extensively  discussed 
with the participating organizations; it finally became clear that setting 
up such a design was the only possible way for obtaining participation. 
It  does,  however,  raise  some  concerns  about  the  ethicality  of  such 
‘secretive’ talent categorization procedures in organizations.  
 
Either way, the current paper represents a rigorous attempt to answer 
some  of  the  most  urgent  questions  about  talent  management  and  high 
potentials,  based  on  empirical  findings.  In  order  to  advance  the  field, 
however, much more empirical research needs to be done. Longitudinal 
designs  could  provide  further  insight  into  the  dynamics  and 
consequences  of  identifying  employees  as  high  potentials.  Other 
important  topics  for  further  research  include:    the  extent  to  which 
‘generic’  criteria  can  be  conceived  for  identifying  high  potentials,  as 
opposed  to  organization‐specific  criteria,  and  the  interactions  between 
232
person‐centered  and  situational  variables  playing  a  role  in  high 
potential identification (Hollenbeck et al., 2006); psychological contract 
dynamics,  turnover  intentions  and  retention  factors  for  employees  in 
different  talent  categories;  and  the  effects  of  talent  management 
interventions  aimed  at  non‐managerial  talent  (e.g.  the  ‘dual  career 
track’) (Bossuyt & Dries, 2008).    
 
As  for  practical  implications,  our  findings  seem  to  demonstrate  that 
openly  communicating  about  which  employees  are  identified  as  high 
potentials  and  which  are  not  generates  the  most  desirable  outcomes 
overall  for  employee  performance,  success  and  commitment.  Earlier 
research  on  communication  in  organizations  came  to  similar 
conclusions.  Ng  &  Butts  (2009)  found  that  the  perceived  relationship 
between  employees  and  their  organizations  can  be  strengthened  by 
information  sharing;  Vandenberg  et  al.  (1999)  found  that  providing 
employees  with  access  to  information  about  the  organization  is 
negatively  related  to  turnover  intentions.  Cope  (1998)  identified  three 
benefits of communicating openly about talent management procedures 
based  on  the  case  of  one  large  organization  that  made  the  transition 
from  a  closed  to  an  open  communication  system.  First  of  all,  the 
increased  transparency  of  the  leadership  development  process  caused 
higher  appreciation  and  more  respect  for  the  procedures  in  line 
managers;  second,  pre‐  and  post‐transition  employee  surveys  revealed 
more  positive  employee  perceptions  of  the  leadership  development 
programs;  and  third,  the  public  nature  of  the  talent  management 
procedures encouraged development of a rigorous process, well able to 
withstand  scrutiny.  The  business  ethics  literature  (e.g.  Van  Buren, 
2003), alternatively, argues that information about internal employment 
prospects must be offered to allow employees to assess whether or not 
to  maintain  the  employment  relationship  with  the  organization  or  to 
disengage.  In  turn,  of  course,  employees  owe  the  organization  an 
obligation  to  use  their  skills  and  abilities  as  contracted  for  in  the 
employment agreement, whether written or oral. 
 
However,  the  literature  also  identifies  possible  pitfalls  of  adopting  an 
open  communication  about  organizational  talent  management 
procedures.  Cope  (1998)  identified  two  issues  with  open 
communication.  First,  as  organizations’  high  potential  identification 
procedures  are  often  based  on  annual  assessments  of  potential,  it  is 
possible to be seen as a high potential one year, but not the next – and 
an  open  communication  system  would  require  communicating  such 
“demotions”  as  well.  Second,  communication  to  employees  about  their 
high potential status might create false expectations about guarantees in 
233
terms  of  promotions  or  successful  careers  (Baruch  &  Peiperl,  1997). 
Eisenberg  &  Witten  (1987),  in  their  seminal  article  about  strategic 
ambiguity  as  a  communication  strategy,  described  how  managers  who 
had  been  overly  explicit  in  their  policies  paid  dearly  later  on,  when  a 
violation by a valued employee forced them to choose between making a 
good decision and remaining consistent with previous pronouncements. 
The  authors  state  that  ambiguous  communication  can  thus  allow 
organizations greater freedom to respond to environmental changes. An 
important  factor  to  take  into  account  in  adopting  an  open  or  closed 
communication strategy about talent management is the organization’s 
general  culture.  Benefits  of  open  communication  may  be  more  easily 
reaped  when  the  company  culture  supports  open  communication  (i.e. 
whether  communication  about  high  potential  status  is  consistent  with 
everyday practices) (Burke, 1997). Further research needs to be done in 
order  to  examine  which  types  of  organizational  cultures  facilitate 
implementation of an open communication climate – and conversely, in 
which  types  of  organizations  open  communication  is  not  feasible  or 
desirable.   
 
References 
 
Arthur,  M.B.,  Khapova,  S.N.  &  Wilderom,  C.P.M.  (2005).  Career  success  in  a  boundaryless  career 
world. Journal of Organizational Behavior, 26 (2), 177‐202.  
Aston,  C.  &  Morton,  L.  (2005).  Managing  talent  for  competitive  advantage.  Strategic  HR  Review,  4 
(5), 28‐31. 
Baruch,  Y.  &  Peiperl,  M.  (1997).  High‐flyers:  Glorious  past,  gloomy  present,  any  future?  Career 
Development International, 2 (7), 354‐358. 
Bossuyt, T. & Dries, N. (2008). Talent management en flexibele loopbaanpaden voor de werknemers 
van morgen [Talent management and flexible career tracks for tomorrow's workforce]. In H. 
Calmeyn, De Witte, K. and Weverbergh, J. (Eds.), Licht op leren 2008: Leren en ontwikkelen in 
een  talentgerichte  maatschappij  [Spotlight  on  learning  2008:  Learning  and  developing  in  a 
talent­oriented society]. Leuven, Belgium: LannooCampus.  
Boudreau,  J.W.  &  Ramstad,  P.M.  (2005).  Where’s  your  pivotal  talent?  Harvard  Business  Review,  83 
(4), 23‐24. 
Briscoe, J.P., Hall, D.T. & Frautschy DeMuth, R.L. (2006). Protean and boundaryless career attitudes: 
An empirical exploration. Journal of Vocational Behavior, 69 (1), 30‐47. 
Buckingham,  M.  &  Vosburgh,  R.M.  (2001).  The  21st  century  Human  Resources  function:  It's  the 
talent, stupid! Human Resource Planning, 24 (4), 17‐23. 
Burke,  L.A.  (1997).  Developing  high  potential  employees  in  the  new  business  reality.  Business 
Horizons, 40 (2), 18‐24.  
Collings,  D.G.  &  Mellahi,  K.  (2009).  Strategic  talent  management:  A  review  and  research  agenda. 
Human Resource Management Review, early online, corrected proof.  
Cope, F. (1998). Current issues in selecting high potentials. Human Resource Planning, 21 (3), 15‐17. 
Day,  R.  &  Allen,  T.D.  (2004).  The  relationship  between  career  motivation  and  self‐efficacy  with 
protégé career success. Journal of Vocational Behavior, 64 (1), 72‐91.  
Derr, C.B.,  Jones, C. & Toomey, E. (1988). Managing high potential employees: Current practices in 
thirty‐three US corporations. Human Resource Management, 27 (3), 273‐290. 
Dries,  N.  &  Pepermans,  R.  (2007).  Using  emotional  intelligence  to  identify  high  potentials:  A 
metacompetency perspective. Leadership and Organization Development Journal, 28 (8), 749‐
770. 
Dries,  N.  &  Pepermans,  R.  (2008).  "Real"  high  potential  careers:  An  empirical  study  into  the 
perspectives of organizations and high potentials. Personnel Review, 37 (1), 85‐108. 
234
Dries,  N.,  Van  Loo,  S.  &  Pepermans,  R.  (2009).  Towards  a  truly  subjective  conception  of  career 
success.  Paper  presented  at  the  25th  annual  conference  of  the  European  Group  for 
Organization Studies in Barcelona, Spain.  
Dubouloy, M. (2004). The transitional space and self‐recovery: A psychoanalytical approach to high 
potential managers’ training. Human Relations, 57 (4), 467‐496.  
Eden, D. (1984).  Self‐fulfilling prophecy as a management tool: Harnessing Pygmalion. Academy of 
Management Review, 9, 64–73.  
Eisenberg,  E.M.  (1984).  Ambiguity  as  strategy  in  organizational  communication.  Communication 
Monographs, 51, 227‐242.  
Eisenberg,  E.M.  &  Witten,  M.  (1987).  Reconsidering  openness  in  organizational  communication. 
Academy of Management Review, 12, 418‐426.  
Festinger, L. (1957). A theory of cognitive dissonance. Stanford, CA: Stanford University Press. 
Fields, G.S. (2002). Predicting potential for promotion: How the data in human resource information 
systems can be used to help organizations gain competitive advantage. Working paper no. 02‐
14, Cornell University, Ithaca, NY.  
Fresina,  T.  (1987).  The  identification  &  development  of  high  potential  managers:  A  benchmarking 
report  on  midsized  &  Fortune  500  companies  in  all  major  industries.  Palatine,  IL:  Executive 
Knowledgeworks.  
Göbel‐Kobialka,  S.  (1998).  Reaching  business  excellence  through  sound  people  management. 
European Journal of Work and Organizational Psychology, 7 (4), 549‐556.  
Greenhaus,  J.H.,  Parasuraman,  S.  &  Wormley,  W.M.  (1990).  Effects  of  race  on  organizational 
experiences,  job  performance  evaluations,  and  career  outcomes.  Academy  of  Management 
Journal, 33 (1), 64‐86. 
Goddard, R.W. (1985). The Pygmalion effect. Personnel Journal, 64, 10‐16.  
Hair, J.F., Anderson, R.E., Tatham, R.L. & Black, W.C. (1998). Multivariate data analysis with readings. 
Englewood Cliffs, NJ: Prentice Hall.  
Hall,  D.T.  &  Chandler,  D.E.  (2005).  Psychological  success:  When  the  career  is  a  calling.  Journal  of 
Organizational Behavior, 26 (2), 155‐176.  
Hall, D.T. & Seibert, K.W. (1992). Strategic management development: linking strategy, succession 
planning,  and  managerial  learning.  In  D.H.  Montross  and  C.J.  Shinkman  (Eds.),  Career 
Development: Theory and Practice. Springfield, IL: Charles C. Thomas Press. 
Heslin,  P.A.  (2005).  Conceptualizing  and  evaluating  career  success.  Journal  of  Organizational 
Behavior, 26 (2), 113‐136. 
Hollenbeck, G.P., McCall,  M.W. & Silzer, R.F. (2006). Theoretical and  practitioner letters: Leadership 
competency models. The Leadership Quarterly, 17, 398‐413.  
Hu,  L.  &  Bentler,  P.  M.  (1999).  Cutoff  criteria  for  fit  indexes  in  covariance  structure  analysis: 
Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1‐55. 
Iles,  P.  (1997).  Sustainable  high  potential  career  development:  A  resource‐based  view.  Career 
Development International, 2 (7), 347‐353.  
Jerusalim,  R.S.  &  Hausdorf,  P.A.  (2007).  Managers'  justice  perceptions  of  high  potential 
identification practices. Journal of Management Development, 26 (10), 933‐950.  
Kierein,  N.M.  &  Gold,  M.A.  (2000).  Pygmalion  in  work  organizations:  A  meta‐analysis.  Journal  of 
Organizational Behavior, 21, 913‐928.  
Lewis,  R.E.  &  Heckman,  R.J.  (2006).  Talent  management:  A  critical  review.  Human  Resource 
Management Review, 16, 139‐154.  
McCall, M.W. (1988). Developing  executives through work experiences. Human Resource Planning, 
11 (1), 1‐11.  
Michaels,  E.,  Handfield‐Jones,  H.,  &  Axelrod,  B.  (2001).  The  war  for  talent.  Boston,  MA:  Harvard 
Business School Press. 
Ng, T. & Butts, M.M. (2009). Effectiveness of organizational efforts to lower turnover intentions: The 
moderating role of employee locus of control. Human Resource Management, 48 (2), 289‐310.  
Ng, T., Eby,  L. T.,  Sorensen,  K.  & Feldman, D. (2005). Predictors of objective and subjective career 
success: A meta‐analysis. Personnel Psychology, 58 (2), 367‐408.  
Nicholson, N. & de Waal‐Andrews, W. (2005). Playing to win: Biological imperatives, self‐regulation, 
and trade‐offs in the game of career success. Journal of Organizational Behavior, 26 (2), 137‐
154.  
Pallant, J. (2001). SPSS Survival Manual. Philadelphia, PA: Open University Press.  
Pepermans,  R.,  Vloeberghs,  D.  &  Perkisas,  B.  (2003).  High  potential  identification  policies:  An 
empirical study among Belgian companies. Journal of Management Development, 22 (8), 660‐
678. 
235
Peteraf, M.A. (1993). The cornerstones of competitive advantage: A resource‐based view. Strategic 
Management Journal, 14 (3), 170‐181.  
Rottinghaus, P. J., Day, S. & Borgen, F. H. (2005). The Career Futures Inventory: A measure of career‐
related adaptability and optimism. Journal of Career Assessment, 13 (1), 3‐24. 
Segalla, M., Rouziès, D. & Flory, M. (2001). Culture and career advancement in Europe: Promoting 
team players vs. fast trackers. European Management Journal, 19 (1), 44‐57. 
Shore,  L.  M.  &  Barksdale,  K.  (1998).  Examining  degree  of  balance  and  level  of  obligation  in  the 
employment relationship: A social exchange approach. Journal of Organizational Behavior, 19, 
731‐744.  
Snipes,  J.  (2005).  Identifying  and  Cultivating  High  potential  Employees.  Available  at: 
http://www.ninthhouse.com/papers/2%20CLO_HIPOsArticle.pdf (Accessed July 22nd, 2009). 
Spreitzer, G.M., McCall, M.W. & Mahoney, J.D. (1997). Early identification of international executive 
potential. Journal of Applied Psychology, 82 (1), 6‐29. 
Sturges,  J. (1999). What it means to succeed: Personal conceptions of career success held by male 
and female managers at different ages. British Journal of Management, 10 (3), 239‐252.  
Van  Buren,  H.J.  III  (2003).  Boundaryless  careers  and  employability  obligations.  Business  Ethics 
Quarterly, 13 (2), 131‐49. 
Vandenberg,  R.J.,  Richardson,  H.A.  &  Eastman,  L.J.  (1999).  The  impact  of  high  involvement  work 
processes on organizational effectiveness: A second order latent variable approach. Group & 
Organization Management, 24(3), 300‐339. 
Viney, C., Adamson, S. & Doherty, N. (1997). Paradoxes of fast‐track career management. Personnel 
Review, 26 (3), 174‐186. 
Zuckerman, E.W., Kim, T.Y., Ukanwa, K. & von Rittmann, J. (2003). Robust identities or non‐entities? 
Typecasting in the feature film labor market. American Journal of Sociology, 108, 1018 ‐ 1075. 
 
236
237
 
 
 
 
Chapter IV.  
Constructing ‘career’ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Paper 7. 
Dries, N., Pepermans, R. & De Kerpel, E. (2008). Exploring four generations' beliefs about 
career: Is ‘satisfied’ the new ‘successful’? Journal of Managerial Psychology, 23 (8), 907‐
928. 
 
Paper 8. 
Dries,  N.,  Pepermans,  R.  &  Carlier,  O.  (2008).  Career  success:  Constructing  a 
multidimensional model. Journal of Vocational Behavior, 73 (2), 254‐267. 
238
 
239
Paper 7. 
 

Exploring four generations' beliefs about 
career: Is ‘satisfied’ the new ‘successful’? 
 
 
The current study examined whether four different generations (Silent Generation, Baby 
Boomers,  Generation  X  and  Generation  Y)  hold  different  beliefs  about  career.  Career 
type, career success evaluation and importance attached to organizational security were 
scrutinized for each generation. 750 people completed a vignette task, rating the career 
success  of  32  fictitious  people.  Each  vignette  contained  a  different  combination  of  five 
career  features  (functional  level,  salary,  number  of  promotions,  promotion  speed  and 
satisfaction)  at  two  levels  (low  and  high).  Furthermore,  several  items  were  added  in 
order to determine each participant’s career type and the extent to which they attached 
importance  to  organizational  security.  The  majority  of  participants  still  had  rather 
‘traditional’  careers,  although  younger  generations  seemed  to  exhibit  larger 
discrepancies  between  career  preferences  and  actual  career  situation.  Overall, 
satisfaction  appeared  to  be  the  overriding  criterion  used  to  evaluate  other  people’s 
career success. No significant differences were found between generations. With regard 
to importance attached to organizational security, the Silent Generation and Generation 
Y  scored  significantly  higher  than  the  other  generations.  However,  the  convenience 
sampling strategy led to large differences in sample size per generation. Using a vignette 
design  limited  the  amount  and  richness  of  information  that  could  be  offered  to 
participants. Perhaps other criteria relevant to real‐life career success evaluation were 
not  incorporated  in  this  study.  Nonetheless,  this  study  raises  questions  about  the 
validity of career success operationalizations frequently used in research. It is the first 
study that examines career success evaluation by means of vignettes. 
 
 

 
240

Figure VIII. Topicality of Paper 7.
241
Paper 7. 
 
Exploring four generations' beliefs about career: Is ‘satisfied’ the 
new ‘successful’? 
 
Career as a dynamic social construct 
 
Careers  are  not  free  of  social  context  (Higgins,  2001).  Rather,  they  are 
influenced  by  political,  economic,  historical  and  socio‐cultural 
developments in society (Chen, 1997; Collin and Young, 1992; Herr and 
Cramer,  1992;  Zunker,  1994).  Over  the  last  couple  of  decades, 
phenomena  such  as  economic  globalization,  organizational 
restructuring  (mergers,  horizontal  and  vertical  integrations,  re‐
engineering) and the growth of services have, indeed, drastically altered 
the  face  of  careers  (Barley,  1989;  DeMeuse  and  Tornow,  1990;  Mirvis 
and  Hall,  1994;  Rousseau,  1990).  As  a  result  the  recent  ‘postmodern’ 
career  literature  has  focused,  for  the  most  part,  on  the  shift  from 
traditional (organizational, linear) careers to more ‘boundaryless’ (non‐
linear)  career  types  (e.g.  Arthur  and  Rousseau,  1996;  Briscoe  and  Hall, 
2005; Collin, 1998). While the old, linear career focused on progressive 
steps  upward  in  an  organizational  hierarchy  to  positions  of  greater 
authority  (Brousseau  et  al.,  1996),  the  new  career  defies  traditional 
employment  assumptions,  emphasizing  continuously  changing  career 
paths  and  possibilities  (Arthur  and  Rousseau,  1996;  Littleton  et  al., 
2000).  
 
When  considering  the  definitions  given  to  the  social  construct  that  is 
‘career’ (Arthur et al., 2005) over the years, it is noticeable that, parallel 
to the changes going on in society, there has been a shift in terminology: 
Wilensky  (1961)  referred  to  career  as  a  succession  of  related  jobs, 
arranged in a hierarchy of prestige, through which persons move in an 
ordered  (more‐or‐less  predictable)  sequence  (p.  523);  Super  (1980) 
defined  career  as  the  combination  and  sequence  of  roles  played  by  a 
person  during  the  course  of  a  lifetime  (p.  282);  Arthur  et  al.  (1989) 
defined career as the evolving sequence of a person’s work experiences 
over time (p. 8) – the latter definition being most frequently used today.  
 
The  evolution  from  ‘jobs’  to  ‘experiences’  was  reflected  in  the 
postmodern turn in the social sciences during the late 1980s (Savickas, 
1995),  when  attention  increasingly  shifted  from  the  objective  to  the 
subjective  world  of  work.  While  the  objective  career  is  publicly 
accessible  (Arthur  et  al.,  2005),  as  it  is  expressed  by  symbols  such  as 
more  or  less  identifiable  positions,  offices,  statuses  and  situations 
serving  as  markers  for  assessing  a  person’s  movement  throughout 
242
society  (Barley,  1989;  Ng  et  al.,  2005),  the  subjective  career  relies  on 
individuals’ internal apprehensions and evaluations of their own career, 
across any dimensions that are important to them (Arthur et al., 2005; 
Bozionelos,  2004).  When  assessing,  then,  the  amount  of  success  there 
has been in a career, one must consider first who is doing the assessing 
(Jaskolka et al., 1985). 
 
Since  traditional  public  symbols  of  career  (i.e.  job  titles  referring  to 
hierarchical  positions,  continuity  and  pace  of  promotions,  salary)  are 
losing  relevance  in  the  postmodern  world  of  work,  however,  reference 
points for career success evaluation are disappearing, and it seems that 
a  clear  and  comprehensive  understanding  of  what  ‘career’  and  ‘career 
success’ mean is no longer self‐evident (Adamson et al., 1998; Osterman, 
1996;  Spilerman,  1977).  There  has  been  surprisingly  little  research, 
however,  devoted  to  the  nature  of  career  success  (Greenhaus,  2003; 
Heslin, 2003; Sturges, 1999), although the construct itself has often been 
used  as  a  research  variable  (e.g.  Eby  et  al.,  2003;  Kirchmeyer,  1998, 
2002; Ng et al., 2005; Wayne et al., 1999).  
 
There are, however, some exceptions. Cangemini and Guttschalk (1986) 
discovered  that,  based  on  a  survey  of  35  000  employees,  what 
employees most desired (i.e. appreciation for their work, being involved, 
and  receiving  a  personal  treatment)  differed  substantially  from  what 
supervisors thought they wanted (i.e. money). Friedman and Greenhaus 
(2000)  let  over  800  professionals  rate  the  relative  importance  of  15 
potential  indicators  of  their  career  success.  Five  dimensions  of  career 
success  were  established:  status,  time  for  self,  challenge,  security  and 
social.  With the  exception  of  status,  these  results  reveal  a  considerable 
emphasis  on  subjective  criteria  rather  than  focusing  on  objective 
indicators  of  career  success  such  as  prestige,  power,  money  and 
advancement.  Finally,  Finegold  and  Mohrman  (2001)  concluded  that 
among  4  500  knowledge  workers  and  managers  from  eight  countries, 
work‐life balance was considered most important out of all aspects of a 
career. Although these studies have all made significant contributions to 
the career literature, it is clear that they focus mostly on the increasing 
diversity  in  idiosyncratic  evaluations  of  what  career  success  means  to 
the  individual,  rather  than  looking  at  societal  trends,  the  collective,  or 
reference groups and subcultures (Barley, 1989; Chen, 1997; Shibutani, 
1962).  
 
Alternatively,  the  current  study  aims  to  call  attention  to  the  shared 
social  understanding  of  what  ‘career  success’  means  (Arthur  et  al., 
2005)  and  to  the  question  as  to  whether  this  understanding,  too,  has 
243
shifted. Although individuals tend to see themselves as operating within 
structures that have an objective existence, these structures are actually 
constituted  by  their  own  ‘instantiation’  of  the  social  reality  of  career  – 
i.e.  the  collective  underwriting  of  its  terms.  In  that  way,  career 
structures  are  reproduced  in  the  minds,  actions  and  interpretations  of 
the  people  that  have  the  careers  (Evetts,  1992).  In  order  to  be  able  to 
investigate  whether  the  shared  social  understanding  of  what  career 
success is has shifted during the last few decades, this study turns to the 
literature on generations and their differences. Generations’ beliefs and 
value  systems  are  believed  to  epitomize  societal  trends;  as  a  result  of 
the  experiences  that  have  been  shared  by  people  from  the  same 
generational cohort it is, indeed, plausible that they would develop a so‐
called ‘peer personality’ or certain shared ‘generational characteristics’, 
at least to some extent (De Kort, 2004; Kupperschmidt, 2000). 
 
Looking at career through a generational lens 
 
A generation is defined as “an identifiable group that shares birth year, 
age  location,  and  significant  life  events  at  critical  development  stages, 
divided  by  five‐seven  years  into  the  first  wave,  core  group,  and  last 
wave”  (Kupperschmidt,  2000,  p.  364).  Our  society  has  labelled  the 
generations  of  the  20th  century  –  however,  these  labels  and  the  years 
they represent are not always consistent among authors (de Kort, 2004; 
Smola  and  Sutton,  2002).  Table  1  provides  an  overview  of  the  four 
generations  that  will  be  the  focus  of  the  current  study:  the  Silent 
Generation, the Baby Boomers, Generation X and Generation Y, together 
making  up  the  vast  majority  of  the  world’s  currently  living  population. 
This study has opted for these specific labels and time periods based on 
a thorough search of the  literature (e.g. de Kort, 2004; Kupperschmidt, 
2000;  Lankard,  1995;  Parker  and  Cusmir,  1991;  Smola  and  Sutton, 
2002)  –  however,  a  list  of  other  frequently  used  labels  was  added  for 
each generation.    
 
Furthermore,  Table  1  offers  an  overview  of  the  salient  events 
experienced  by  people  from  the  same  generational  cohort,  beginning 
with  the  Silent  Generation,  which  entered  the  workforce  in  the  post‐
World War II era, when the notion of ‘career’ was forged. The increase in 
opportunities  for  advancement,  and  the  broadening  of  occupations  to 
choose  from  appealed  greatly  to  this  generation  that  had  grown  up 
experiencing or hearing about the Great Depression and the Great War 
(Barley,  1989;  Mirvis  and  Hall,  1994).  Following  this  era  of  economic 
prosperity,  the  late  1970s  and  1980s  were  characterized  by  economic 
recession.  
Table 1. Synopsis of the four generations under study 
 
Generation  Equivalent labels  Birth  Salient events  General values  Work­related values  Credo 
year 
Silent  Traditional Generation  1925‐ Great Depression Conformism Obedience  “We must pay 
Generation  Conservatives  1945  World War II  Maturity  Loyalty  our dues and 
Matures    Conscientiousness  Obligation  work hard.” 
Thrift  Security (stability) 
Baby  Boom(er) Generation  1946‐ Kennedy‐King murders Idealism Challenge  “If you have it, 
Boomers  ‘Me’ Generation  1964  Moon landing  Creativity  Workaholism    flash it.” 
Vietnam War  Tolerance  Criticism 
1960s social revolution  Freedom  Innovativeness 
Self‐fulfillment  Advancement 
Materialism 
Generation X  X‐ers  1965‐ AIDS Individualism Free agency  “Whatever.”
244

13th Generation  1980  Oral contraceptive pills  Skepticism  Learning 


1973 oil crisis   Flexibility  Entrepreneurship 
Cold War  Control  Materialism  
Fun  Balance 
Generation Y  Millenium Generation  1981‐ Fall of the Berlin Wall Collectivism Balance  “Let’s make 
Generation Next  2001  MTV  Positivity  Passion  this world a 
Internet  Moralism  Learning   better place.” 
9/11‐War on Terror  Confidence   Security (not stability) 
Civic mindedness  Willingness to work 
 
 
 
245
Organizational  restructuring  and  downsizing  caused  the  first  cracks  in 
the  image  of  the  lifetime  career  with  one  employer  (Mirvis  and  Hall, 
1994). As a result, Baby Boomers are often portrayed as ‘free agents’ in 
the workplace, i.e. “radical individualists who advocate individual rights 
over  family  needs  and  the  rights  of  the  team  or  organization” 
(Kupperschmidt,  2000,  p.  69).  This  trend  of  employees  distancing 
oneself  from  organizations  persisted  in  Generation  X  and  Y.  These 
generations have learned not to take anything for granted, as the future 
is  unpredictable.  They  do  not  count  on  organizations  to  take  care  of 
them  and  offer  them  security  and  stability  in  their  career;  rather,  they 
tend  to  seek  out  employers  that  offer  challenging  jobs  and  sufficient 
training  so  that  they  gain  ‘employability’  in  the  job  market  –  thus 
acquiring career security in lieu of job security (Kupperschmidt, 2000). 
 
Finally,  Table  1  lists  the  general  and  work‐related  values  attributed  to 
the  different  generations  in  the  literature  (e.g.  de  Kort,  2004; 
Kupperschmidt, 2000; Lankard, 1995; Parker and Cusmir, 1991; Smola 
and Sutton, 2002). Employees from different generations are believed to 
have different value systems and react differently to common life events 
(Kupperschmidt,  2000).  Especially  the  experiences  and  events  that 
occur  during  the  formative  years  of  development  would  influence 
individuals during subsequent life experiences (Super et al., 1996). It is 
thus well possible that peoples’ beliefs about career and career success 
reflect  the  social  context  in  which  they  have  developed  as  adults 
(Sturges, 1999).  
 
Research questions 
 
Based  on  the  foregoing  review  of  the  literature  on  careers,  career 
success  and  generations,  the  following  research  questions  were 
formulated: 
 
(1) Do people from different generations have different career types? 
(2)  Does  the  importance  attached  to  organizational  security  differ  between 
generations? 
(3) Do people from different generations evaluate career success differently? 
 
The  first  research  question  stems  from  the  need  to  examine  whether 
there has, indeed, been a shift in society from ‘traditional, linear’ career 
types  to  ‘new,  non‐linear’  career  types  (Arthur  and  Rousseau,  1996; 
Briscoe and Hall, 2005; Brousseau et al., 1996; Collin, 1998; Littleton et 
al.,  2000);  the  second  research  question  scrutinizes  each  sub  sample’s 
preference  vis‐à‐vis  organizational  security,  which  is  also  believed  to 
differ  between  generations  (Kupperschmidt,  2000;  Mirvis  and  Hall, 
246
1994); finally, the third research question aims to examine whether the 
shared  social  understanding  of  what  career  success  means  has  shifted 
throughout  the  four  generations  under  study  in  this  paper  (Barley, 
1989;  Mirvis  and  Hall,  1994;  Sturges,  1999).  It  may  be  interesting  to 
note that the first research question focuses on the actual career of the 
participants,  while  the  second  and  the  third  research  question  centre 
around preferences and beliefs regarding career. 
 
Subsequently, the study will check if gender, functional level and career 
type  also  influence  how  people  make  career  success  evaluations. 
Women  have  been  found  to  view  career  success  more  as  a  process  of 
personal  development,  involving  interesting  and  challenging  work  and 
an  acceptable  work‐life  balance,  rather  than  salary  and  rank  that  have 
been  shown  to  correlate  with  career  satisfaction  for  men  (Asplund, 
1988;  Hennig  and  Jardim,  1978;  Nicholson  and  West,  1988;  Marshall, 
1984;  Russo  et  al.,  1991).  Women  managers’  different  ideas  about 
career success are likely  to be influenced both by  their socialization  as 
women  (Chodorow,  1974;  Gallos,  1989;  Gilligan,  1982)  and  the 
constraints they perceive as likely to affect their careers (Davidson and 
Cooper,  1992).  With  regard  to  functional  level,  two  outcomes  are 
possible.  On  the  one  hand,  people  who  are  successful  in  hierarchical 
terms may value the ‘traditional’ criteria for success less once they have 
achieved  them  –  in  accordance  to  goal  setting  theory  (Locke  and 
Latham,  1990).  However,  people  who  have  not  achieved  this  kind  of 
career success may still judge such criteria as important (Sturges, 1999). 
On the other hand, Festinger’s (1957) cognitive dissonance theory gives 
input to the opposite hypothesis: that people will adjust their personal 
definition  of  what  career  success  is  based  on  their  actual  situation, 
allowing  them  to  feel  good  about  themselves.  If  this  is  the  case,  then 
people  at higher  functional levels as well as people in more  traditional 
career  types  will  indicate  to  attach  more  importance  to  ‘traditional’ 
markers  of  success  than  people  at  lower  levels  or  in  ‘newer’  career 
types. Gunz and Heslin (2005) have raised a similar point in questioning 
how easy it is for people to develop their own personal standards in the 
face of pervasive social influences. They hypothesize that maybe people 
“manipulate  their  feelings  of  subjective  success  in  the  light  of  their 
objective situation” (p. 106). 
 
 
 
 
 
 
247
Methodology 
 
Participants 
 
An  online  survey  was  set  up,  investigating  four  different  generations’ 
beliefs  about  career  and  career  success.  In  order  to  obtain  the  largest 
possible  sample,  a  strategy  of  convenience  (snowball)  sampling  was 
employed.  All  students  and  all  employees  of  a  large  Belgian  university 
received  an  e‐mail  asking  for  their  participation  in  the  study.  Each  e‐
mail  contained  a  URL  to  the  online  survey.  Furthermore,  participants 
were informed of the fact that this study sought to collect data from four 
different  generations,  and  requested  to  forward  the  e‐mail  to  family 
members and acquaintances of various age groups.  
 
A total of 750 individuals completed the online survey in a valid manner. 
349  respondents  were  male  (47%)  and  401  were  female  (53%).  Of 
these 750 respondents, 437 (58%) indicated to be working at the time 
of  survey  administration.  For  the  career  type  variable  the  sample  size 
was 380 due to missing data.  
 
The average age of the study sample was 36.51 (sd = 15.51) – one must 
take  into  account,  however,  that  there  were  more  participants  in  the 
Generation Y respondent category than in the other categories. Further 
inspection of the background variables revealed that the majority of the 
study  sample  (67%)  had  completed  secondary  school  and  received 
some form of higher education (i.e. a Bachelor’s or Master’s degree). In 
the  subsample  of  the  ‘working  people’,  219  respondents  were  male 
(50%),  and  218  female  (50%);  64  were  employed  in  the  industrial 
sector  (15%),  186  in  services  (43%)  and  148  in  the  non‐profit  sector 
(34%) (39 unreported).  
 
In  Table  2,  more  detailed  information  is  provided  on  the  sample’s 
composition  by  crossing  the  ‘generation’  variable  with  gender, 
functional level and career type respectively.  
 
Procedure and measures 
 
Career  type.  In  analogy  with  Verbruggen  et  al.  (2007),  this  study 
used  several  variables  to  determine  which  of  six  career  types  fit  each 
respondent’s career best.  
 
 
Table 2. Cross‐tabulation of generation, gender, functional level and career type frequencies 
 
    Gendera    Functional levelb   
    Man  Woman    Not working Executive Management Employee  Worker  
Silent Generation  n  73  13    62 6 7 8 3 
    85%  15%    72% 7% 8% 9%  4%   
Baby Boomers  n  109  62    27 36 46 59 3 
    64%  36%    16% 21% 27% 35%  2%   
Generation X  n  91  114    23 12 41 126  3 
    44%  56%    11% 6% 20% 62%  2%   
Generation Y  n  76  212    202 1 3 81 1 
 
    26%  74%    70% 0% 1% 28%  0% 
Total  n  349  401    314 55 97 274  10 
 
    47%  53%    42% 7% 13% 37%  1% 
248

    Career typec 
    Bounded  Staying  Homeless Trapped Released Boundaryless 
Silent Generation  n  10  0  0 0 0 1 
    91%  0%  0% 0% 0% 9% 
Baby Boomers  n  89  12  7 8 3 3 
    73%  10%  6% 7% 3% 3% 
Generation X  n  76  23  43 2 6 16 
    46%  14%  26% 1% 4% 10% 
Generation Y  n  41  12  23 0 2 3 
    51%  15%  28% 0% 3% 4% 
Total  n  216  47  73 10 11 23  Notes. a, b n = 750; c n = 380 
    57%  12%  19% 3% 3% 6% 
 
 
 
249
The  authors  distinguish  between  bounded,  staying,  homeless,  trapped, 
released  and  boundaryless  careers  as  they  believe  simply  categorizing 
careers as ‘traditional’ versus ‘new’ would be overly simplistic – a point 
raised earlier in this paper (Arthur and Rousseau, 1996). The ‘bounded’ 
follow  the  traditional  career  path:  they  work  for  only  one  or  two 
organizations during the course of a lifetime, expect their employers to 
offer  them  job  security  and  standard  career  tracks,  and  derive  career 
satisfaction  from  pay,  promotion  and  status.  On  the  other  hand  of  the 
spectrum  are  the  ‘boundaryless’,  who  work  for  multiple  firms,  have 
transferable  skills,  manage  their  own  careers  and  feel  rewarded  by 
psychologically  meaningful  work.  Between  these  two  extremes  are  the 
other  four  career  types  that  are  all  characterized  by  some  sort  of 
discrepancy  between  their  career  aspirations  and  their  actual  career 
path. The ‘staying’ have changed employers regularly although they long 
for  security  and  stability.  However,  they  expect  this  discrepancy  to  be 
solved  as  they  believe  they  will  be  able  to  stay  with  their  current 
employer. The ‘homeless’ are in the same situation, but do not expect to 
be  able  to  stay  with  their  current  organization.  The  ‘trapped’  and  the 
‘released’  build  their  careers  under  opposite  circumstances:  they  have 
been  working  with  the  same  employer  over  a  long  period  of  time, 
although  they  aspire  change.  While  the  former  feel  they  are  ‘trapped’ 
with their current employer, the latter believe they will be able to solve 
the discrepancy between their aspirations and their actual career in the 
future.  
 
In order to classify every working respondent into one of the six career 
types,  we  first  categorized  them  as  having  either  stable  or  multiple‐
employer  careers.  In  order  to  do  so,  participants  to  the  survey  were 
asked how often they had changed employers during the last five years. 
Answer categories were ‘never’, ‘once’, ‘twice’ and ‘three times or more’. 
Following  Verbruggen  et  al.  (2007),  all  persons  who  indicated  to  have 
changed  employers  at  least  once  over  the  last  fast  years  were 
categorized  as  having  a  multiple‐employer  career.  Although  rather 
strict, this measure does represent an easy way to assess the extent to 
which a person has demonstrated mobility in his or her career.  
 
Secondly, we drew on the five Security/Stability items of Schein’s Career 
Orientation  Inventory  (1990)  so  as  to  obtain  a  score  of  (importance 
attached  to)  organizational  security.  Respondents  were  asked  to 
indicate their agreement to the items on a 7‐point Likert scale, ranging 
from  ‘Totally  disagree’  to  ‘Totally  agree’.  A  sample  item  is  “I  am  most 
fulfilled  in  my  work  when  I  feel  that  I  have  complete  financial  and 
employment security”. Internal consistency of the scale proved high (α= 
250
0.83).  This  variable  was  included  in  the  analyses  as  both  a  continuous 
variable (organizational security scale score) and a categorical variable 
(high  vs.  low  value  attached  to  organizational  security,  which  was 
operationalized  by  recoding  the  scale  scores  to  10  and  categorizing 
values lower than or equal to 5 as ‘low’ and higher than 5 as ‘high’).  
 
Thirdly, the survey also assessed respondents’ expectation to leave their 
employer within the next five years. Three items assessed this variable: 
“Do  you  expect  to  change  employers  in  the  next  five  years?”,  “Do  you 
expect to become self‐employed in the next five years?”, “Do you expect 
to  become  unemployed  in  the  next  five  years?”.  Answering  categories 
were  ‘yes’  or  ‘no’.  A  positive  answer  to  at  least  one  of  these  questions 
would  lead  to  the  categorization  ‘expects  to  leave’  (as  opposed  to 
‘expects to stay’). 
 
Combining  the  three  measures  above  then  made  it  possible  to 
determine each participant’s career type as described in Verbruggen et 
al. (2007): 
 
1.  Bounded:  Stable  career;  High  importance  attached  to  organizational 
security  
2.  Staying:  Multiple‐employer  career;  High  importance  attached  to 
organizational security; Expects to stay 
3.  Homeless:  Multiple‐employer  career;  High  importance  attached  to 
organizational security; Expects to leave 
4.  Trapped:  Stable  career;  Low  importance  attached  to  organizational 
security; Expects to stay 
5.  Released:  Stable  career;  Low  importance  attached  to  organizational 
security; Expects to leave 
6.  Boundaryless:  Multiple‐employer  career;  Low  importance  attached  to 
organizational security 
 
Evaluation of career success. In order to assess whether people from 
different  generations  hold  different  beliefs  about  what  ‘career  success’ 
entails,  a  series  of  vignettes  was  incorporated  in  the  online  survey. 
Vignette (or: factorial) surveys  are  generally considered to be  effective 
for assessing judgments on a wide array of complex issues (Ganong and 
Coleman,  2006),  and  to  have  high  internal  and  external  validity 
(Ludwick and Zeller, 2001; Lee and Stolte, 1994). 
 
The  design  of  this  part  of  the  study  was  a  4  (generation:  Silent 
Generation,  Baby  Boomers,  Generation  X  or  Generation  Y)  x  5  (career 
feature:  functional  level,  salary,  number  of  promotions,  promotion 
speed  or  satisfaction)  x  2  (career  feature  level:  high  or  low)  mixed 
design,  in  which  ‘generation’  was  the  between‐subjects  factor  and  the 
‘career  feature’  and  the  ‘career  feature  level’  factors  were  within‐
251
subject.  Functional  level,  salary,  number  of  promotions,  promotion 
speed  and  satisfaction  were  chosen  as  the  career  features  since  these 
are, by far, the most frequently reported measures of career success in 
literature  (Arthur  et  al.,  2005;  Greenhaus  et  al.,  1990;  Hurley  & 
Sonnenfeld,  1998;  Hurley‐Hanson  et  al.,  2005;  Judge  et  al.,  1995; 
Kirchmeyer,  1998;  Orpen,  1998;  Seibert  et  al.,  1999;  Tharenou,  1999; 
Wayne  et  al.,  1999).  While  functional  level,  salary,  number  of 
promotions and promotion speed represent rather ‘traditional’ markers 
of career success (Adamson et al., 1998), satisfaction is the postmodern 
career success criterion of choice, as it may incorporate a whole range of 
different opinions as to what career success means (Arthur et al., 2005; 
Bozionelos, 2004; Greenhaus et al., 1990).  
 
Participants were told that the vignette task was designed to assess how 
people  evaluate  other  people’s  career  success  based  on  a  limited 
amount of information available about that person’s career. Specifically, 
they  were  instructed  to  imagine  that  they  would  meet  the  people 
described in each of the vignettes, and so infer how they would evaluate 
their  career  success  in  a  real‐life  situation.  Their  final  evaluation  of 
career  success  then  needed  to  be  indicated  on  a  7‐point  Likert  scale 
going from “This person’s career is, in your opinion, very unsuccessful” 
to  “This  person’s  career  is,  in  your  opinion,  very  successful”.  The 
instructions  stressed  that  the  study  was  interested  in  their  own 
personal opinions, rather than there being an optimal solution of some 
sort.  Furthermore,  the  importance  of  subjectivity  was  underlined 
further  by  adding  that  participants  should  rely  on  one’s  own 
interpretation of what ‘high’ or ‘low’ is (i.e. in the career feature levels). 
We  came  to  this  conclusion  as  we  found  it  very  difficult  to  establish 
more objective levels (for instance, of salary) that would yield the same 
notion  of  ‘high’  and  ‘low’  across  all  participants  –  unless  very  extreme 
values would be used, which would, ultimately, not improve the validity 
of  the  vignette  task.  The  career  features  themselves,  however,  were 
clearly  defined  in  the  instructions,  in  order  to  ensure  that  all 
participants  would  make  their  evaluations  departing  from  the  same 
‘mindset’. Functional level was defined as ‘the level this person holds in 
the  hierarchy  of  the  organization  by  which  he  or  she  is  employed,  as 
associated  with  a  certain  status,  power  and  influence’;  salary  as  ‘the 
gross  year  salary  this  person  receives  for  his  or  her  work’;  number  of 
promotions as ‘the number of times this person has been promoted to a 
higher  functional  level’;  promotion  speed  as  ‘the  pace  by  which  this 
person  has  climbed  the  ladder  as  opposed  to  stagnating  in  the  same 
function for a long time’; and satisfaction as ‘the satisfaction this person 
feels when he or she considers all aspects of his or her career’. Finally, 
252
the  two  most  extreme  vignettes  were  presented  and  participants  were 
instructed  to  rate  all  other  combinations  as  lying  between  these  two 
extremes: 
 
“This person is at a low functional level and receives a low salary,  
has  gotten  promoted  a  small  number  of  times,  in  a  low  pace,  and  is 
dissatisfied” 
 
“This person is at a high functional level and receives a high salary, has gotten 
promoted a large number of times, in a fast pace, and is satisfied” 
 
The 32 vignettes that resulted from all possible combinations of career 
features  and  career  feature  levels  were  presented  in  a  randomly 
generated order to each participant (Ployhart et al., 1999).  
 
Background variables. All respondents were instructed to designate 
their  year  of  birth.  This  data  was  then  recoded  into  the  categorical 
‘generation’  variable  using  the  birth  year  ranges  listed  in  Table  1. 
Furthermore,  gender,  educational  level,  work  status  (working  or  not 
working),  functional  level  and  employment  sector  were  surveyed  in  a 
multiple‐choice format.  
 
Results 
 
Do people from different generations have different career types? 
 
In  order  to  assess  whether  working  respondents  from  different 
generations  reported  having  different  career  types,  the  data  was 
analyzed with a Pearson chi‐square test of independence. This analysis 
technique  was  the  most  appropriate  as  both  the  ‘generation’  variable 
and the ‘career type’ variable were categorical. The chi‐square test (n = 
357)  yielded  a  significant  value  (χ²  (15)  =  52.91,  p  <  .001),  indicating 
that  the  percentage  of  working  people  in  each  career  type  differed  by 
generation.  However,  as  can  be  inferred  from  the  cross  tabulation  in 
Table 2, the assumption of adequate cell frequencies was violated – over 
50%  of  cells  had  an  expected  count  lower  than  5  –  and  so  we  must 
interpret these findings  with caution. Moreover, the cross tabulation is 
difficult  to  interpret  as  the  four  generations  differ  strongly  in  terms  of 
sample  size.  Nonetheless,  we  can  observe  some  trends.  Looking  at  the 
overall sample, it is clear that the vast majority of working respondents 
(57%) still have bounded career types; 13% are staying, 19% homeless, 
3% trapped, 2% released and 6% boundaryless.  
 
 
 
253
Does the importance attached to organizational security differ between 
people from different generations? 
 
Analysis  of  the  study  sample’s  frequencies  and  descriptives  revealed 
that,  for  the  continuous  7‐point  organizational  security  variable,  the 
overall  mean  was  5.33  (sd  =  1.40);  the  means  and  standard  deviations 
per generation are reported in Figure 1. With regard to the categorical 
variable,  the  data  showed  that  90%  of  the  overall  sample  attached 
importance  to  organizational  security.  Looking  at  the  data  through  a 
generational lens, we found that 92% of the Silent Generation attached 
importance to organizational security, as did 85% of the Baby Boomers, 
84% of the Generation X‐ers and 95% of the Generation Y‐ers.  
 
A one‐way ANOVA was conducted in order to determine whether people 
from  different  generations  attach  significantly  different  amounts  of 
value  to  organizational  security.  As  the  assumption  of  homogeneity  of 
variances  was  violated  (possibly  due  to  the  fact  that  the  sample  size 
differed  strongly  per  generation)  only  Scheffé  corrected  values  are 
reported.  There  was,  indeed,  a  small  significant  difference  (η²p  =  0.04) 
between  generations (F(3,717) =  10.04, p  < .001). Figure  1 shows how 
the  generations’  mean  scores  differed.  Post‐hoc  tests  revealed  that 
respondents  from  the  Silent  Generation  attached  significantly  more 
importance  to  organizational  security  than  did  Baby  Boomers  and 
people from Generation  X. Furthermore, Generation X had significantly 
lower scores than Generation Y. There were no differences between the 
Silent Generation and Generation Y, nor between the Baby Boomers and 
Generation  X.  There  were  no  significant  differences  between  men  and 
women. We did find a small significant effect (η²p = 0.04) for functional 
level, where inspection of the data revealed that people at management 
levels  attach  significantly  less  importance  to  organizational  security 
than  do  employees  and  workers.  Differences  in  terms  of  importance 
attached to organizational security were not tested for career type, as it 
is one of the variables used to determine career type, as described in the 
Methodology section.  
 
Do people from different generations evaluate career success differently? 
 
Several  one‐way  repeated  measures  ANOVAs  were  conducted  to 
determine  how  the  different  groups  of  respondents  (the  four 
generations,  the  two  sexes,  the  four  functional  levels,  the  six  career 
types) came to their evaluation of career success for each vignette. Main 
effects (‘within’ and ‘between’) and interaction effects were scrutinized. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
254

 
 
 
 
 
 
 
 
 
 
 
Figure 1. Means per generation for importance attached to organizational security (OS) 
 
255
Figure 2 offers a visual presentation of the effects found in the repeated 
measures ANOVAs conducted for each generation. Below the graphs are 
the mean career success scores for each career feature (functional level, 
salary, number of promotions, promotion speed and satisfaction) at the 
‘low’ and the ‘high’ level, as well as the F and η²p (partial Eta squared) 
values  for  each  career  feature’s  main  effect.  All  career  features 
demonstrated  significant  main  effects  (p  <  .001)  for  each  generation; 
taking into consideration the η²p values for each effect we can then infer 
the proportion of total variability in career success score attributable to 
each  factor.  When  we  rank  these  main  effects  from  largest  to  smallest 
we  achieve  the  following  order  for  the  overall  sample,  and  for  each 
generational  subsample  as  well  (although  for  the  Silent  Generation, 
satisfaction  and  salary,  and  functional  level  and  promotions  had  equal 
effect sizes) :  
 
1. Satisfaction (η²p ranging from 0.69 to 0.81; 0.72 in the overall sample) 
2. Salary (η²p ranging from 0.69 to 0.74); 0.68 in the overall sample) 
3. Functional level (η²p ranging from 0.27 to 0.53; 0.38 in the overall sample) 
4. Number of promotions (η²p  ranging from 0.27 to 0.41; 0.27 in the overall 
sample) 
5. Promotion speed (η²p ranging from .20 to .35; 0.22 in the overall sample) 
 
Following  the  widely  accepted  guidelines  of  Cohen  for  partial  Eta 
squared  values  (1988,  1992),  all  of  these  main  effects  can  be  classified 
as large (≥ .14).  
  
Closer inspection of the graphs and the η²p values reveals that the main 
effects get  more prominent with each generation.  Furthermore, for the 
overall  sample,  all  factors  except  salary  display  significant  interaction 
effects with generation. However, these effects are quite small, with η²p 
values  around  .02.  No  between‐subjects  effects  were  found  for 
generation (F(3, 708) =  1.61, ns), nor for gender (F(1, 710)  = 1.38, ns), 
nor for career type (F(5, 351) = 0.28, ns) or functional level (F(4, 406) = 
1.01, ns).  
 
Besides  looking  into  the  main  effects  of  the  career  feature  factors  and 
the  between‐subjects  effects  for  generation,  gender,  career  type  and 
functional  level,  we  carefully  examined  all  interaction  effects  in  the 
study’s  data  –  as  such  effects  may  be  warnings  that  main  effects  may 
lack  generality  (Stevens,  2002).  However,  closer  inspection  of  the 
significant  interaction  effects  (at  p  <  .05)  reported  in  Table  3  revealed 
that  all  of  the  interaction  effects  were  ordinal;  this  type  of  effects 
generally  does  not  undermine  or  alter  the  findings  of  the  main  effects 
(Stevens, 2002).  
 
 
 
 
 
 
 
 
 
 
 
 
 
256

 
 
 
 
 
 
 
 
 
 
 
Figure 2. Factorial plot for career success (CS) score as a function of career feature (CF) and career feature level (CFL) per generation 
257
Ordinal  effects  signify  that  the  effect  of  one  variable  intensifies  the 
other.  For  instance,  a  strong  interaction  effect  was  found  for  each 
generation  between  functional  level  and  salary,  which  is  to  be 
interpreted as follows: “the effect of a high salary, combined with a high 
functional level, on career success evaluation is larger than just the sum 
of both separate effects (i.e. a high salary always leads to a higher career 
success score than does a low one, but this effect is blown up at a higher 
functional  level”).  Inspection  of  Table  3,  however,  discloses  that  the 
effect  sizes  of  these  interactions  are  much  smaller  than  those  of  the 
main  effects  reported  above.  It  is  therefore  questionable  whether 
interpreting these interactions in addition to the main effects would add 
much valuable information (Stevens, 2002).  
 
Table 3. Significant interactions from the repeated measures ANOVAs (p < .05) 
 
Sample  Interaction*  df  F η²p 
Overall sample  Fl x Sal 1,708 73.29 .09
  Fl x Sal x Sat 46.03 .06
  Fl x Sal x Pr 8.40 .01
  Ps x Sat 6.10 .01
  Fl x Sal x Ps 5.74 .01
  Sal x Ps 4.83 .01
  Sal x Ps x Sat 4.26 .01
Silent Generation Fl x Sal x Ps 1,80 8.38 .10
  Fl x Sal 4.50 .05
Baby Boomers  Fl x Sal 1,158 21.59 .12
  Fl x Sal x Sat 15.24 .09
  Pr x Ps 4.22 .03
Generation X  Fl x Sal 1,188 41.06 .18
  Fl x Sal x Sat 19.65 .10
  Fl x Ps x Sat 4.49 .02
  Fl x Pr x Ps 4.16 .02
Generation Y   Fl x Sal 1,282 30.03 .10
  Fl x Sal x Pr 16.81 .06
  Sal x Sat 13.14 .05
  Fl x Sal x Sat 14.76 .05
  Pr x Ps x Sat 6.21 .02
  Ps x Sat 6.09 .02
Note.  *  Fl  =  functional  level;  Sal  =  salary;  Pr  =  number  of  promotions;  Ps  =  promotion 
speed; Sat = satisfaction. 
 
Discussion 
 
The  current  study  aimed  to  investigate  whether  people  from  four 
different  generations  (Silent  Generation,  Baby  Boomers,  Generation  X 
and  Generation  Y)  hold  different  beliefs  about  the  meanings  of  ‘career’ 
258
and  ‘career  success’.  Three  basic  research  questions  arose  from  our 
review  of  the  literature:  do  people  from  different  generations  have 
different  career  types;  does  the  importance  attached  to  organizational 
security  differ  between  people  from  different  generations;  and  do 
people from different generations evaluate career success differently. 
 
The answer to the first research question, about career type, appears to 
be  a  cautious  ‘yes’.  Questions  may  be  raised  about  the  chi‐square 
analysis  performed.  There  were,  however,  some  observable  trends. 
Remarkably, the three most prevalent career types in the study sample 
(bounded,  staying  and  homeless)  have  in  common  that  they  all  attach 
high importance to organizational security, indicating that the majority 
of  respondents  did  find  this  very  important  (these  results  will  be 
discussed  below  in  light  of  the  third  research  question).  Furthermore, 
the  relative  decrease  of  bounded  career  types  with  generation  and 
simultaneous  increase  of  staying  and  homeless  career  types  (Table  2) 
may  be  evidence  of  the  fact  that  although  certain  career  aspirations 
(such as achieving organizational security) may not have changed much, 
career  reality  has  –  causing  a  shift  from  traditional,  bounded  career 
types to staying  and homeless career  types (where there is a  multiple‐
employer  career,  but  still  a  longing  for  stability  and  security).  The 
numbers  for  trapped,  released  and  boundaryless  career  types,  on  the 
other hand, appeared rather inconclusive.    
 
The  current  findings  are  consistent  with  those  of  Verbruggen  et  al. 
(2007),  who  found  that  that  only  6  percent  of  the  957  respondents  in 
their  study  had  a  boundaryless  career,  while  60  percent  still 
experienced  bounded  careers.  Despite  the  growing  popularity  of  the 
boundaryless career concept in literature (Arthur and Rousseau, 1996; 
Briscoe  and  Hall,  2005;  Collin,  1998),  several  authors  are  now  saying 
that the traditional career is ‘far from dead’ and that one must be careful 
when making claims about the speed and inevitability of the shift from 
traditional to ‘new’ career types (e.g. Guest and Mackenzie Davey, 1996; 
Verbruggen et al., 2007; Walton and Mallon, 2004).  
 
The results for the second research question demonstrate that, if there 
is  indeed  something  as  a  generational  ‘shift’  in  beliefs  about  career,  it 
does not necessarily have to be a linear one. Instead we found that a U‐
shaped  function  better  captures  the  relationship  between  generation 
and  importance  attached  to  organizational  security,  implying  that  this 
‘old‐fashioned’ belief or preference about career (i.e. that it is desirable 
to  work  for  an  organization  that  can  offer  long‐term  security  and 
stability)  does  not  only  prevail  in  the  Silent  Generation,  but  also  in 
259
Generation  Y,  the  youngest  generation  of  all.  A  possible  explanation  is 
that the Silent Generation grounds its beliefs in the past (a career reality 
that  is  long  gone),  and  the  Baby  Boomers  and  Generation  X  in  the 
present (as they represent the bulk of the present‐day workforce), while 
Generation  Y  (for  the  most  part)  is  just  dreaming  about  their  future 
career (not yet having been confronted with career reality today). It will 
be interesting to see future studies of Generation Y come about, as they 
are only just now entering the workforce and not much is known about 
their  work  values  and  career  preferences  (Jennings,  2000;  Smola  and 
Sutton, 2002).  
 
The third research question was operationalized by offering a series of 
vignettes  to  respondents  from  different  generations,  which  contained 
abstract  descriptions  of  a  fictitious  person’s  career  (combining 
information  about  functional  level,  salary,  number  of  promotions, 
promotion  speed  and  satisfaction).  No  significant  differences  were 
found between generations (nor between sexes, nor between people at 
different functional levels or in different career types) when it comes to 
how they evaluate career success. The within effects, on the other hand, 
did  have  a  story  to  tell.  For  the  overall  sample  as  well  as  for  each 
generational  subsample  we  found  quite  large  main  effects  for  each 
career  feature,  ‘satisfaction’  being  the  overriding  criterion  each  time. 
This would mean that people of all generations, when evaluating other 
people’s  career  success,  would  rely  primarily  on  those  people’s  own 
evaluations of their careers. However, it is noticeable that for the Silent 
Generation, the effect sizes for the main effects of satisfaction and salary 
are equal (η²p = 0.69). Furthermore, the effect sizes of each of the main 
effects, including satisfaction, increase with generation. This could fit in 
with the literature stating that ‘free career agency’ took off in the era of 
the Baby Boomers, and continued in Generation X and Y, when all career 
securities  disappeared  (Kupperschmidt,  2000)  –  which  would  imply 
that, although satisfaction is quite a universal concept, its prominence in 
career success evaluation increased as the traditional markers of career 
disappeared.  It  is  clear,  however,  that  also  salary  continues  to  play  an 
important role when people evaluate other people’s careers.  
 
Limitations and future research 
 
Although  vignette  studies  have  proven  to  be  effective  for  assessing 
judgments on a wide array of complex issues and to have high internal 
and  external  validity  (Ganong  and  Coleman,  2006;  Ludwick  and  Zeller, 
2001;  Lee  and  Stolte,  1994),  they  are  not  without  disadvantage.  One 
possible  problem  is  the  tendency  of  respondents  to  process  the 
260
information  in  the  vignettes  less  carefully  and  effectively  than  they 
would  under  real‐life  conditions  (Lee  and  Stolte,  1994).  Vignette 
characteristics  such  as  the  length,  complexity  and  cognitive  load  of  its 
content will, then, affect  the quality of the data  gathered (Esposito and 
Jobe, 1991; Krosnick, 1991). In the current study, the vignettes were not 
very long and simple language was used. However, the imagery that was 
evoked was quite abstract, so that a fairly high cognitive load was placed 
on the study participants (Krosnick, 1991; Lee and Stolte, 1994). On the 
other  hand,  this  load  was  the  same  for  each  participant  as  they  all 
received the exact same vignettes. Furthermore, in a vignette study one 
must  always  choose,  at  least  to  some  extent,  between  experimental 
control  and  realism.  The  use  of  vignettes  restricts  the  amount  and 
richness  of  information  that  can  be  offered  to  participants,  so that  it  is 
often not possible to include all variables that may play a role in real‐life 
decision  making  on  the  topic  of  study.  Therefore,  respondents  often 
communicate  finding  the  task  very  difficult,  as  they  feel  it  cannot  fully 
capture their opinions and beliefs (Furnham and Petrides, 2006).  
 
Perhaps  more  differences  would  have  been  found  between  the  four 
generations  under  study  if  a  broader  range  of  possible  career  success 
criteria  would  have  been  included  in  the  survey.  There  may  be  other 
criteria that are very relevant to real‐life career success evaluation that 
have  not  been  incorporated  in  this  study.  It  would  be  particularly 
interesting  to  explore  the  dimensions  of  career  satisfaction  (Arthur  et 
al.,  2005)  and  to  which  extent  these  differ  between  different  (sub) 
populations.  Furthermore,  future  studies  should  try  to  attain  a  more 
balanced sample (so that sample sizes per generation would differ less) 
and  more  ‘working’  people,  so  that  the  data  for  career  type  would  be 
more reliable.  
 
Further research is necessary in order to  establish the  extent to which 
evolving  work  values  in  society  translate  into  shifts  in  beliefs  about 
career  and  career  success.  It  would  be  interesting  to  see  if  people  are 
more influenced in this respect by the experiences they had growing up, 
or  by  the  current  economic  climate.  In  the  former  case,  one  would 
expect  to  find  inter‐generational  differences.  Another  viewpoint  to 
consider  in  this  respect  is  that  of  life‐span,  life‐stage  developmental 
theory, which suggests that life‐style priorities change with age and that 
consequently,  individuals'  career  interests  and  aspirations  evolve  over 
time  (Super  et  al.,  1996).  A  longitudinal  design  would  be  more 
appropriate  to  scrutinize  such  developmental  processes.  In  the  latter 
case,  all  people  in  the  present‐day  workforce  would  hold  similar 
opinions.  
261
Conclusion 
 
Previous studies have focused mostly on asking people what they feel is 
important  in  their  own  careers  (e.g.  Cangemini  and  Guttschalk,  1986; 
Finegold  and  Mohrman,  2001;  Friedman  and  Greenhaus,  2000).  It  is 
quite  logical  that  they  have  concluded  that  an  increasing  number  of 
idiosyncratic  opinions  about  what  career  success  means  is  arising. 
However,  these  studies  do  not  paint  the  whole  picture,  as  they  only 
focus on the internal aspect of career evaluation. We have attempted to 
avoid  somewhat  individuals  projecting  their  feelings  about  their  own 
career on our vignette task by instructing them to judge other peoples’ 
careers.  We  used  tangible  (functional  level,  salary,  number  of 
promotions,  promotion  speed)  as  well  as  intangible  criteria 
(satisfaction)  of  career  success.  Satisfaction  came  out  as  being  ‘the’ 
overriding criterion by which people evaluate others’ career success.  
 
If our design accurately presented the reality of career evaluation, then, 
this  would  mean  that  the  shared  social  understanding  agreed  upon  by 
all  generations  tends  to  validate  the  internal  evaluations  individuals 
make  about  their  own  careers,  no  matter  what  their  objective 
characteristics (‘if they are satisfied with their careers, who am I to say 
that  they  are  not  successful?’).  This  would,  then,  be  evidence  that  the 
‘new’ career (in which everything goes) is well embedded in the shared 
social  understanding  of  career  and  career  success.  But  is  this  true  in 
reality? We see two alternative explanations for our findings. The first is 
that the lack of information presented in the vignettes causes people to 
rely  upon  the  judgement  of  the  fictitious  career  holder  in  the  vignette 
(i.e.  their  career  satisfaction);  indeed,  the  instructions  did  say  that 
satisfaction should be seen as ‘the satisfaction this person feels when he 
or she considers all aspects of his or her career’. In this case, our finding 
that satisfaction is the overriding criterion for career success evaluation 
might  partly  be  an  artefact  of  the  study  design.  The  second  alternative 
explanation  is  that  processes  of  social  desirability  (Phillips  and  Clancy, 
1972) and cognitive dissonance reduction (Festinger, 1957) play a role 
when people make evaluations of career success. Indeed, agreeing to the 
statement  “A  cleaning  lady  that  is  satisfied  with  her  career  is  more 
successful than a CEO that is not satisfied with her career” would seem 
to be induced by social desirability, and thus may not accurately reflect 
how  one  would  really  evaluate  both  careers  in  a  real‐life  setting. 
Cognitive  dissonance  reduction,  on  the  other  hand,  would  occur  when 
people would project their own career situation onto the vignettes they 
are  asked  to  rate.  What  would  happen  is,  then,  that  people  would 
262
manipulate  their  feelings  of  what  career  success  means  in  the  light  of 
their own objective situation (since no one likes to feel unsuccessful). 
 
If  anything,  we  must  call  into  question  whether  the  most  frequently 
used  operationalizations  of  career  success  (i.e.  functional  level,  salary, 
promotions)  are  still  adequate.  How  will  we  study  this  variable  in  the 
future  when  there  is  increasing  evidence  that  satisfaction  is  generally 
considered  as  the  overriding  criterion  for  evaluating  career  success 
(Arthur  and  Rousseau,  1996;  Arthur  et  al.,  2005;  Bozionelos,  2004; 
Cangemini  and  Guttschalk,  1986;  Finegold  and  Mohrman,  2001; 
Friedman  and  Greenhaus,  2000)?  This  notion  may  well  be  accepted  by 
many  postmodern  career  theorists,  but  recent  studies  using  career 
success  as  a  dependent  variable  are  continuing  to  operationalize  it  in 
terms  of  the  traditional  markers  of  career  success  (e.g.  Seibert  et  al., 
1999; Tharenou, 1999; Wayne et al., 1999). Using career satisfaction as 
a  dependent  variable  would  most  likely  yield  very  different  study 
outcomes,  as  objective  and  subjective  measures  of  career  success  have 
been found to be only moderately related (Arthur et al., 2005; Boudreau 
et  al.,  2001).  Each  of  these  facts  must  be  considered  when  conducting 
studies  about  career  success  in  the  future,  as  they  might  gravely  affect 
their (construct and external) validity. 
 
Although studies about the career success construct in itself may not be 
directly  applicable  to  practice,  they  might  help  organizations  design 
career management programs and individuals manage their own career 
processes. Organizations in the present day still contribute considerably 
to the persistence of the more traditional (linear) career concept by the 
depreciation  and  inadequate  rewarding  of  those  in  non‐linear  careers, 
e.g.  experts  in  ‘lateral’  career  tracks,  or  those  who  take  a  break  from 
paid employment for personal reasons. However, in the current reality, 
job  security,  advancement  opportunities  and  regular  salary  increases 
are  no  longer  available  to  all  employees  (Rousseau,  1990). 
Consequently,  a  more  diversified  perception  of  what  career  success 
might  mean  to  different  groups  of  employees  (and  how  their  careers 
should be managed accordingly) becomes inevitable. 
 
References 
 
Adamson, S.J., Doherty, N. and Viney, C. (1998), “The meanings of career revisited: Implications for 
theory and practice”, British Journal of Management, Vol. 9, pp. 251‐59. 
Arthur, M.B, Hall, D. and Lawrence, B. (1989), "Generating new directions in career theory: the case 
for  a  transdisciplinary  approach",  in  Arthur,  M.B.,  Hall,  D.T.  and  Lawrence,  B.S.  (Eds), 
Handbook of Career Theory, Cambridge University Press, Cambridge, pp.7‐25. 
Arthur, M.B., Khapova, S.N. and Wilderom, C.P.M. (2005), “Career success in a boundaryless career 
world”, Journal of Organizational Behavior, Vol. 26, pp. 177‐202.  
263
Arthur,  M.B.  and  Rousseau,  D.M.  (1996),  The  boundaryless  career:  A  new  employment  principle  for 
new organizational era, Oxford University Press, New York., NY.  
Asplund,  G.  (1988),  Women.  managers:  Changing  organizational  cultures,  John  Wiley  &  Sons,  New 
York, NY. 
Barley,  S.  R.  (1989),  “Careers,  identities,  and  institutions:  the  legacy  of  the  Chicago  School  of 
Sociology”, in Arthur, M. B., Hall, D. T. and Lawrence, B. S. (Eds), Handbook of Career Theory, 
Cambridge University Press, Cambridge, pp. 41‐59.  
Bozionelos, N. (2004), “The relationship between disposition and career success: A British study”, 
Journal of Occupational and Organizational Psychology, Vol. 77, pp. 403‐20.  
Briscoe,  J.P.  and  Hall,  D.T.  (2005),  “The  interplay  of  boundaryless  and  protean  careers: 
Combinations and implications”, Journal of Vocational Behavior, Vol. 69, pp. 4‐18. 
Brousseau,  K.,  Driver,  M.J.,  Eneroth,  K.  and  Larsson,  R.  (1996),  “Career  Pandemonium:  realigning 
organizations and individuals”, Academy of Management Executive, Vol. 10, pp. 52‐66. 
Cangemini,  J.P.  and  Guttschalk,  G.E.  (1986),  “What  employees  really  want  from  their  jobs”, 
Psychology: A Journal of Human Behavior, Vol. 23, pp. 57‐61. 
Chen,  C.  (1997),  “Career  projection:  Narrative  in  context”,  Journal  of  Vocational  Education  and 
Training, Vol. 49, pp. 311‐26. 
Chodorow, N. (1974), “Family structure and feminine personality”, in Resaldo, M. Z. and Lamphere, 
L. (Eds), Women, Culture and Society, Stanford, University Press, Stanford, CA. 
Cohen, J. (1988), Statistical power analysis for the behavioral sciences (2nd ed.), Erlbaum, Hillsdale, 
NJ. 
Cohen, J. (1992), “A power primer”, Psychological Bulletin, Vol. 112, pp.155‐9.  
Collin, A. (1998), “New challenges in the study of career”, Personnel Review, Vol. 27, pp. 412–25.  
Collin, A. and Young, R.A. (2000), The future of career, Cambridge University Press, Cambridge. 
Davidson,  M.J.  and  Cooper,  C.L.  (1992),  Shattering  the  Glass  Ceiling:  the  Woman  Manager,  Paul 
Chapman, London. 
De  Kort,  L.  (2004),  “Generations  at  work”,  Retrieved  September  7th,  2007  from 
www.aimnt.com.au/ntatwork/generations_at_work.pdf. 
DeMeuse,  K.P.  and  Tornow,  W.W.  (1990),  “The  Tie  That  Binds—Has  Become  Very,  Very  Frayed”, 
Human Resource Planning, Vol. 13, pp. 203‐13. 
Eby, L. T., Butts, M., and Lockwood, A. (2003), “Predictors of success in the age of the boundaryless 
career”, Journal of Organizational Behavior, Vol. 24, pp. 689‐708. 
Esposito, J.L. and Jobe, J.B. (1991), "A General Model of the Survey Interaction Process", Proceedings 
of the Census Bureau’s 1991 Annual Research Conference, pp. 537‐60. 
Evetts, J. (1992), “Dimensions of Career: Avoiding Reification  in the Analysis of Change”, Sociology, 
Vol. 26, pp. 1‐21. 
Festinger, L. (1957), A Theory of Cognitive Dissonance, Row, Perterson & Company, Evanston, IL. 
Finegold, D. and Mohrman, S.A. (2001), “What Do Employees Really Want? The Perception vs. the 
Reality,”  Paper  presented  at  the  annual  meeting  of  the  World  Economic  Forum,  Davos, 
Switzerland.  
Friedman, S.D. and Greenhaus, J.H. (2000),  Allies  or  Enemies? How Choices about Work and Family 
Affect the Quality of Men’s and Women’s Lives, Oxford University Press, New York, NY.  
Furnham,  A.  and  Petrides,  K.V.  (2006),   “Deciding  on  promotions  and  redundancies:  Promoting 
people by ability, experience, gender, and motivation”, Journal of Managerial Psychology, Vol. 
21, pp. 6‐18. 
Gallos,  J.V.  (1989),  "Exploring  women’s  development:  implications  for  career  theory,  practice  and 
research", in Arthur, M. B., Hall, D. T. and Lawrence, B. S. (Eds), Handbook of Career Theory, 
Cambridge University Press, Cambridge. 
Ganong,  L.H.  and  Coleman,  M.  (2006),  “Multiple  Segment  Factorial  Vignette  Designs”,  Journal  of 
Marriage and Family, Vol. 68, pp. 455‐ 68. 
Gilligan,  C.  (1982),  In  a  Different  Voice:  Psychological  Theory  and  Women’s  Development,  Harvard 
University Press, Cambridge, MA.  
Greenhaus,  J.H.  (2003),  "Career  dynamics",  in  Borman,  W.C.,  Ilgen,  D.R.  and  Klimoski,  R.J.  (Eds), 
Handbook  of  Psychology,  Industrial  and  Organizational  Psychology,  Wiley,  New  York,  NY, 
pp.519‐40.  
Greenhaus, J.H., Parasuraman, S. and Wormley, W.M. (1990), "Race, Organizational Experiences, and 
Career Outcomes," Academy of Management Journal, Vol. 33, pp. 64‐86.  
Guest, D., Mackenzie Davey, K. (1996), "Don’t write off the traditional career", People Management, 
No. February, pp. 22‐5.  
264
Gunz,  H.P.  and  Heslin,  P.A.  (2005),  "Reconceptualizing  career  success",  Journal  of  Organizational 
Behavior, Vol. 26, pp.105‐11. 
Hennig, M. and Jardim, A. (1978), The Managerial Woman, Marion Boyars, London. 
Herr, E.  L., and Cramer, S. H.  (1992), Career guidance and  counseling through the life span (4th ed.), 
HarperCollins, New York, NY. 
Heslin,  P.A.  (2003),  “Self‐  and  other‐referent  criteria  of  career  success”,  Journal  of  Career 
Assessment, Vol. 11, pp. 262–. 286.  
Higgins,  M.  C.  (2001),  “Changing  careers:  The  effect  of  social  context”,  Journal  of  Organizational 
Behavior, Vol. 22, pp.595‐618. 
Hurley,  A.E.  and  Sonnenfeld,  J.A.  (1998),  “The  Effect  of  Organizational  Experience  on  Managerial 
Career Attainment in an Internal Labor Market”,  Journal of Vocational Behavior,  Vol. 19, pp. 
172‐90.   
Hurley‐Hanson, A. E., Wally, S., Segrest Purkiss, S.L. and Sonnenfeld, J.A. (2005), “The changing role 
of education on managerial career attainment”, Personnel Review, Vol. 34, pp. 517‐33.  
 Jaskolka,  G.,  Beyer,  J.M.  and  Trice,  H.M.  (1985),  "Measuring  and  predicting  managerial  success", 
Journal of Vocational Behavior, Vol. 26, pp.189‐205. 
Jennings, A.T. (2000), "Hiring Generation‐X", Journal of Accountancy, Vol. 189, pp. 55‐9.  
Judge,  T.A.,  Cable,  D.M.,  Boudreau,  J.W.  and  Bretz,  R.D.  (1995),  “An  empirical  investigation  of  the 
predictors of executive career success”, Personnel Psychology, Vol. 48, pp. 485–519. 
Kirchmeyer,  C.  (1998),  “Determinants  of  managerial  career  success:  Evidence  and  explanation  of 
male/female differences”, Journal of Management, Vol. 24, pp. 673‐92.  
Kirchmeyer, C.  (2000), “Managers’  gender roles across career experiences: Evidence for the plaster 
vs. plasticity debate”, Paper presented at the 2000 Academy of Management Meetings, Toronto.  
Krosnick,  J.  A.  (1991),  “Response  strategies  for  coping  with  the  cognitive  demands  of  attitude 
measures in surveys”, Applied Cognitive Psychology, Vol. 5, pp. 213‐36. 
Kupperschmidt,  B.  (2000),  “Multi‐generation  employees:  Strategies  for  effective  management”, 
Health Care Manager, Vol. 19, pp. 65‐76.  
Lankard,  B.A.  (1995),  “Career  development  in  Generation  X:  Myths  and  realities”,  ERIC 
Clearinghouse on Adult, Career, and Vocational Education, Columbus, OH. 
Lee,  J.S.  and  Stolte,  J.F.  (1994),  “Cultural  and  Structural  Determinants  of  Justice  Reactions  in  the 
Economic Domain”, Social Behavior and Personality, Vol. 22, pp. 319‐28. 
Littleton, S. M.,  Arthur,  M. B. and  Rousseau, D.  M. (2000), “The future of boundaryless careers”, in 
Collin, A. and Young, R.A. (Eds), The Future of Career, Cambridge University Press, Cambridge, 
pp. 101‐14.   
Ludwick, R., & Zeller, R. A. (2001), “The factorial survey: An experimental method to replicate real 
world problems”, Nursing Research, Vol. 50, pp. 129‐33. 
Marshall, J. (1984), Women Travellers in a Male World, Wiley, London. 
Mirvis,  P.H.  and  Hall,  D.T.  (1994),  "Psychological  success  and  the  boundaryless  career",  Journal  of 
Organizational Behavior, Vol. 15, pp.365‐80.  
Ng, T., Eby, L. T., Sorensen, K. and Feldman, D. (2005), “Predictors of objective and subjective career 
success: A meta‐analysis”, Personnel Psychology, Vol. 58, pp. 367‐408.  
Nicholson,  N.,  and  West,  M.A.  (1988),  Managerial  job  change:  Men  and  women  in  transition, 
Cambridge University Press, Cambridge.  
Orpen,  C.  (1998),  “The  effects  of  organizational  centrality  on  employee  career  success  and 
satisfaction”, Social Behavior & Personality, Vol. 26, pp. 85‐8. 
Osterman, R. (1996), Broken ladders, Oxford University Press, New York, NY.  
Parker B. and  Cusmir  L.H. (1990), “A generational and sex‐based  view of managerial work values”, 
Psychological Reports, Vol. 66, pp. 947‐51.  
Phillips,  D.  L.  and  Clancy,  K.J.  (1972),  “Some  Effects  of  ‘Social  Desirability’  in  Survey  Studies”, 
American Journal of Sociology, Vol. 77, pp. 921–41. 
Ployhart, R.E., Ryan, A.M., and Bennett, M. (1999), “Explanation for selection decisions: Applicants' 
reactions  to  informational  and  sensitivity  features  of  explanations”,  Journal  of  Applied 
Psychology, Vol. 84, pp. 87‐106.  
Rousseau,  D.M.  (1990),  “New  hire  perceptions  of  their  own  and  their  employer's  obligations:  A 
study of psychological contracts”, Journal of Organizational Behavior, Vol. 11, pp. 389‐400. 
Russo,  N.,  Kelly,  M.  and  Deacon,  M.  (1991),  "Gender  and  sex  related  attribution:  beyond 
individualistic conceptions of achievement", Sex Roles, Vol. 25, pp.331‐50. 
Savickas,  M.L.  (1995),  “Constructivist  counseling  for  career  indecision”,  Career  Development 
Quarterly, Vol. 43, pp. 363–73. 
Schein, E. H. (1990), Career Anchors (Rev. Ed.), Pfeiffer, Inc, San Diego. 
265
Seibert,  S.E.,  Crant,  J.M.,  and  Kraimer,  M.L.  (1999),  “Proactive  personality  and  career  success”, 
Journal of Applied Psychology, Vol. 84, pp. 416‐27.  
Shibutani, T. (1962), "Reference Groups and Social Control", in Rose, A. (Ed), Human Behavior and 
Social Processes, Houghton Mifflin, Boston, MA, pp. 128‐45. 
Smola, K.W. and Sutton, C.D. (2002), “Generational differences: Revisiting generational work values 
for the new millennium”, Journal of Organizational Behavior, Vol. 23, pp. 363‐82. 
Spilerman, S. (1977), “Careers, labor market structure, and socio‐economic achievement”, American 
Journal of Sociology, Vol. 83, pp. 551–93. 
Stevens,  J.  (2002),  Applied  Multivariate  Statistics  for  the  Social  Sciences,  Lawrence  Erlbaum 
Associates, Mahwah, NJ. 
Sturges, J. (1999), “What it means to succeed: Personal conceptions of career  success held by male 
and female managers at different ages”, British Journal of Management, Vol. 10, pp. 239‐52.  
Super, D.E. (1980), “A life‐span, life‐space approach to career development”, Journal  of Vocational 
Behavior, Vol. 16, pp. 282–98.  
Super, D. E., Savickas, M. L. and Super, C. M. (1996). “The life‐span, life‐space approach to careers”, 
in  Brown,  D.  &  Brooks,  L.  (Eds),  Career  choice  and  development,  Jossey‐Bass,  San  Francisco, 
pp. 121‐78. 
Tharenou,  P.  (1999).  Is  there  a  link  between  family  structures  and  women’s  and  men’s  career 
advancement? Journal of Organizational Behavior, Vol. 20, pp. 837‐63.  
Verbruggen, M., Sels, L. and Forrier, A. (2007). “Unraveling the relationship between organizational 
career  management  and  the  need  for  external  career  counseling”,  Journal  of  Vocational 
Behavior, in press.  
Walton,  S.,  and  Mallon,  M.  (2004),  “Redefining  the  boundaries?  Making  sense  of  career  in 
contemporary New Zealand”, Asia Pacific Journal of Human Resources, Vol. 42, pp. 75‐95. 
Wayne, S.J., Liden, R.C., Kraimer, M.L. and Graf, I.K. (1999), “The Role of Human Capital, Motivation 
and  Supervisor  Sponsorship  in  Predicting  Career  Success”,  Journal  of  Organizational 
Behavior, Vol. 20, pp. 577‐95. 
Wilensky,  H.L.  (1961),  “Careers,  lifestyles,  and  social  integration”,  International  Social  Science 
Journal, Vol. 12, pp. 553–8. 
Zunker, V. (1994), Career counseling: Applied concepts of life planning (4th ed.), Brooks/Cole, Pacific 
Grove, CA. 
266
 
267
Paper 8. 
 

Career success: Constructing a 
multidimensional model 
 
 
A multidimensional model of career success was developed aiming to be more inclusive 
than existing models. In a first study, 22 managers were asked to tell the story of their 
careers.  At  the  end  of  each  interview,  idiosyncratic  career  success  ‘construct  ladders’ 
were  constructed  for  each  interviewee  through  an  interactive  process  with  the 
interviewer. The 42 superordinate career success operationalizations that came forward 
through  this  process  were  then  used  as  input  in  a  Q‐sort  study  in  which  30  subject 
matter  experts  (SMEXs)  served  as  judges.  Through  multidimensional  scaling  (MDS),  a 
model  incorporating  the  different  idiosyncratic  meanings  the  interviewees  attached  to 
the  career  success  construct  was  obtained.  A  two‐dimensional  configuration  (affect  – 
achievement  vs.  intra‐personal  –  inter‐personal)  comprising  9  regions  (performance, 
advancement,  self‐development,  creativity,  security,  satisfaction,  recognition, 
cooperation  and  contribution)  was  concluded  upon  as  the  optimal  solution. 
Correspondences with earlier conceptual models of career success are spelled out. 
 
 

 
268

Figure IX. Topicality of Paper 8.
269
Paper 8. 
 
Career success: Constructing a multidimensional model 
 
The  ‘new’  careers  literature  is  grounded  in  the  supposition  of  a 
substantial  change  affecting  careers  and  the  career  success  construct 
(Adamson,  Doherty  &  Viney,  1998;  Dany,  2003;  Sullivan,  1999).  In 
particular, much has been written about the gradual displacement of the 
traditional,  organization‐based  career  and  the  old  psychological 
contract with new types of career and the new, employability‐orientated 
psychological  contract  (Argyris,  1960;  Arthur,  1994;  Briscoe,  Hall  & 
DeMuth,  2006;  Fugate,  Kinicki  &  Ashfort,  2004;  Rousseau,  1995; 
Sullivan, 1999). Consequently, there has been  a noticeable evolution in 
the  way  career  and  career  success  are  defined.  While  Wilensky  (1961) 
referred  to  career  as  a  succession  of  related  jobs,  arranged  in  a 
hierarchy  of  prestige,  through  which  persons  move  in  an  ordered 
(more‐or‐less  predictable)  sequence  (p.  523),  Super  (1980)  defined 
career  as  the  combination  and  sequence  of  roles  played  by  a  person 
during  the  course  of  a  lifetime  (p.  282).  Finally,  Arthur,  Hall  and 
Lawrence (1989) defined career as the evolving sequence of a person’s 
work experiences over time (p. 8), which is an established definition of 
career today (Arthur, Khapova & Wilderom, 2005). This evolution from 
‘jobs’ to ‘experiences’ was reflected in the postmodern turn in the social 
sciences  during  the  late  1980s  (Savickas,  1995),  when  attention 
increasingly shifted from the objective to the subjective world of work.  
 
The  objective  face  of  career  is  mostly  concerned  with  observable, 
measurable  and  verifiable  attainments  such  as  pay,  promotion  and 
occupational status, which have long been recognized as ‘the’ hallmarks 
of  career  in  many  different  societies  (Nicholson,  2000).  However, 
economic  trends  such  as  organizational  delayering,  downsizing  and 
outsourcing  have  diminished  the  opportunity  as  well  as  the  desire  to 
have ‘traditional’, upwardly mobile careers, and as such, the experience 
of  objective  career  success  (Evans,  Gunz  &  Jalland,  1997;  Hall,  2002; 
Heslin,  2005;  Reitman  &  Schneer,  2003).  Subjective  career  success,  on 
the  other  hand,  refers  to  career  satisfaction  about  all  aspects  of  career 
relevant  to  a  specific  individual  (Greenhaus,  Parasuraman  &  Wormley, 
1990).  It  is  measured  in  terms  of  an  individual’s  perceptions  of  his  or 
her  own  success,  based  on  perceptions  of  personal  career 
accomplishments and future prospects (Aryee, Chay & Tan, 1994; Judge, 
Cable,  Boudreau  &  Bretz,  1995;  Melamed,  1996;  Nabi,  1999).  It  is  a 
combination  of  the  internalized  career  success  evaluations  made  by 
significant  others  and  one’s  own  evaluation  of  one’s  success,  weighted 
against  the  careers  of  peers,  and  certain  age‐related  or  personal 
270
expectations  about  career  (Betz  &  Fitzgerald,  1987)  and  a  sense  of 
progress  towards  personally  defined  career  goals  (Hall,  1976;  Wiese, 
Freund & Baltes, 2002). Although the distinction between objective and 
subjective  career  success  is  generally  underwritten  by  career  scholars, 
Hall  and  Chandler  (2005)  stressed  that  careers  are  two‐sided  and  that 
the  literature  should  steer  clear  of  ‘either‐or’  discourse:  “not  that  one 
vantage‐point  is  more  pertinent,  but  rather  that  the  subjective  career 
cannot be ignored – as it was in prior decades – particularly in today’s 
career environment” (p. 156).    
 
Rather  than  stating  that  the  subjective  career  is  ‘taking  over’  from  the 
objective career, it would be more correct to assert that career scholars 
today are becoming increasingly aware of the fact that career success is 
as a social construction rather than an objective reality (Adamson et al., 
1998; Chen, 1997; Collin & Young, 1992; Mallon & Cohen, 2000; Young, 
Valach  &  Collin,  1996),  a  dynamic  concept  rather  than  a  static  truth 
(Savickas, 2005), evolving along with the historical and cultural contexts 
surrounding  it  (Stead,  2004;  Young  &  Collin,  2004).  Older  research 
departed too much from the assumption of career success as a variable 
that  is  perceived  consensually  by  persons  evaluating  others’  careers 
(Gattiker  &  Larwood,  1986),  as  a  result  laying  the  foundation  for  the 
reification of career – the supposition that  a career is a ‘real’,  material, 
demonstrable entity together with a fixation on terminology, which one 
considers  as  factual  rather  than  as  a  complex  and  dynamic  reality  that 
can be interpreted in different ways (Evetts, 1992).  
 
The construction of career success contextualized 
 
The  current  study  aims  to  dissect  the  meaning  of  the  career  success 
construct  by  applying  the  laddering  interview  technique  as  well  as  Q‐
sort  methodology  to  generate  a  more  inclusive  model  of  what  career 
success might mean to different people. Several recent publications have 
made  apparent  that  career  success  can  have  a  multitude  of  meanings, 
depending  on  the  individual  or  subgroup  constructing  the  meaning 
(Adamson et al., 1998; Collin & Young, 1992).  
 
First of all, the construction of meaning for career and career success is 
a  historically  and  culturally  specific  process  (Young  &  Collin,  2004). 
Savickas (2000) described the ‘rise and fall’ of career in North America 
by outlining the historical evolution from an agricultural to an industrial 
economy  and  finally,  to  the  current‐day  knowledge  economy,  causing 
careers to be decreasingly regulated and linear (Arthur, 1994; Briscoe et 
al.,  2006;  Evetts,  1992;  Fugate  et  al.,  2004;  Rousseau,  1995;  Sullivan, 
271
1999).  With  regard  to  culture,  several  authors  have  demonstrated  that 
the  career  narratives  of  people  from  different  countries  vary.  Dany 
(2003)  described  the  European  culture  as  bureaucratic,  with  servile 
obedience, risk aversion and strong commitment as key features. The US 
culture, on the other hand, is described as ‘contractual’; risk‐taking and 
entrepreneurial,  individualist  attitudes  are  valued  (Briscoe,  Hall,  Las 
Heras  &  Unite,  2007).  Other  researchers  compared  prevalent  career 
theories  –  that  are  mostly  contextualized  for  White  people  in  middle‐
class US environments (Blustein, 2001) – to career narratives studied in 
Austria (Chudzikowski, Demel & Mayrhofer, 2007), China (Shen, Hall & 
Fei, 2007), Costa Rica (Ogliastri, Caballero, Rodríguez & Sabando, 2007), 
Japan (Taniguchi, 2007), Malaysia (Abdul‐Ghani, Poon, Noordin, Briscoe 
&  Jones,  2007),  Spain  (Las  Heras  &  Zikic,  2007),  and  among  Northern 
Plains American Indians (Juntunen et al., 2001).  
 
Furthermore,  sub‐culture  and  socio‐economic  background  can  play  an 
important  role  in  the  way  people  construct  the  meaning  of  career 
success  (Stead,  2004).  For  instance,  Sturges  (1999)  reported  that 
interviewees  who  grew  up  in  blue‐collar  environments  indicated  that 
their  parents  ‘did  not  have  careers’.  Hennequin  (2007)  found,  in  line 
with  Thomas  (1989),  that  blue‐collar  workers  create  their  own 
definition of career success, in which seniority, technical competence or 
the amount of danger experienced at work play a role.  
 
Finally, and on a more individual level, developmental theories such as 
socialization  and  life‐cycle  processes  may  be  relevant.  Socialization 
processes are particularly relevant when it comes to the career success 
definitions of men versus those of women. For instance, certain cultures 
limit  the  desirability  and  possibility  of  women  pursuing  careers  in 
‘masculine’  sectors  by  emphasizing  their  responsibility  as  ‘wife  and 
mother’  (Stead,  2004),  while  men  are  ‘breadwinners’  or  ‘providers’ 
(Dyke  &  Murphy,  2006;  Parker  &  Chusmir,  1992;  Powell  &  Mainiero, 
1992;  Russo,  Kelly  &  Deacon,  1991;  Sturges,  1999).  Moreover,  several 
authors  have  found  that  the  meaning  attributed  to  career  success  can 
shift  across  generations  (e.g.  Dries,  Pepermans  &  De  Kerpel,  2008)  as 
well as during one person’s life cycle, depending on the career stage, the 
different  life  tasks  and  the  development  of  the  self‐concept  (Lee  et  al., 
2006; Poole, Langan‐Fox & Omodei, 1993).  
 
Recent studies on the meaning of career success 
 
There have been a few studies scrutinizing the actual topic of the career 
success construct and its dimensionality in itself (as opposed to its many 
272
research applications as a vaguely operationalized dependent variable). 
Based  on  an  open‐ended  survey  study,  Gattiker  and  Larwood  (1986) 
compiled  a  measure  of  career  success  consisting  of  five  factors:  job 
success,  interpersonal  success,  financial  success,  hierarchical  success 
and life success. Parker and Chusmir (1990), on the other hand, named 
their  conception  of  the  construct  ‘life  success’,  arguing  that  subjective 
career success tends to encompass factors from outside the career. They 
identified  status/wealth,  contribution  to  society,  family  relationships, 
personal  fulfilment,  professional  fulfilment  and  security  as  relevant 
factors.  Sturges  (1999)  and  Nabi  (2001)  both  reported  finding  two 
factors  in  their  qualitative  studies,  i.e.  external,  or  extrinsic,  career 
success  versus  internal,  or  intrinsic,  career  success.  Dyke  and  Murphy 
(2006) classified their findings on the meanings of career success under 
the four factors balance, relationships, recognition and material success; 
Lee et al. (2006) identified organization‐based, personal and interlinked 
themes  in  interviewees’  discourse  about  career  success.  Finally, 
Hennequin  (2007)  established  three  factors  within  the  career  success 
construct:  material  career  success,  psychological  career  success  and 
social career success.  
 
Much  recent  work  refers  to  the  urgent  need  for  an  adequate 
conceptualization of what career success means, both qualitatively and 
quantitatively (Poole  et  al., 1993; Heslin, 2005). According to  Duxbury, 
Dyke  and  Lam  (1999),  as  well  as  Heslin  (2005),  an  expansion  of  the 
definition of career success is necessary, but more empirical work needs 
to  be  done  to  determine  what  such  an  expansion  might  entail.  The 
current  study  answers  calls  for  more  studies  that  let  people  define 
career  success  in  their  own  words  and  a  more  fine‐grained  analysis  of 
different career success criteria (Heslin, 2005).  
 
Methods and results 
 
Two  consecutive  studies  were  conducted:  the  first  study  using  the 
laddering  interview  technique  to  generate  idiosyncratic  meanings 
attributed  to  the  career  success  construct;  and  the  second  study 
attempting  to  fit  these  meanings  into  an  inclusive  model  of  career 
success  by  using  Q‐sort  methodology  and  multidimensional  scaling 
(MDS). 
 
 
 
 
 
273
Study 1: Laddering interview technique 
 
Methods 
 
Measures  and  procedures.  The  laddering  interview  technique  (e.g. 
Bourne  &  Jenkins,  2005)  was  designed  to  access  higher 
(‘superordinate’)  levels  of  abstraction  when  interviewing  people  about 
the  construct  systems  by  which  they  organize  their  worlds.  It  has  its 
roots in the repertory grid technique (Kelly, 1955). Study 1 started with 
interviewees being asked to tell the stories of their careers, choose their 
own  themes  and  find  coherence  in  their  narratives.  The  reasons  for 
starting  off  with  career  narratives  were  twofold:  first,  by  letting  the 
interviewees  talk  freely  about  their  careers  a  certain  ‘rapport’  was 
created between researcher and interviewee that greatly facilitated the 
interactive  laddering  process  in  a  later  stage  of  the  interview;  second, 
allowing  the  interviewees  to  construct  their  career  narratives  at  their 
own  pace  was  necessary  to  be  able  to  generate  relevant  constructs  for 
the laddering exercise. The majority of interviewees began their career 
narratives  by  listing  their  jobs  and  employers  from  graduation  to 
present; gradually they talked more about challenges and opportunities, 
turning points and decisions, and success and failure.  
 
After  completion  of  each  narrative,  which  took  about  one  hour  on 
average, interviewee and researcher took a few minutes to recapitulate 
the  main  themes  that  arose  from  the  career  story.  At  this  point,  the 
interviewees were asked if they wanted to ‘do some ladders’ – we used 
this phrasing as Bourne and Jenkins (2005) emphasized that an informal 
tone  during  the  laddering  interviews  is  beneficial  to  interviewees’ 
openness  to  active  participation.  They  were  told  that  the  goal  of  the 
exercise was to come to their own personal definition of career success. 
In  a  first  step,  each  interviewee  was  asked  to  reflect  again  about  the 
different meanings of career success that came forward in their personal 
and  spontaneous  account  of  their  career.  He  or  she  was  then  asked  to 
select  the  three  most  central  constructs.  Subsequently,  these  three 
constructs  were  the  starting  points  of  the  laddering  exercise  for  this 
interviewee, and were noted on three separate pages, at the bottom left. 
For each chosen career success construct, the question was asked “And 
what is the opposite of [...]”? The answer was then recorded at the right 
side  of  the  page  and  a  double  arrow  was  drawn  between  the  two 
constructs to illustrate the dichotomy. Interviewees were then asked to 
choose which pole they preferred and why. The pole that was preferred 
got  the  title  ‘preferred  pole’,  the  other  ‘opposite  pole’.  The  ‘why’ 
question  was  used  to  find  the  superordinate  construct  for  the  first 
274
concept  on  the  preferred  pole.  The  process  continued  in  this  manner, 
until no more superordinate constructs for a ladder could be found. This 
method  was  applied  to  all  three  career  success  ladders  for  each 
participant.  Upon  completion,  the  interviewees  were  asked  to  go  over 
‘their’  ladders  and  to  assess  whether  or  not  these  actually  conveyed 
their personal definition of career success. If, at the end of the exercise, 
an  interviewee  believed  that  his  or  her  personal  definition  of  career 
success was not adequately captured by his or her three ladders, one or 
more  extra  constructs  from  the  career  narrative  were  elaborated.  For 
most interviewees, however, three constructs were sufficient.  
 
Participants. Participation for the study was established within five 
Belgian  organizations  (from  the  banking,  food,  medical,  metal  and 
tobacco industry, respectively) located in the Dutch‐speaking part of the 
country.  Twenty‐two  managers  participated  in  the  laddering  interview 
study;  17  were  men  (78%)  and  5  were  women  (22%).  The  age  of  the 
participants  ranged  from  28  to  59;  their  average  age  was  40.05  (sd  = 
7.25).  
 
Results 
 
Sixty‐nine career success ladders were constructed during the laddering 
interviews.  In  the  content  analyses  process  (conducted  by  each  of  the 
three  authors  to  establish  a  certain  degree  of  inter‐rater  reliability)  no 
notable differences were found between contents of the ladders of men 
and women, nor between those of managers of different ages.  
 
Below,  we  have  elaborated  a  ladder  from  one  interviewee  to 
demonstrate  how  we  arrived  at  the  superordinate  career  success 
constructs  used  throughout  the  follow‐up  study  (see  Figure  1).  It 
illustrates  how  important  it  is  to  access  superordinate  meanings 
attributed  to  career  success,  as  the  interviewee  started  of  with 
‘learning’,  yet  realized  at  the  end  of  the  exercise  –  to  her  own 
amazement – that what she really valued was ‘proving one’s worth’.  
 
One  of  the  themes  that  arose  as  central  in  the  interviewee’s  views  on 
career  success  was  ‘learning’.  The  laddering  interview  proceeded  as 
follows: 
 
Interviewer (IR): As opposed to...? 
Interviewee (IE): Not learning . . . or better yet, ‘routine’. When you get stuck 
in your daily routine and never learn anything new . . . I prefer ‘learning’. 
IR: And why do you prefer this pole over the other? 
IE: I guess because I don’t want to get stuck in that kind of routine . . . and by 
learning I can develop myself as a person . . . as opposed to stagnating in life, 
275
and  in  my  career.  The  next  level  is  immediately  obvious  to  me!  When  I 
develop myself as a person, I feel like there is no stopping me, that I can keep 
going further. It makes me feel like I’m pushing back all boundaries.  
IR: As opposed to what? 
IE: I would say stagnation, but that was the previous level . . . When someone 
is not crossing that boundary, they’re just standing still. 
IR: Which do you prefer and why? 
IE: When you’re standing still, you’re just another face in the crowd . . . This 
process is really confronting me with how much I value acknowledgement! I 
really  don’t  want  to  be  just  a  face  in  the  crowd  .  .  .  When  I  push  back 
boundaries,  that  gets  me  acknowledgement.  My  boss  and  co‐workers  will 
then show me respect and show appreciation for what I do. The opposite is 
being a ‘grey mouse’, when no one notices you or acknowledges what you do. 
IR: And why is this so important to you? 
IE: Acknowledgement is very important to me because I am so competitive; I 
want to win the  internal competition. My competitiveness really shows itself 
in many areas of my life. Don’t challenge me to go for a run . . .  I hate losing!  
IR: Why is that? 
IE: You know, I guess maybe it has something to do with when I grew up . . . I 
was  the  youngest  in  my  family  and I  had  five  older  brothers!  It  wasn’t  easy 
for me to get noticed or heard. I taught myself to speak up against all those 
‘tough’ boys in my house. I learned how to make clear that I was a part of the 
family too! I always had this urge to prove my worth. I  never want to be the 
smallest! 
IR: So, ‘proving one’s worth’ versus ‘being the smallest’? 
IE: Yes! Looking at the ladder, I’m kind of surprised. I started of with learning 
and ended up with proving one’s worth! Not what I expected, but I guess it’s 
accurate.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 1. Example of a career success construct ladder obtained in Study 1 
276
Although looking into the career success value systems of each of the 22 
participants  separately does provide  interesting information for career 
scholars, it is not possible to report all of these findings here. 
 
Study 2: Q­sort technique and MDS 
 
Methods 
 
Measures  and  procedures.  The  69  career  success  ladders  –  in 
particular, the highest superordinate constructs of each preferred pole – 
that came forward in Study 1 were re‐assessed for use in Study 2, where 
they were used as input for expert analysis. The three authors served as 
independent  assessors  in  the  process  of  removing  career  success 
constructs that were identical to another construct, and thus redundant. 
Once inter‐rater consensus was reached, 27 redundant constructs were 
removed.  The  remaining  42  career  success  constructs  were  used  to 
prepare  a  Q‐sort  exercise  (Parker,  2006).  In  compiling  the  Q‐sort  card 
deck,  the  three  authors  focused  mainly  on  the  highest  superordinate 
career  success  constructs  of  the  preferred  poles.  However,  when  the 
meaning  of  a  construct  was  unclear  in  itself,  or  when  an  incorrect 
interpretation was possible, they made sure to label the corresponding 
Q‐sort  card so,  that  the  meaning  of  the  interviewee  who  generated  the 
construct was adequately conveyed.  Some  examples are:  ‘Being able to 
manage  and  shape  one’s  own  career;  achieving  everything  single‐
handedly’  and  ‘Standing  out;  being  the  best  at  what  one  does’.  The  full 
list  of  all  42  Q‐sort  cards  originating  from  the  laddering  exercise  in 
Study 1 can be consulted in the Appendix.  
 
The 42 career success constructs were printed onto separate numbered 
cards  and  sent  by  post,  along  with  an  instructions  letter,  to  a  selected 
group  of  experts  from  the  career  counselling  field.  The  experts  were 
instructed to sort all cards into structured piles or ‘Q‐sorts’. They were 
told  to  put  two  cards  in  the  same  pile  if  they  were  similar  in  meaning, 
and  in  a  different  pile  if  they  were  dissimilar  (Parker,  2006). 
Furthermore,  they  were  advised  to  only  have  piles  of  one  card  if  they 
were  absolutely  certain  they  perceived  it  as  dissimilar  from  all  other 
cards. After sorting all 42 cards into piles, they were instructed to now 
see  each  pile  as  a  category  that  required  a  label  to  capture  the  shared 
meaning  of  the  constructs  assigned  to  it.  In  order  to  ensure  that  the 
experts  did  not  lose  sight  of  the  fact  that  the  exercise  revolved  around 
the definition of career success, each card contained the phrasing ‘career 
success is…’ in the top left corner. They were then asked to  make up a 
277
document  presenting  their  categorizations  (labels  and  card  numbers) 
and return it to the researchers.  
 
Participants. Thirty subject matter experts (SMEXs) participated in 
the  Q‐sort  study.  We  chose  to  work  with  career  counselling  experts  in 
Study  2  as  they  possess  much  more  knowledge  of  the  way  different 
people  might  look  at  the  career  success  concept  than  does  the  general 
population, that might bias this type of exercise in using their personal 
preconceptions  about  success.  To  facilitate  diverse  perspectives  half  of 
the  sample  were  students  in  their  final  Master  year  of  Industrial  and 
Organizational psychology at a large Belgian University (all students had 
received  several  courses  on  the  topic  of  career  counselling  and  the 
psychology  of  careers);  the  other  half  were  Human  Resources 
professionals  specialized  in  career  counselling.  Nineteen  of  the  SMEXs 
were  women  (63%),  eleven  were  men  (37%).  The  age  of  the  students 
varied  between  23  and  25  (m  =  23.4  ;  sd  =  0.74);  the  HR  professionals 
were all between 25 and 50 years old (m = 37.07; sd = 8.94).  
 
Analysis. Statistical and interpretative analyses were performed on 
the  career  success  categorizations  that  were  obtained  from  the  Q‐sort 
study.  The  goal  of  these  analyses  was  to  determine  the  underlying 
structure of the 42 career success constructs in the form of an inclusive 
model.  To  achieve  this,  we  investigated  the  appropriate  graphical 
configuration of the career success constructs, and more specifically, the 
amount of underlying dimensions and clusters (i.e. ‘regions’) in the data 
by  applying  multidimensional  scaling  (MDS).  Multidimensional  scaling 
is  an  exploratory  technique  that  helps  researchers  determine  the 
underlying structure in sets of data, and is considered particularly useful 
for  the  development  of  theory  (Borg  &  Groenen,  1997).  As  the  aim  of 
MDS  is  to  represent  perceived  dissimilarities  between  constructs  as 
metric  distances  in  an  n‐dimensional  space,  the  data  from  Study  2 
needed to be transformed into a dissimilarity matrix first. For each pair 
of  constructs,  dissimilarities  were  calculated  as  the  total  number  of 
experts  that  had  not  sorted  them  into  the  same  category.  It  logically 
follows that values in the obtained dissimilarity matrix varied between 
0 (i.e. all experts put constructs x and y together) and 30 (i.e. no expert 
put  constructs  x  and  y  together).  A  classical  ordinal  MDS  analysis  was 
applied to  the data  matrix. The calculations were carried  out using the 
ALSCAL  algorithm  of  the  SPSS  program  (SPSS  15.0).  Subsequently, 
adequate  labels  were  determined  for  the  dimensions  and  regions 
through content analysis so as to add meaning to the model.  
 
 
278
Results 
 
Selection of the most appropriate MDS solution.  In  a  first  step,  
solutions  were  computed  for  a  one‐  to  a  five‐dimensional  model.  To 
determine the appropriate number of dimensions for the career success 
model we first considered the observed stress values and R² values for 
each number of dimensions, which were .56 (.31), .32 (.57), .24 (.69), .18 
(.77) and .13 (.83) respectively. We then plotted these stress values in a 
dimension  x  stress  graph  to  look  for  the  ‘elbow’.  This  method  of 
determining optimal dimensionality is based on Cattell’s scree test (see 
Derous, De Witte & Stroobants, 2003). The resulting plot showed a slight 
elbow  for  the  two‐dimensional  solution.  However,  some  authors  have 
stated  that  an  elbow  should  seldom  be  accepted  if  stress  values  are 
higher  than  .10  (Kruskal  &  Wish,  1978),  which  is  definitely  the  case 
here.  However,  other  authors  have  advised  researchers  using  MDS  not 
to  consider  Kruskal’s  rule‐of‐thumb  alone  in  determining  the 
appropriate  number  of  dimensions  for  a  model.  First  of  all,  simulation 
studies  have  demonstrated  that  stress  values  systematically  increase 
when  the  number  of  constructs  in  the  model  increases  (indicating  that 
higher  critical  stress  values  would  apply  for  a  42x42  matrix  such  as 
ours), and decrease when the number of dimensions increases (Spence 
&  Ogilvie,  1973;  Wagenaar  &  Padmos,  1971).  Secondly,  Borg  and 
Groenen  (1997)  suggest  that  the  ‘principle  of  simple  structure’  should 
be  applied  when  selecting  the  appropriate  dimensionality;  the  authors 
state that both over‐ and underestimation errors in proximities arise in 
MDS  spaces  of  high  dimensionality.  Finally,  a  more  substantive 
consideration  is  that  of  interpretability  (Kruskal  &  Wish,  1978).  Based 
on  all  these  concerns,  a  two‐dimensional  configuration  was  decided 
upon as the most appropriate solution. 
 
Interpretation  of  the  regions.  A  depiction  of  the  two‐dimensional 
spatial configuration of all 42 career success constructs is presented in 
Figure  3  (each  construct  is  represented  by  a  number;  the  associated 
labels  can  be  consulted  in  the  Appendix).  To  determine  the  number  of 
regions in the model, a combination of methods was used.  
 
Visual  inspection  of  the  two‐dimensional  configuration,  along  with 
additive  tree  modelling  (T‐Rex  4.0a1),  resulted  in  four  acceptable 
suggestions  for  delineation  of  the  regions.  In  order  to  decide  which  of 
these four solutions was optimal, consistency‐in‐meaning (CIM) indices 
were calculated for all suggested regions. The CIM index was created by 
Derous  et  al.  (2003)  to  provide  some  measure  of  the  consistency  in 
meaning  of  constructs  that  spatially  belong  to  the  same  region.  This  is 
279
an important step in the analysis of a model, as MDS does not offer any 
information on the meaning of constructs, only on their proximity. Thus, 
the  CIM  index  is  a  useful  aid  in  deciding  which  constructs  to  group 
together into one region. According to Derous et al. (2003), “regions are 
interpreted  consistently  if  all  items  belonging  to  that  region  (i.e.  in  the 
MDS  solution)  were  sorted  into  piles  which  were  labelled  in  a  very 
similar  way  by  experts”  (p.  188).  The  CIM  index  is  represented  in  the 
following equation:  
 
CIM =   ⎡ (δL − 1)i ⎤  
⎢1 − (NiNε ) − 1⎥ × 100
⎣ ⎦
 
CIM  indices  are  calculated  using  three  variables:  δL,  the  number  of 
labels with  different  meanings of  all items  graphically belonging to the 
same  region  i;  Ni,  the  number  of  items  belonging  to  the  same  region  i; 
and Nε, the maximum number of different labels that can be ascribed to 
the items of region i – it follows logically that Nε is a constant value that 
equals  the  number  of  experts  in  the  study  (n  =  30).  In  order  to 
determine  the  δL  value  for  each  region  of  each  of  the  four  solutions, 
content  analysis  was  necessary.  The  three  authors  conducted  this 
analysis separately: in a first step, we each made up a list of all category 
labels  assigned  by  the  SMEXs  who  completed  the  Q‐sort  exercise.  We 
then grouped these labels together by meaning. Upon completion of this 
task, the three authors sat down together to reach inter‐rater agreement 
on the number and content of different meanings that arose from the Q‐
sort  exercise.  We  then  made  up  a  final  list  of  meanings  (17)  and 
associated labels (173). In a next step we made a list, for each of the 42 
career success constructs, of all category labels the constructs had been 
sorted  under  in  Study  2.  All  of  the  above  steps  made  it  possible  to 
‘objectively’  count  the  number  of  labels  for  each  solution,  so  that  CIM 
indices  could  be  calculated  (the  CIM  indices  are  reported  in  the 
Appendix). 
 
However,  we  included  one  additional  step  in  our  content  analysis  to 
determine  which  constructs  to  group  together  into  regions.  As  is,  the 
CIM  index  does  not  take  into  account  the  degree  of  consensus  or 
dispersion  between  experts  about  the  meaning  of  a  construct.  For 
instance, it is possible that constructs x and y have both been attributed 
10  different  meanings;  however,  it  is  well  possible  that  x  was  placed 
under  a  similar  category  by  21  SMEXs,  and  that  the  9  other  meanings 
attributed to x were each only mentioned by 1 expert. For y on the other 
hand, maybe each of the 10 meanings was agreed upon by 3 SMEXs.  
 
 
 
 
 
 
 
 
 
 
 
 
 
280

 
 
 
 
 
 
 
 
 
 
 
Figure 2. Final MDS model of all 42 career success constructs mapped onto two dimensions and into 9 regions 
281
Intuitively,  this  would  mean  that  there  is  more  consensus  on  the 
meaning of x than there is on the meaning of y, although this variable is 
not  incorporated  into  the  CIM  index.  In  order  to  solve  this  issue,  we 
mapped out each of the 42 career success constructs onto each of the 17 
different  meanings  in  a  frequency  matrix  (the  frequencies  being  the 
number  of  SMEXs  that  placed  the  construct  under  a  category  with  a 
label falling under meanings 1‐17). 
 
Combining  the  results  of  all  of  the  above  analyses  –  visual  inspection, 
additive  tree  modelling,  and  the  two  types  of  consistency‐in‐meaning 
evaluation – a final solution was decided upon by the authors, which is 
presented  in  Figure  3.  Together  with  the  information  in  the  Appendix, 
Figure  3  presents  the  final  solution  consisting  of  nine  regions: 
performance (1), advancement (2), self‐development (3), creativity (4), 
security  (5),  satisfaction  (6),  recognition  (7),  cooperation  (8)  and 
contribution  (9).  The  regions  self‐development,  satisfaction  and 
contribution were further divided into sub‐regions based on the content 
analyses. 
 
Interpretation  of  the  dimensions.  The  first,  horizontal,  dimension  is 
called  affect  versus  achievement.  Constructs  at  the  affect  end  of  this 
dimension  belong  to  regions  5  (security),  6  (satisfaction),  7 
(recognition),  8  (cooperation)  and  9a  (experienced  contribution).  The 
affect  end  of  this  dimension  stands,  more  specifically,  for  feelings  and 
perceptions  people  may  have  had  throughout  their  careers  that  cause 
them  to  evaluate  their  career  success  as  high  or  low.  Constructs 
belonging  to  these  regions  are  labelled  ‘experiencing  stability  and  job 
security’  (n°28)  and  ‘believing  to  have  achieved  positive  matters,  in  an 
ethical  way’  (n°27),  among  others.  The  other  end  of  the  dimension, 
achievement,  refers  to  factual  accomplishments  characteristic  of 
peoples’ careers that cause them to evaluate their career success as high 
or  low.  Constructs  at  this  end  of  the  scale  belong  to  regions  1 
(performance),  2  (advancement),  3  (self‐development),  4  (creativity) 
and 9b (factual contributions). Examples of construct labels are ‘getting 
promoted; climbing the ladder’ (n°30) and ‘acquiring many experiences, 
inside one’s job and out’ (n°25).  
 
The  second,  vertical,  dimension  is  labelled  intra‐personal  versus  inter‐
personal. Constructs on the intra‐personal side of the dimension belong 
to  regions  3  (self‐development),  4  (creativity),  5  (security)  and  6 
(satisfaction).  The  focus  on  this  end  of  the  dimension  is  on  the  career 
actor’s  ‘self’,  on  aspects  of  his  or  her  internal  world  that  would  cause 
him or her to evaluate career success as high or low. This is illustrated 
282
by  the  category  labels  belonging  to  these  regions,  e.g.  ‘accomplishing 
one’s own goals’ (n°4) and ‘feeling healthy and happy, at home as well as 
at  work’  (n°9).  Constructs  on  the  inter‐personal  side  of  this  dimension 
belonged to regions 1 (performance), 2 (advancement), 7 (recognition), 
8  (cooperation)  and  9  (contribution).  Construct  labels  were  ‘having  a 
good  understanding  with  one’s  employer  –  one  that  does  not  always 
require  words’  (n°8)  and  ‘being  an  important  link  in  the  process  that 
leads  to  organizational  success’  (n°35),  among  others.  Therefore,  the 
inter‐personal  side  of  the  dimension  illustrates  a  focus  on  the 
relationships  with  the  outside  world  career  actors  engage  in  – 
validations of their careers that take place outside their ‘self’ that cause 
them to evaluate their own career success as high or low. 
 
Discussion 
 
Discussion of the proposed career success model 
 
The aim of the current paper was to generate a more inclusive model of 
career  success.  A  model  with  four  quadrants  was  decided  upon  as 
relevant for categorizing the different meanings attributed to the career 
success construct obtained from the two studies presented in this paper: 
 
(I) inter­personal – achievement. Career success constructs belonging to 
this  quadrant  refer  to  factual  accomplishments  that  characterize  a 
career; the world external to the career actor’s ‘self’ acts as the source of 
validation.  This  quadrant  contains  three  distinct  ‘regions’  of  meaning: 
performance  (i.e.  success  in  terms  of  attaining  verifiable  results  and 
meeting  set  goals);  advancement  (i.e.  success  in  terms  of  progressing 
and growing, both in terms of level and experience); factual contribution 
(i.e.  success  in  terms  of  contributing  something  tangible  to  the 
collective, for instance to an organization, as an individual); 
 
(II) intra­personal – achievement. Career success constructs belonging to 
this  quadrant  refer  to  factual  accomplishments  that  characterize  a 
career;  the  career  actor’s  ‘self’  acts  as  the  source  of  validation.  This 
quadrant  contains  two  distinct  ‘regions’  of  meaning:  self‐development 
(i.e.  success  in  terms  of  reaching  one’s  full  potential  through  self‐
management  of  challenges  and  learning  experiences);  creativity  (i.e. 
success in terms of creating something innovative and extraordinary); 
 
(III)  intra­personal  –  affect.  Career  success  constructs  belonging  to  this 
quadrant  refer  to  feelings  and  perceptions  that  characterize  a  career; 
the  career  actor’s  ‘self’  acts  as  the  source  of  validation.  This  quadrant 
283
contains two distinct ‘regions’ of meaning: security (i.e. success in terms 
of  being  able  to  meet  one’s  financial  and  employment  needs); 
satisfaction (i.e. success in terms of achieving personal satisfaction and 
happiness, both in the family and in the work domain); 
 
(IV)  inter­personal  –  affect.  Career  success  constructs  belonging  to  this 
quadrant  refer  to  feelings  and  perceptions  that  characterize  a  career; 
the  world  external  to  the  career  actor’s  ‘self’  acts  as  the  source  of 
validation.  This  quadrant  contains  three  distinct  ‘regions’  of  meaning: 
recognition  (i.e.  success  in  terms  of  being  adequately  rewarded  and 
appreciated  for  one’s  efforts  and  talents);  cooperation  (i.e.  success  in 
terms of working well together with peers, superiors, subordinates and 
clients); perceived contribution (i.e. success in terms of serving society 
through work, in an ethical way). 
 
Links with existing models of career success 
 
When  putting  our  two‐dimensional  (affect  –  achievement  vs.  intra‐
personal – inter‐personal) model of career success side by side with the 
models  found  in  earlier  studies  (Dyke  &  Murphy,  2006;  Gattiker  & 
Larwood, 1986; Hennequin, 2007; Lee et al., 2006; Nabi, 2001; Parker & 
Chusmir,  1991;  Schein,  1978,  1990;  Sturges,  1999),  we  see  that  the 
model  proposed  in  this  paper  seems  to  be  rather  inclusive,  in  that  it 
allows  to  fit  all  of  the  concepts  of  previous  studies  within  its  nine 
proposed  ‘regions’  (performance,  advancement,  self‐development, 
creativity,  security,  satisfaction,  recognition,  cooperation  and 
contribution).  We  established  these  correspondences  with  existing 
models  of  career  success  by  carefully  scrutinizing  factors  and  items  of 
the career success scales constructed by Gattiker and Larwood (1986), 
Schein  (1978,  1990),  Parker  and  Chusmir  (1991)  and  Nabi  (2001);  in 
the case of Sturges (1999), Dyke  and Murphy (2006), Lee et  al. (2006) 
and Hennequin (2007) we dissected the qualitative data reported. Table 
1  presents  the  hypothesized  correspondences  between  the  meanings 
attributed to career success in previous studies (wording is based on the 
original texts) and the model presented in this paper.  
 
When considering the distribution of the correspondences between the 
career success constructs of earlier studies and the model presented in 
this paper, we see that constructs fitting into our ‘satisfaction’ region are 
by far most prevalent (24% of the 114 constructs in Table 1 correspond 
to  our  conception  of  ‘satisfaction’);  somewhat  less  prevalent  were 
correspondences  with  the  regions  ‘contribution’  (16%),  ‘recognition’ 
(15%) and ‘cooperation’ (14%).  
Table 1. Hypothesized correspondences between existing models of career success and the model presented in the current study 
 
Existing models of career success  Correspondences with model of career success presented in the current study a
Gattiker & Larwood (1986)b   
   1. Job success  performance1; development opportunities3; responsibility9; management support8; happiness at 
work6; dedication9  
   2. Interpersonal success  peer respect and acceptance8; good evaluations1; supervisor confidence8
   3. Financial success  fair compensation compared to peers7; high income compared to peers7 
   4. Hierarchical success  (opportunities for) promotion2; meeting career goals as planned3 
   5. Life success  happiness in private life6
Schein (1987, 1990) b   
   1. Technical/functional competence  being good at something1; achieve a high level of competence1; become an expert1
   2. General managerial competence  being a general manager2; managing the efforts of others8; making decisions that affect people9
   3. Security/stability  stability, continuity5; financial and employment security5; avoid risks5 
   4. Autonomy/independence  freedom6; working without rules and constraints6
284

   5. Entrepreneurial creativity  creativity4; building a business based on ideas4
   6. Service/dedication to a cause  helping other people9; contributing to society, making the world a better place9
   7. Pure challenge  seeking constant stimulation3; challenging problem‐solving and competitive skills1
   8. Life style  integrating work and personal life6

Notes.  The numbers next to each attributed meaning refer to supposed correspondences with the nine regions of the career success model 
presented in Figure 2 and in the Appendix; b Survey study: the attributed meanings per factor were derived from the items in the career success 
scale presented in this paper (wording based on original text); c Interview study: the attributed meanings per factor were derived from the career 
success constructs reported in the results section (wording based on original text). 
Table 1. Continued 
 
Existing models of career success  Correspondences with model of career success presented in the current study a
Parker & Chusmir (1991) b   
   1. Status/wealth  power, influence9; high income compared to peers7; good benefits7; recognition7; salary increases2
   2. Contribution to society  helping others9; being useful9; making a difference9; improving well‐being of the workforce9
   3. Family relationships  happy and stable marriage6; being a good parent6; raising children successfully6
   4. Personal fulfilment  inner peace, contentment, happiness6; personal worth6, creativity 4; satisfaction, happiness6; self‐
respect6; enjoying non‐work activities6 
   5. Professional fulfilment  competence, performance1; commitment9; supervisor confidence8; acceptance by colleagues8; job 
satisfaction6 
   6. Security  economic security5, job security5
Sturges (1999)c   
   1. External career success  financial rewards7; hierarchical level2; advancement2
   2. Internal career success  influence9; recognition7; competence1; personal achievement3; enjoyment6; integrity9; balance6; 
285

receiving feedback8; responsibility9 
Nabi (2001) b   
   1. Extrinsic job success  equitable pay, fair income7; responsibility9; promotional opportunities2 
   2. Intrinsic job success  job happiness6; management support8; likable work6; likable colleagues8; good performance 
evaluations1; supervisor confidence8 

Notes.  The numbers next to each attributed meaning refer to supposed correspondences with the nine regions of the career success model 
presented in Figure 2 and in the Appendix; b Survey study: the attributed meanings per factor were derived from the items in the career success 
scale presented in this paper (wording based on original text); c Interview study: the attributed meanings per factor were derived from the career 
success constructs reported in the results section (wording based on original text). 
 
 
Table 1. Continued 
 
Existing models of career success  Correspondences with model of career success presented in the current study a
Dyke & Murphy (2006) c   
   1. Balance  happiness6; personal fulfilment6; content with life’s direction6; freedom6
   2. Relationships  stable relationship with partner and children6, community9; contribution9; respect8; support8; team 
achievements8; recognition, status7 
   3. Recognition  being good at a challenging job1; being recognized as competent7; acceptance by clients8; 
recognition, acknowledgement7 
   4. Material success  good salary7; being able to live good5; no financial burdens5 
Lee et al. (2006) c   
   1. Organization‐based themes  peer respect8; upward mobility2; appreciation/recognition7 
   2. Personal themes  having a life outside work6; learning, growing and being challenged3; fun and enjoyment/doing 
interesting work6 
   3. Interlinked themes  performing well1; having an impact/making a contribution9 
286

Hennequin (2007) c   
   1. Material career success  monetary rewards7; fringe benefits; hierarchical position2; number of promotions2
   2. Psychological career success  career satisfaction6; job success1; interpersonal success8; life balance6 
   3. Social career success  social status7; recognition7; reputation7

Notes.  The numbers next to each attributed meaning refer to supposed correspondences with the nine regions of the career success model 
presented in Figure 2 and in the Appendix; b Survey study: the attributed meanings per factor were derived from the items in the career success 
scale presented in this paper (wording based on original text); c Interview study: the attributed meanings per factor were derived from the career 
success constructs reported in the results section (wording based on original text). 
 
 
287
Least  prevalent  were  correspondences  with  ‘performance’  (10%), 
‘advancement’  (8%)  and  ‘self‐development’  (4%);  the  lowest 
correspondence  was  found  with  ‘creativity’  (3%).  These  numbers 
indicate  that  previous  research  into  the  meaning  of  the  career  success 
construct has focused mainly on the inter‐personal – affect quadrant of 
career  success,  whereas  the  opposite  quadrant,  intra‐personal 
achievement,  has  been  stressed  much  less  in  the  literature  on  career 
success  –  this  quadrant,  however,  does  link  well  to  the  literature  on 
career self‐management and protean careers (e.g. Hall, 2002). 
 
Remarkably, the one career success construct that is most difficult to fit 
into our model of career success, is salary – one of the most frequently 
used  measures  in  the  literature  (Heslin,  2005).  Without  further 
information  on  what  is  meant  exactly  ‘salary’,  it  could  be  fitted  into 
several different regions: in the sense of ‘receiving a fair salary’ it would 
fit  best  within  region  7,  recognition;  in  the  sense  of  ‘not  having  any 
financial burdens’ it would be better fitted into region 5, security; and in 
the  sense  of  ‘salary  increases’  we  would  fit  it  into  region  2, 
advancement. In contrast, the current study attempted to use relatively 
unambiguous  career  success  constructs,  which  was  demonstrated  by 
the high inter‐rater agreement (i.e. between the three authors) reached 
at different points during the content analyses of the data and reflected 
in  the  high  consistency‐in‐meaning  (CIM)  parameters  (Derous  et  al., 
2003) reported in the Appendix.  
 
Another  finding  of  interest  is  that  our  model  of  career  success 
corresponds specifically well with the career anchors conceptualized by 
Schein  (1978,  1990):  his  technical/functional  competence  anchor 
corresponds  quite  well  with  our  ‘performance’  region;  the 
security/stability  anchor  with  ‘security’;  the  autonomy/independence 
anchor  with  the  ‘self‐development’  sub‐region  ‘independence’;  the 
entrepreneurial  creativity  anchor  with  ‘creativity’;  and  the 
service/dedication  to  a  cause  anchor  with  ‘contribution’.  This  is  an 
interesting finding, as career anchors literature has mostly been linked 
to career management, decision‐making, and planning (King, 2004), but 
never explicitly to definitions of career success. It does seem that career 
anchors put a stronger emphasis on action, while career success refers 
to  construct  systems  that  help  people  organize  their  worlds;  the 
emphasis  on  career  identity  is  also  less  pronounced  in  definitions  of 
career success, as people do not only use their career success construct 
systems to evaluate their own careers, but also those of others. Finally, 
career anchors are considered input variables of career, which precede 
action,  whereas  career  success  is  generally  conceptualized  as  a 
288
destination,  a  desired  outcome  of  career  (e.g.  Hall,  1976).  Nonetheless, 
the  extensive  and  well‐established  literature  on  career  anchors  might 
provide  useful  input  for  answering  the  calls  in  the  literature  for  an 
expansion  of  the  definition  of  career  success  (Duxbury  et  al.,  1999; 
Heslin, 2005). 
 
Limitations 
 
Although the proposed model of career success does appear to be more 
inclusive than previous models, it is certainly true that further external 
validation is necessary. By no means do we claim to have developed ‘the’ 
model  of  career  success.  As  with  all  single‐standing  studies,  variables 
from  the  research  context  (i.e.  culture,  language,  sample  selection, 
researchers’ preconceptions) have undoubtedly affected our model. The 
study  was  conducted  in  a  homogeneous  sample  of  Belgian  managers, 
and  should  thus  not  be  generalized  to  blue‐collar  workers  or  people 
from  different  cultural  backgrounds  without  further  validation.  Also, 
variations in organizational settings and cultures, as well as industries, 
might  affect  career  success  conceptualizations  of  respondents. 
Furthermore,  the  majority  of  our  sample  were  men,  which  may  also 
have  coloured  the  resulting  findings.  Also,  the  majority  of  our  sample 
was  in  their  thirties,  which  means  that  the  distribution  in  terms  of 
generations,  and  life  and  career  stages  was  limited.  In  order  to  fully 
grasp  the  dynamics  of  career  success  definitions  across  and  within  life 
spans  however,  different  types  of  designs  (i.e.  cross‐sectional  and/or 
longitudinal) would be needed. Finally, although applying the findings of 
Study 1 in a Q‐sort follow‐up study reduced the richness of information 
that  was  obtained  from  the  laddering  interviews,  combining  the 
constructionist approach with more positivist methods made it easier to 
come  to  a  synopsis  of  its  findings.  Notwithstanding  possible 
epistemological  incongruities  in  combining  these  two  paradigms,  any 
qualitative study on the meaning of career success would at some point 
be subject to a loss of information richness, but by using expert analysis 
in  the  Q‐sorts  and,  at  a  later  point  of  analysis,  multiple  raters  (i.e.  the 
three authors) during the content analysis of the data, subjectivity of the 
data reduction was curtailed.  
 
Implications 
 
Implications for research. The most important research implication 
of  a  study  looking  into  the  meaning  of  the  career  success  construct  is 
how  it  should  be  measured  consequently.  The  subjective  career 
measures currently used in research (e.g. the career satisfaction scale by 
289
Greenhaus et al., 1990) have been criticized for their lack of broadness, 
as  they  do  not  encompass  all  possible  dimensions  people  may  use  to 
define and evaluate career success (Arthur et al., 2005; Heslin, 2005).  
 
However,  even  if  it  were  possible  to  come  to  a  list  of  ‘all  possible’ 
idiosyncratic  criteria  used  by  people  to  evaluate  career  success, 
transforming this knowledge into a workable measure of career success 
would  still  remain  quite  difficult.  Further  research  is  necessary  to 
evaluate  how  new  career  measures  can  be  developed  that  are  both 
inclusive and workable. One suggestion is to work with weighted scales, 
in which all possible definitions of career success are incorporated, and 
respondents  must  indicate  both  the  applicability  of  each  item  to  their 
own career, as well as their perceived relevance in light of their personal 
definition  of  career  success  (Heslin,  2005).  With  such  a  design,  both 
differences  in  career  success  (as  measured  by  the  weighted  or 
unweighted  applicability  of  the  constructs  to  respondents’  actual 
careers)  and  differences  in  career  success  definitions  (as  measured  by 
the evaluated relevance of each construct) could be compared between 
groups  (e.g.  men  versus  women,  white‐collar  versus  blue‐collar 
employees,  older  versus  younger  employees).  However,  in  order  to  be 
truly  inclusive,  such  a  measure  would  have  to  contain  the  option  of 
respondents  adding  more  constructs  to  the  scale  while  they  are 
answering  to  it,  should  they  feel  that  not  all  relevant  constructs  have 
been  surveyed.  It  is  clear  that  this  would  cause  some  methodological 
intricacies;  nonetheless  it  would  be  interesting  to  test  this  type  of 
designs in further research, and validate it alongside the other measures 
of  career  success  in  a  variety  of  research  samples.  In  any  case,  it  is 
important  that  careers  researchers  reflect  thoroughly  about  the 
implications different operationalizations of career success may have on 
subsequent research findings. It has been repeatedly demonstrated that 
different  measures  of  career  success  (i.e.  objective  versus  subjective 
measures,  one‐item  measures  versus  multiple‐item  measures)  yield 
different,  sometimes  opposing  conclusions  (Arthur  et  al.,  2005; 
Boudreau  et  al.,  2001;  Hall  &  Chandler,  2005;  Judge  et  al.,  1995;  Nabi, 
1999, 2001).  
 
Another  important  topic  concerns  the  dynamics  of  career  success 
definitions: which contextual systems might influence the career success 
construct  systems  held  by  individuals?  The  generations  literature 
argues  that  historical  evolutions  in  society  would  cause  inter‐
generational  differences  in  conceptions  of  career  and  career  success 
(Mirvis  &  Hall,  1994;  Dries  et  al.,  2008).  On  the  other  hand,  life‐cycle 
processes influence career success construct systems within individuals 
290
as  life  priorities  shift  alongside  their  career  stages  (Lee  et  al.,  2006; 
Poole,  Langan‐Fox  &  Omodei,  1993).  It  is  probable,  for  instance,  that 
career  actors  entering  parenthood  will  emphasize  the  ‘satisfaction’ 
region  more  than  young  graduates,  who  might  be  more  focused  on 
‘performance’  (Lee  et  al.,  2006).  Finally,  the  ‘self‐enforcing’  nature  of 
career  must  also  be  entered  into  the  equation  (Luhmann,  1995). 
Cognitive dissonance reduction processes most likely play a substantial 
role  in  individuals’  constructions  of  career  success  (Dries  et  al.,  2008; 
Sturges,  1999):  to  what  extent  do  individuals  seek  out  careers  that  fit 
their  career  preferences,  or  conversely,  adapt  their  subjective 
definitions of what career success means to their own, objective career 
realities?  Several  past  publications  have  tackled  the  interdependence 
between  the  subjective  and  the  objective  face  of  career  (e.g.  Hall  & 
Chandler,  2005).  Furthermore,  studies  conducted  in  populations  with 
careers  that  digress  from  the  norm  (e.g.  blue‐collar  workers,  part‐time 
workers,  people  that  have  ‘derailed’  or  hit  a  glass  ceiling)  have  been 
found  to  develop  different  definitions  to  better  fit  their  own  situations 
(Dany,  2003;  Hennequin,  2007;  Lee  et  al.,  2006;  Lucas  &  Buzzanell, 
2004;  Nicholson  &  de  Waal‐Andrews,  2005;  Sturges,  1999;  Thomas, 
1989). 
 
Implications  for  practice.  Although  studies  on  the  meanings 
attributed to the career success construct may not be directly applicable 
to  practice,  they  may  help  organizations  design  career  systems,  and 
individuals guide their own career self‐management (Aryee et al., 1994). 
One  of  the  most  significant  consequences  of  reification  for  both 
individuals  and  organizations  is  the  assumption  regarding  what  is  the 
‘normal’ career and hence, what are ‘abnormal’ career patterns (Evetts, 
1992).  One  consequence  of  this  is  the  devaluation  and,  as  a  result, 
inadequate  rewarding  of  those  in  abnormal  careers,  e.g.  experts  in 
‘lateral’ career tracks or those who take a break from paid employment 
for personal reasons. It is important to acknowledge that organizations 
themselves  contribute  significantly  to  the  continuation  of  the  more 
traditional career concept. In their early careers, people’s own personal 
definitions  of  career  success  are  confronted  with  those  of  the 
organizations they work for, which often causes a re‐definition of their 
own goals and roles (Duxbury et al., 1999). The reward policies of most 
organizations  seem  more  accommodating  for  ‘masculine’  goals;  for 
example,  parental  leaves  are  often  penalized  through  a  subsequent 
stagnation  in  the  career  (Duxbury  et  al.,  1999).  However,  at  the  same 
time it is a fact that job security, advancement opportunities and regular 
salary  increases  are  no  longer  available  to  all  employees  (Rousseau, 
1995).  As  a  consequence,  workforce  segmentation  –  e.g.  in  the  form  of 
291
talent  management  programs  (Dries  &  Pepermans,  2008)  –  together 
with a more diversified perception of what career success could mean to 
these  different  groups  of  employees  and  how  their  careers  can  be 
managed, becomes inevitable. An interesting implementation  would be 
for  organizations  to  establish  frameworks  of  what  they  understand 
under  ‘career  success’  utilizing  the  four  quadrants  of  the  model 
presented  in  the  current  paper.  The  adequate  input  for  these 
frameworks  could  be  drawn  from  HR  strategy,  and  the  organization’s 
culture and values. Personnel evaluations could then incorporate these 
organizational career success frameworks, allowing for gap analysis and 
policy adaptations.  
 
References 
 
Abdul‐Ghani, R.,  Poon, J.M.L., Noordin, F.,  Briscoe, J.P. & Jones, E.A. (2007, August). Career success 
from a Malaysian perspective: Doing well by doing good. In J.P. Briscoe (Chair), Doing well by 
doing  good  across  cultures?  A  global  perspective  on  career  success.  Symposium  conducted  at 
the annual meeting of the Academy Of Management, Philadelphia, PA.  
Adamson,  S.J.,  Doherty,  N.  &  Viney,  C.  (1998).  The  meanings  of  career  revisited:  Implications  for 
theory and practice. British Journal of Management, 9 (4), 251‐259. 
Argyris, C. (1960). Understanding organizational behavior. Homewood, IL: Dorsey Press.  
Arthur, M.B. (1994). The boundaryless career: A new perspective for organizational inquiry. Journal 
of Organizational Behavior, 15 (4), 295‐306. 
Arthur,  M.B.,  Hall,  D.T.  &  Lawrence,  B.S.  (1989).  Handbook  of  career  theory.  New  York,  NY: 
Cambridge University Press. 
Arthur,  M.B.,  Khapova,  S.N.  &  Wilderom,  C.P.M.  (2005).  Career  success  in  a  boundaryless  career 
world. Journal of Organizational Behavior, 26 (2), 177‐202.  
Aryee,  S.,  Chay,  Y.W.  &  Tan,  H.H.  (1994).  An  examination  of  the  antecedents  of  subjective  career 
success among a managerial sample in Singapore. Human Relations, 47 (5), 487‐510. 
Baruch,  Y.  (2004).  Transforming  Careers:  From  Linear  to  Multidirectional  Career  Paths.  Career 
Development International, 9 (1), 58‐73. 
Betz, N., & Fitzgerald, L. (1987). The career psychology of women. Orlando, FL: Academic Press.  
Blustein,  D.L.  (2001).  Extending  the  reach  of  vocational  psychology:  Toward  an  inclusive  and 
integrative psychology of working. Journal of Vocational Behavior, 59 (2), 171‐182.  
Briscoe,  J.P.,  Hall,  D.T.,  DeMuth,  R.L.F.  (2006).  Protean  and  boundaryless  career  attitudes:  An 
empirical exploration. Journal of Vocational Behavior, 69 (1), 30‐47. 
Briscoe,  J.P.,  Hall,  D.T.,  Las  Heras,  M.  &  Unite,  J.A.  (2007,  August).  Doing  well  and  doing  good: 
Equations for career success in the United States. In J.P. Briscoe (Chair), Doing well by doing 
good  across  cultures?  A  global  perspective  on  career  success.  Symposium  conducted  at  the 
annual meeting of the Academy Of Management, Philadelphia, PA. 
Borg, I., & Groenen, P. (1997). Modern multidimensional scaling: Theory and applications. New York, 
NY: Springer.  
Boudreau, J.W., Boswell, W.R., & Judge, T.J. (2001). Effects of personality on executive career success 
in the U.S. and Europe. Journal of Vocational Behavior, 58 (1), 53‐81.  
Bourne, H. & Jenkins, M. (2005). Eliciting managers’ personal values: An adaptation of the laddering 
interview method. Organizational Research Methods, 8 (4), 410‐428.  
Chen,  C.  (1997).  Career  projection:  Narrative  in  context.  Journal  of  Vocational  Education  and 
Training, 49 (2), 311‐326. 
Chudzikowski,  K.,  Demel,  B.  &  Mayrhofer,  W.  (2007,  August).  Do  I  look  as  if  I  care?  Doing  well  in 
careers of three professional groups on Austria – A qualitative study. In J.P. Briscoe (Chair), 
Doing well by doing good across cultures? A global perspective  on career success.  Symposium 
conducted at the annual meeting of the Academy Of Management, Philadelphia, PA. 
Collin,  A.  &  Young,  R.A.  (1992).  Constructing  career  through  narrative  and  context:  An 
interpretative perspective. In R.A. Young & A. Collin (Eds.). Interpreting career: Hermeneutical 
studies of lives in context (pp. 1­12). Westport, CT: Praeger. 
292
Dany,  F. (2003). 'Free actors' and  organizations: Critical remarks about the new career literature, 
based on French insights. International Journal of Human Resource Management, 14 (5), 821‐
838.  
Davidson, M.J. & Cooper, C.L. (1992). Shattering the Glass Ceiling: The Woman Manager. London, UK: 
Paul Chapman Publishing. 
Derous,  E.,  De  Witte,  K.  &  Stroobants,  R.  (2003).  Testing  the  Social  Process  model  on  Selection 
through expert analysis. Journal of Occupational and Organizational Psychology, 76, 179‐199. 
Duxbury, L., Dyke, L., & Lam, N. (1999). Career development in the federal public service: Building a 
world­class workforce. Ottawa, Canada: Treasury Board of Canada Secretariat. 
Dyke, L.S. & Murphy, S.A. (2006). How we define success: A qualitative study of what matters most 
to women and men. Sex Roles ­ A Journal of Research, 55 (5), 357‐372.  
Dries,  N.  &  Pepermans,  R.  (2008).  "Real"  high  potential  careers:  An  empirical  study  into  the 
perspectives of organizations and high potentials. Personnel Review, 37 (1), 85‐108. 
Dries, N., Pepermans, R. & De Kerpel, E. (2008). Exploring four generations' beliefs about career:  Is 
‘satisfied’ the new ‘successful’? Journal of Managerial Psychology, 23 (8), 907‐928. 
Evetts, J. (1992). Dimensions of Career: Avoiding Reification in the Analysis of Change. Sociology, 26 
(1), 1‐21. 
Evans,  M.G.,  Gunz,  H.P.  &  Jalland,  R.M.  (1997).  Implications  of  organizational  downsizing  for 
managerial careers. Canadian Journal of Administrative Sciences, 14, 359‐371. 
Fugate,  M.,  Kinicki,  A.J.  &  Ashfort,  B.E.  (2004).  Employability:  A  psycho‐social  construct,  its 
dimensions, and applications. Journal of Vocational Behavior, 65 (1), 14‐38.  
Gattiker,  U. E.,  &  Larwood,  L. (1986). Subjective career success: A study of managers  and support 
personnel. Journal of Business and Psychology, 1, 78‐94. 
Gallos,  J.V.  (1989).  Exploring  women’s  development:  implications  for  career  theory,  practice  and 
research.  In  M.B.  Arthur,  D.T.  Hall.  &  B.S.  Lawrence  (Eds.),  Handbook  of  Career  Theory  (pp. 
110­132). Cambridge, UK: Cambridge University Press. 
Gilligan, C. (1982). In a Different  Voice: Psychological Theory and  Women’s Development. Cambridge, 
MA: Harvard University Press.  
Greenhaus,  J.H.,  Parasuraman,  S.  &  Wormley,  W.M.  (1990).  Effects  of  race  on  organizational 
experiences,  job  performance  evaluations,  and  career  outcomes.  Academy  of  Management 
Journal, 33 (1), 64‐86. 
Hall, D. T. (1976), Careers in organizations. Glenview, IL: Scott, Foresman and Company. 
Hall, D.T. (2002), Careers in and out of organizations. Thousand Oaks, CA: Sage Publications. 
Hall,  D.T.  &  Chandler,  D.E.  (2005).  Psychological  success:  When  the  career  is  a  calling.  Journal  of 
Organizational Behavior, 26 (2), 155‐176.  
Hennequin,  E.  (2007).  What  'career  success'  means  to  blue‐collar  workers.  Career  Development 
International, 12 (6), 565‐581.  
Heslin,  P.A.  (2005).  Conceptualizing  and  evaluating  career  success.  Journal  of  Organizational 
Behavior, 26 (2), 113‐136. 
Judge,  T.A.,  Cable,  D.M.,  Boudreau,  J.W.  &  Bretz,  R.D.  (1995).  An  empirical  investigation  of  the 
predictors of executive career success. Personnel Psychology, 48 (3), 485–519. 
Juntunen,  C.L.,  Barraclough,  D.J.,  Broneck,  C.L.,  Seibel,  G.A.,  Winrow,  S.A.,  &  Morin,  P.M.  (2001). 
American Indian perspectives on the career journey. Journal of Counseling Psychology, 48(3), 
274‐285. 
Kelly, G. A. (1955). The psychology of personal constructs. New York, NY: Norton.  
King, Z. (2004). Career self‐management: Its nature, causes and consequences. Journal of Vocational 
Behavior, 65, 112–133. 
Kruskal, J.B. & Wish, M. (1978). Multidimensional scaling. Beverly Hills, CA: Sage. 
Las  Heras,  M.  &  Zikic,  J.  (2007,  August).  To  feel  successful  in  one’s  career  by  doing  good:  Career 
success  in  Spain.  In  J.P.  Briscoe  (Chair),  Doing  well  by  doing  good  across  cultures?  A  global 
perspective on career success. Symposium conducted at the annual meeting of the Academy Of 
Management, Philadelphia, PA. 
Lee, M.D., Lirio, P., Karakas, F., MacDermid, S.M., Buck, M.L. & Kossek, E.E. (2006). Exploring career 
and personal outcomes and the meaning of career success among part‐time professionals in 
organizations. In R.J. Burke (Ed.), Research companion to work hours and work addiction (pp. 
284­309). Cheltenham, UK: Edward Elgar. 
Lucas, K., & Buzzanell, P. M. (2004). Blue‐collar work, career, and success: Occupational narratives 
of sisu. Journal of Applied Communication Research, 32, 273‐292. 
293
Lucas,  K.,  Liu,  M.,  &  Buzzanell,  P.  M.  (2006).  No  limits  careers:  A  critical  examination  of  career 
discourse in the U.S. and China. In M. Orbe, B. J. Allen, & L. A. Flores (Eds.), International and 
intercultural communication annual 28 (pp. 217­242). Thousand Oaks, CA: Sage.  
Luhmann, N. (1995). Social systems. Stanford, CA: Stanford University Press. 
Mallon,  M.  &  Cohen,  L.  (2000,  July).  My  brilliant  career?  Using  stories  as  a  methodological  tool  in 
careers  research.  Paper  presented  at  the  annual  International  Conference  on  Organizational 
Discourse, London, UK.  
Melamed,  T.  (1996).  Career  success:  an  assessment  of  a  gender‐specific  model.  Journal  of 
Occupational and Organizational Psychology, 69 (3), 217–235. 
Mirvis,  P.H.  &  Hall,  D.T.  (1994).  Psychological  success  and  the  boundaryless  career.  Journal  of 
Organizational Behavior, 15 (4), 365‐380.  
Nabi,  G.  (1999).  An  investigation  into  the  differential  profile  of  predictors  of  objective  and 
subjective career success. Career Development International, 4 (4), 212‐224.  
Nabi,  G.R.  (2001).  The  relationship  between  HRM,  social  support,  and  subjective  career  success 
among men and women. International Journal of Manpower, 22 (5), 457–474. 
Ng, T., Eby,  L. T.,  Sorensen,  K.  & Feldman, D. (2005). Predictors  of objective and subjective  career 
success: A meta‐analysis. Personnel Psychology, 58 (2), 367‐408.  
Nicholson,  N.  (2000)  Motivation‐Selection‐Connection:  An  Evolutionary  Model  of  Career 
Development.  In  M.  Peiperl,  M.  Arthur,  R.  Goffee,  &  T.  Morris  (Eds.).  Career  Frontiers:  New 
concepts of working life (pp. 54­75). Oxford, UK: Oxford University Press.  
Nicholson, N. & de Waal‐Andrews, W. (2005). Playing to win: Biological imperatives, self‐regulation, 
and trade‐offs in the game of career success. Journal of Organizational Behavior, 26 (2), 137‐
154.  
Ogliastri, E., Caballero, K. Rodríguez, C. & Sabando, C. (2007, August). Doing well by doing good in 
personal careers: career success in Costa Rica. In J.P. Briscoe (Chair), Doing well by doing good 
across  cultures?  A  global  perspective  on  career  success.  Symposium  conducted  at  the  annual 
meeting of the Academy Of Management, Philadelphia, PA. 
Parker  P.  (2006).  Card  sorts:  A  process  of  constructing  personal  narrative.  In  M.  McMahon,  &  W. 
Patton  (Eds.).  Career  Counselling:  Constructivist  approaches  (pp.  176­186).  New  York,  NY: 
Routledge. 
Parker  B.  &  Chusmir  L.H.  (1990).  A  generational  and  sex‐based  view  of  managerial  work  values. 
Psychological Reports, 66, 947‐951.  
Poole  M.E.,  Langan‐Fox  J.,  &  Omodei  M.  (1993).  Contrasting  subjective  and  objective  criteria  as 
determinants  of  perceived  career  success.  Journal  of  Occupational  and  Organizational 
Psychology. 66 (1), 39‐54. 
Powell,  G.N.  &  Mainiero,  L.A.  (1992).  Cross‐currents  in  the  river  of  time:  Conceptualizing  the 
complexities of women’s careers. Journal of Management, 18215–18237. 
Reitman,  F.,  Schneer,  J.A.  (2003).  The  promised  path:  a  longitudinal  study  of  managerial  careers. 
Journal of Managerial Psychology, 18 (1), 60‐75.  
Rousseau,  D.M.  (1995),  Psychological  Contracts  in  Organizations:  Understanding  Written  and 
Unwritten Agreements. Newbury Park, CA: Sage. 
Russo, N., Kelly, M. & Deacon, M. (1991). Gender and sex related attribution: beyond individualistic 
conceptions of achievement. Sex Roles, 25 (5­6), 331‐350. 
Savickas, M. (1995). Current theoretical issues in vocational psychology: convergence, divergence, 
and schism. In W.B. Walsh & S.H. Osipow (Eds.), Handbook of Vocational Psychology: theory, 
research and practice (2nd Edition, pp. 1­34), Mahwah, NJ : Lawrence Erlbaum Associates. 
Savickas, M. L. (2000). Renovating the psychology of careers for the 21st century. In A. Collin & R. 
Young (Eds.), The future of career (pp. 53­68). Cambridge, UK: Cambridge University Press.  
Savickas,  M.L.  (2005).  The  theory  and  practice  of  career  construction.  In  S.S.  Brown  &  R.W.  Lent 
(Eds.) Career Development and Counseling: Putting Theory And Research To Work (pp. 42­70). 
Hoboken, NJ: John Wiley & Sons. 
Schein,  E.H.  (1978).  Career  dynamics:  Matching  individual  and  organizational  needs.  Reading,  MA: 
Addison‐Wesley.  
Schein, E. H. (1990), Career Anchors (Rev. Ed.), Pfeiffer, Inc, San Diego. 
Shen, Y., Hall, D.T. & Fei, Z. (2007, August). The evolution of career success: Doing well in China. In 
J.P.  Briscoe  (Chair),  Doing  well  by  doing  good  across  cultures?  A  global  perspective  on  career 
success.  Symposium  conducted  at  the  annual  meeting  of  the  Academy  Of  Management, 
Philadelphia, PA. 
Spence, I. & Ogilvie, J.C. (1973). A table of expected stress values for random rankings in nonmetric 
multidimensional scaling. Multivariate Behavioral Research, 14, 355‐365. 
294
Stead,  G.B.  (2004).  Culture  and  career  psychology:  A  social  constructionist  perspective.  Journal  of 
Vocational Behavior, 64 (3), 389‐406.  
Sturges, J. (1999). What it means to succeed: Personal conceptions of career success held by male 
and female managers at different ages. British Journal of Management, 10 (3), 239‐252.  
Sullivan,  S.E.  (1999).  The  changing  nature  of  careers:  a  review  and  research  agenda.  Journal  of 
Management, 25 (3), 457‐484. 
Super,  D.E.  (1980).  A  life  span,  life  space  approach  to  career  development.  Journal  of  Vocational 
Behavior, 13, 282‐298. 
Taniguchi,  M.  (2007,  August).  Change  and  continuity  in  career  success  in  Japan.  In  J.P.  Briscoe 
(Chair),  Doing  well  by  doing  good  across  cultures?  A  global  perspective  on  career  success. 
Symposium conducted at the annual meeting  of the Academy Of Management, Philadelphia, 
PA. 
Thomas,  R.J.  (1989).  Blue‐collar  careers:  meaning  and  choice  in  a  world  of  constraints.  In  M.B. 
Arthur, D.T. Hall & B.S. Lawrence (Eds.). Handbook of Career Theory (pp. 354­379), Cambridge, 
UK: Cambridge University Press.  
Verbruggen,  M.,  Sels,  L.  &  Forrier,  A.  (2007).  Unraveling  the  relationship  between  organizational 
career  management  and  the  need  for  external  career  counselling.  Journal  of  Vocational 
Behavior, 71 (1), 69 ‐ 83. 
Wagenaar,  W.A.  &  Padmos,  P.  (1971).  Quantitative  interpretation  of  stress  in  Kruskal’s 
multidimensional  scaling  technique.  British  Journal  of  Mathematical  and  Statistical 
Psychology, 24, 101‐110. 
Wiese,  B.  S.,  Freund,  A.  M.,  &  Baltes,  P.  B.  (2002).  Subjective  career  success  and  emotional  well‐
being: Longitudinal predictive power of selection, optimization, and compensation. Journal of 
Vocational Behavior, 60, 321‐335. 
Wilensky, H.L. (1961). Careers, lifestyles, and social integration. International Social Science Journal, 
12 (4), 553–558. 
Young,  R.A.  &  Collin,  A.  (2004).  Introduction:  Constructivism  and  social  constructionism  in  the 
career field. Journal of Vocational Behavior, 64 (3), 373‐388.  
Young R.A., Valach L., Collin A. (1996). A contextualist approach to career analysis and counselling. 
In D. Brown, L. Brooks (Eds.). Career Choice and Development (3rd edition, pp. 477­512). San 
Francisco, CA: Jossey‐Bass. 
Appendix. Synopsis of the proposed career success model 
 
Quadranta  Region  Q sorts  δL Ni Nε CIMb 
I  1 Performance  13 Successfully achieving envisioned goals  13 6 30 94
      16 Getting into a self‐perpetuating cycle of motivation, action, results and 
growth 
      20 Performing well, delivering efforts and getting results; demonstrating 
that one does a good job 
      23 Going to great lengths to achieve good things
      26 Staying motivated and driven, continuously pursuing new goals 
      37 Standing out; being the best at what one does
  2 Advancement  1 Being ambitious; wanting to progress, horizontally as well as vertically  10 3 30 90
      2 Experiencing plenty of opportunities to build a career, to grow within the 
organization 
      30 Getting promoted; climbing the ladder
II  3 Self‐development    14 9 30 96
295

      a Goal attainment 8 1 30 7 6
      4 Accomplishing one’s own goals
      b Continuous learning 10 5 30 94
      10 Going through personal growth and development as a result of work 
experiences 
      18 Achieving personal development by on‐the‐job learning and through 
various experiences 
      19 Continuously learning in various ways so that one can continue to 
develop as a person and get ahead 
      25 Acquiring many experiences, inside one’s job and out
      42 Acquiring knowledge
Notes. a I = inter‐personal – achievement; II = intra‐personal – achievement; III = intra‐personal – affect; IV = inter‐personal – affect; b CIM = 
consistency‐in‐meaning index (Derous et al., 2003). 
Appendix. Continued 
 
Quadranta  Region  Q sorts  δL Ni Nε CIMb 
      c Self­expression 8 1 30 7 6
      22 Being able to express oneself through work and accomplishment 
      d Career self­management 12 2 30 82
      39 Being able to manage and shape one’s own career; achieving everything 
single‐handedly 
      41 Creating opportunities in life
  4 Creativity  38 Working creatively; accomplishing innovative, extraordinary ideas  10 1 30 69
III  5 Security  7 Having the freedom to buy what one desires; not experiencing financial  12 2 30 82
restrictions 
      28 Experiencing stability and (job) security
  6 Satisfaction    13 8 30 95
      a Work­life balance 9 4 30 94
      3 Being satisfied with one’s work‐life balance
296

      5 Being happy with all that one has and with all the possibilities present 
      9 Feeling healthy and happy, at home as well as at work
      12 Allowing one’s family to play a significant role in life alongside one’s 
career 
      b Achievement satisfaction 11 3 30 89
      11 Experiencing satisfaction and fulfillment brought about by one’s own 
personal achievements 
      24 Feeling good; experiencing a warm and fuzzy feeling on the inside 
      36 Being proud of oneself and one’s achievements
      c Independence 10 1 30 69
      40 Pursuing independence as a major source of happiness
Notes. a I = inter‐personal – achievement; II = intra‐personal – achievement; III = intra‐personal – affect; IV = inter‐personal – affect; b CIM = 
consistency‐in‐meaning index (Derous et al., 2003). 
Appendix. Continued 
 
Quadranta  Region  Q sorts  δL Ni Nε CIMb 
IV  7 Recognition  6 Being recognized for one’s accomplishments; proving one’s worth  14 3 30 86
      31 Being respected and appreciated by one’s co‐workers
      33 Receiving a good, equitable compensation package; getting what 
one deserves, in line with one’s performance 
  8 Cooperation  8 Having a good understanding with one’s employer – one that does  12 3 30 88
not always require words 
      21 Emphasizing people, as they make the difference in an organization 
      32 Being amongst other people and cooperate with them
  9 Contribution      13 7 30 95
      a Experienced contribution 12 3 30 88
      15 Personally contributing to the welfare of the organization ‐ and as 
such, to that of society as a whole 
      27 Believing to have achieved positive matters, in an ethical way 
297

      29 Realizing that one person can make a world of difference in an 
organization 
I      b Factual contribution 13 4 30 90
      14 Demonstrating that one is a valuable asset to the organization 
      17 Steering a work team into a self‐perpetuating cycle of motivation, 
action, results and growth 
      34 Being able to exert power or influence; making one’s mark 
      35 Being an important link in the process that leads to organizational 
success 
Notes. a I = inter‐personal – achievement; II = intra‐personal – achievement; III = intra‐personal – affect; IV = inter‐personal – affect; b CIM = 
consistency‐in‐meaning index (Derous et al., 2003). 
 
 
298
 
299
 
 
 
 
Discussion 
 
 
In  this  dissertation,  eight  studies  were  covered  within  four  chapters, 
each approaching the topic of talent management within the context of 
the  postmodern  career  from  a  different  vantage  point.  The  first  six 
studies  focused  on  talent  management  variables  explicitly  (i.e.  the 
criteria  used  in  high  potential  identification  procedures,  and  career 
outcomes  for  employees  identified  as  high  potentials).  Both 
organizational‐level  variables  (e.g.  open  or  closed  communication 
strategies  concerning  talent  management  procedures  within  the 
organization)  and  individual‐level  variables  (e.g.  organizational 
commitment) were studied. The last two studies looked at the meanings 
individual  career  actors  attribute  to  the  constructs  ‘career’  and  ‘career 
success’.  Thus,  the  postmodern  career  was  addressed  in  this 
dissertation both as context and as focus of research, respectively.   
 
Table II, presented below, summarizes the findings of each of the eight 
studies  briefly.  It  also  provides  an  overview  of  the  academic  output 
generated  by  this  dissertation  project  in  terms  of  publications.  As  the 
table shows, four out of the eight papers were published at the time of 
dissertation  submission,  one  was  in  second  review,  and  three  in  first 
review 5 .  
 
In this concluding chapter, we will first go over the main findings of each 
of  the  eight  papers,  linking  these  back  to  the  research  framework 
depicted in Figures I to IX throughout this dissertation. We will discuss 
the main findings of this dissertation per topic, rather than per chapter, 
in  order  to  achieve  an  integrative  overview.  We  then  continue  this 
Discussion  by  addressing  contributions  and  limitations  of  the  current 
dissertation.  We  finish  with  directions  for  further  research  and 
implications for practice.  

                                                            
5 For a full overview of  past and ongoing projects and publications, please consult the

author's Vita.  
Table II. Overview: main findings and publication output 
 
Chapter  Paper  Main findings  Status  Journal 

I . Exploring ‘real’ high potential careers   
  1. ‘Real’ high potential careers: An empirical study into the perspectives of organizations and high potentials
    High  upward  intra‐organizational  mobility  and  low  inter‐organizational  mobility  arose  as  Published PR
key features in both high potentials’ and organizations’ views about high potential careers. 
High  potentials  stated  not  to  actively  pursue  career  opportunities  outside  of  their 
organizations. They believed their chances for rapid advancement were higher within their 
own  organization  due  to  the  inherent  career  ‘promises’  encapsulated  in  the  high  potential 
label.  The organizational representatives, from their side, reported that they still preferred 
to recruit internal successors for top management positions.  
II. High potential identification   
  2.  Identification of leadership potential: Is there consensus about ‘the’ criteria?
300

    A  model  of  leadership  potential  was  conceptualized,  comprising  77  items  mapped  onto  a  R & R HRM
two‐dimensional structure (cognition‐conation vs. extrapersonal‐intrapersonal). The model 
was found to have 4 factors (information processing, learning agility, drive, and charismatic 
leadership)  and  13  subfactors  (input‐seeking  behavior,  insightfulness,  decision  making, 
problem  solving,  willingness  to  learn,  emotional  stability,  personal  adaptability,  drive  for 
results,  persistence,  personal  dedication,  people  management,  inspirational  management, 
and  stakeholder  management).    High  consensus  was  found  among  top  managers,  line 
managers and HR managers about the practical relevance of the proposed model. 
 
Notes.  R  &  R  =  revise  and  resubmit  (i.e.  second  review);  PR  =  Personnel  Review;  HRM  =  Human  Resource  Management;  LODJ  =  Leadership  and 
Organization Development Journal; EJWOP = European Journal of Work and Organizational Psychology; JOB = Journal of Organizational Behavior; 
JOMP = Journal of Managerial Psychology; JVB = Journal of Vocational Behavior.  
Table II. Continued 
 
Chapter  Paper  Main findings  Status  Journal 
 
  3. Using emotional intelligence to identify high potentials: A metacompetency perspective 
    Assertiveness, independence, optimism, flexibility and social responsibility (all subscales of  Published LODJ
the  EQ‐i)  were  found  to  differ  between  high  potentials  and  non‐high  potentials. 
Furthermore,  we  found  that    high  potentials  at  lower  managerial  levels  displayed  lower 
levels of interpersonal skills and happiness at work than non‐high potentials, but that these 
relationship  was  inverted  at  higher  managerial  levels.  Finally,  results  indicated  that    high 
potentials displayed higher levels of job performance and less boundaryless career attitudes. 
 
  4. High potential identification: Examining the developmental perspective
    Employees identified as high potentials were found to demonstrate higher overall learning  In review PR
agility than non‐high potentials. Furthermore, this effect was only partially mediated by on‐
the‐job learning,  indicating  that  the  effect  could  not  be  fully  explained  by  actual,  observed, 
301

learning behavior on the job. Finally, learning agility was found to be a more or less stable 
trait  that  does  not  increase  with  age  or  experience.  In  addition,  people  with  high  learning 
agility tend to seek out more learning opportunities by pursuing career variety. 
 
Notes.  R  &  R  =  revise  and  resubmit  (i.e.  second  review);  PR  =  Personnel  Review;  HRM  =  Human  Resource  Management;  LODJ  =  Leadership  and 
Organization Development Journal; EJWOP = European Journal of Work and Organizational Psychology; JOB = Journal of Organizational Behavior; 
JOMP = Journal of Managerial Psychology; JVB = Journal of Vocational Behavior.  
 
Table II. Continued 
 
Chapter  Paper  Main findings  Status  Journal 
   
III. Career outcomes in high potential careers   
  5.  The role of employability and firm­specific capital in shaping the psychological contract: Do high potentials and long­
tenured employees get the better deal? 
     No  consistent  support  was  found  for  the  assumption  that  more  employable  employees  In review EJWOP
    perceive  weaker  psychological  contract  obligations  in  terms  of  job  security  and 
organizational loyalty. The idea that the downfall of the organizational career and decreases 
in job security could somehow be counterbalanced by developing individuals’ employability  
seems overly simplistic. Furthermore, high potentials and long‐tenured employees perceived 
stronger obligations from the side of the  employer to provide job security, than from their 
side to stay with their employer, particularly under conditions of high employability. These 
findings  warrant  attention  towards  the  central  role  of  employees’  bargaining  position  in 
shaping the psychological contract. 
302

 
  6. Effects of the high potential label on performance, career success and commitment: A matter of communication?
    We  found  positive  main  effects,  both  of  being  identified  as  a  high  potential  and  of  In review JOB
organizational  communication  openness,  on  job  performance,  career  success,  and 
commitment.  Furthermore,  we  found  interaction  effects  of  being  identified  as  a  high 
potential  with  communication  openness  on  salary,  satisfaction  with  performance,  and 
satisfaction with job security. Overall, results indicate that adopting an open communication 
strategy about talent management generates the most desirable outcomes. 
 
Notes.  R  &  R  =  revise  and  resubmit  (i.e.  second  review);  PR  =  Personnel  Review;  HRM  =  Human  Resource  Management;  LODJ  =  Leadership  and 
Organization Development Journal; EJWOP = European Journal of Work and Organizational Psychology; JOB = Journal of Organizational Behavior; 
JOMP = Journal of Managerial Psychology; JVB = Journal of Vocational Behavior.  
Table II. Continued 
 
Chapter  Paper  Main findings  Status  Journal 
   
IV. Constructing ‘career’   
  7. Exploring four generations' beliefs about career: Is ‘satisfied’ the new ‘successful’?
    The  majority  of  respondents  still  had  rather  ‘traditional’ careers,  although  younger  Published JOMP
generations  did  seem  to  exhibit larger discrepancies between their career preferences  and 
their  actual  career  situation.  Overall,  satisfaction  appeared  to  be  the  overriding  criterion 
used to evaluate other (fictitious) people’s career success. No significant differences in terms 
of  career  success  evaluation  heuristics  were  found  between  generations.  With  regard  to 
importance  attached  to  organizational  security,  the  oldest  and  the  youngest  generations 
scored significantly higher than the other generations. 
 
  8. Career success: Constructing a multidimensional model
    A model incorporating 42 idiosyncratic meanings interviewees attached to the construct of  Published JVB
303

    ‘career  success’  was  obtained.  A  two‐dimensional  configuration  (affect  –  achievement  vs.  


intra‐personal  –  inter‐personal)  comprising  9  regions  (performance,  advancement,  self‐
development,  creativity,  security,  satisfaction,  recognition,  cooperation  and  contribution) 
was concluded upon as the optimal solution. 
Notes.  R  &  R  =  revise  and  resubmit  (i.e.  second  review);  PR  =  Personnel  Review;  HRM  =  Human  Resource  Management;  LODJ  =  Leadership  and 
Organization Development Journal; EJWOP = European Journal of Work and Organizational Psychology; JOB = Journal of Organizational Behavior; 
JOMP = Journal of Managerial Psychology; JVB = Journal of Vocational Behavior.  
304
Overview of study findings 
 
High potential identification 
 
Conceptual  issues.  Mainly  in  paper  1,  some  empirical  input  was 
provided as to how organizations (and individual career actors, as well) 
define  ‘high  potentials’.  The  interview  data  confirmed  that  high 
potentials are seen as (and see themselves as) employees making up an 
elite  employee  segment  destined  for  upward  progression  on  the 
corporate  ladder.  This  finding  mirrors  the  different  definitions  of  high 
potentials found in the literature (e.g. Cope, 1998; Iles, 1997; Roussillon 
& Bournois, 2002).  
 
A  related  finding  was  that  organizations,  in  a  response  to  structural 
decreases in promotion opportunities, have become even more elitist in 
their  definition  of  what  a  high  potential  is.  Organizations  today 
deliberately  identify  a  smaller  percentage  of  their  workforce  as  high 
potentials  compared  to  a  decade  ago.  This  finding  seems  to  confirm 
some of the claims in the literature about the effects of downsizing and 
delayering on talent management policies (e.g. Baruch & Peiperl, 1997).  
 
Another  interesting  finding  is  that,  apart  from  engaging  in  workforce 
segmentation  to  separate  high  potentials  from  average  performers  or 
contributors,  organizations  identify  different  segments  within  the  pool 
of  high  potentials,  as  well.  Horizontal  as  well  as  vertical  talent  pool 
segmentation  is  applied  by  organizations  in  order  to  further 
differentiate  between  the  developmental  and  reward  needs  of  the 
different talent segments. Horizontal segmentation means that all‐round 
managerial  talent  is  distinguished  from  key  experts  and  project 
workers.  In  order  to  accommodate  the  needs  of  the  latter  groups, 
organizations have incorporated ‘dual’ or even ‘triple’ career tracks into 
their  opportunity  structures  (e.g.  Rothwell  &  Poduch,  2004).  However, 
most  organizations  willingly  admit  that  their  non‐managerial  career 
tracks are not (yet) in a mature stage of implementation. As for vertical 
segmentation, organizations report that they distinguish between early‐
career high potentials and more experienced high potentials (sometimes 
referred  to  as  ‘top  potentials’,  see  Vinkenburg  &  Pepermans,  2005). 
Typically,  however,  high  potentials  tend  to  be  younger  staff,  since 
organizations  often  include  age‐related  criteria  in  their  definition  of 
high potentials. It thus appears that ‘potential’ is associated with ‘youth’, 
an  issue  that  might  prove  disadvantageous  in  a  labor  market 
increasingly characterized by an aging workforce (e.g. Sels, Van Woensel 
& Herremans, 2008).  
305
Interestingly,  the  high  potentials  themselves  acknowledge  that  their 
status  is  organization‐specific,  and  that  they  might  not  be  considered 
high  potentials  in  other  organizations.  This  factor  might  play  a  role  in 
the retention process (Cope, 1998).  
 
Identification  criteria.  Throughout  this  dissertation,  a  variety  of 
possible high potential identification criteria were identified. Overall, we 
found  that  although  high  performance  is  a  sine  qua  non  (i.e.  a  high 
potential is per definition also a high performer), it should never be the 
sole  criterion  for  high  potential  identification.  Letting  raters  assess 
performance  and  potential  at  the  same  time  in  the  appraisal  cycle  is 
strongly discouraged, as it increases the risk of halo bias, i.e. a wrongful 
alignment  of  potentiality  with  performance  ratings  (e.g.  Fields,  2002; 
Pepermans,  Vloeberghs  &  Perkisas,  2003;  Segalla,  Rouziès  &  Flory, 
2001). 
 
Paper 2, in particular, focused on establishing a comprehensive model of 
criteria  used  to  assess  leadership  potential  based  on  a  review  of  the 
relevant  literature  from  the  last  20  years.  In  this  paper,  a  two‐
dimensional  model  was  constructed.  The  horizontal  axis  was  labeled 
‘extrapersonal‐intrapersonal’  (or:  ‘context‐self’);  the  vertical  axis  was 
labeled  ‘cognition‐conation’  (or:  ‘head‐heart’).  Whereas  ‘context’  refers 
to the ability to collect input from,  and generate  impact on the context 
within which one functions, ‘self’ refers to the exploration and pursuit of 
personal  goals.  ‘Head’  refers  to  mental  ability  and  information‐
processing  capacities;  ‘heart’  to  motivation,  drive  and  willpower.  Each 
quadrant  contained  multiple  high  potential  identification  criteria. 
Although the overview of high potential identification criteria in paper 2 
is,  without  a  doubt,  the  most  inclusive  in  this  dissertation,  some  of  the 
other papers identified issues related to high potential identification as 
well.  
 
Paper  3  and  4  focused  explicitly  on  the  utility  of  incorporating 
‘metacompetencies’  in  high  potential  identification  processes.  Paper  3 
focused  on  emotional  intelligence  (EQ)  as  a  possible  identification 
criteria,  and  paper  4  focused  on  learning  agility  (LQ).  Both  EQ  and  LQ 
are metacompetencies, in that they are ‘overarching’ competencies that 
enable individual career actors to develop new competencies when their 
environments demand change or adaptation (Briscoe & Hall, 1999). The 
results  of  these  studies  indicate  that  there  are  indeed,  differences 
between high potentials and non‐high potentials in terms of EQ and LQ. 
The  study  on  EQ  found  that  high  potentials  are  more  assertive, 
independent, optimistic, and flexible than non‐high potentials. However, 
306
the  study  also  found  that  high  potentials  at  lower  functional  level 
demonstrate  lower  levels  of  interpersonal  skills  than  non‐high 
potentials,  although  this  relationship  is  inverted  at  higher  functional 
levels. Possibly, early‐career stage high potentials are mainly evaluated 
on performance criteria, leaving little time or space for the development 
of interpersonal relationships on the work floor. The study on LQ found 
that  LQ  is  predictive  of  being  identified  as  a  high  potential  above  and 
beyond  job  performance.  Furthermore,  on‐the‐job  learning  partially 
mediated  the  effect  of  LQ  on  assigned  ‘talent  category’.  However,  only 
on‐the‐job learning in terms of learning about technical aspects of a job 
mediated  the  relationship.  Learning  to  interact  better  with  others 
(which is another aspect of on‐the‐job learning) was not a predictor of 
being identified as a high potential or not – which is a similar finding to 
that  of  the  EQ  study.  It  thus  appears  that  organizations  tend  to 
underemphasize  interpersonal  skills  in  identification  processes  of  high 
potentials.  This  might  be  problematic  considering  the  fact  that  high 
potentials will, inevitably, have to take up leadership roles at one point 
in  their  career  development  (e.g.  Dulewicz,  Young  &  Dulewicz,  2005; 
Van Velsor & Leslie, 1995).  
 
As  for  the  question  of  high  potentials  possessing  ‘innate’  qualities 
separating  them  from  others,  or  rather,  a  series  of  developable 
(meta)competencies,  our  findings  were  inconclusive.  Whereas  high 
potentials  themselves  seemed  to  define  leadership  potential  as  an 
‘innate’  quality  that  not  everyone  can  develop,  their  organizations 
described  high  potentials  as  mainly  characterized  by  a  strong  drive  to 
work  long  hours  and  be  self‐managed  (paper  1).  Paper  4  explicitly 
examined  whether  LQ  was  linked  to  age,  work  experience,  educational 
level,  and  career  variety.  No  relationship  was  found  to  age  or  work 
experience; educational level and career variety were positively related 
to learning agility. However, rather than concluding that learning agility 
can  be  developed  by  education  or  varied  career  experiences,  the 
opposite  conclusion  is  also  plausible.  When  we  assume  a  reversed 
causality  scenario,  the  same  findings  imply  that  LQ  is  more  of  a 
dispositional  characteristic  (unrelated  to  age  or  work  experience,  like 
IQ), and that individuals with high LQ tend to achieve higher educational 
levels and pursue a variety of career experiences.  
   
Finally,  as  concerns  the  actors  accountable  for  high  potential 
identification,  we  found  that  line  managers,  HR  managers  and  top 
managers  (possibly  in  the  form  of  a  selection  committee)  are  typically 
the parties involved. Generally speaking, line managers provide bottom‐
up information about their subordinates’ performance and potential; HR 
307
managers  are  (partly)  accountable  for  developing  the  criteria  and 
procedures  for  identifying  high  potentials;  and  top  managers  provide 
top‐down information (i.e. about the organization’s talent management 
strategy) and hold decision making capacity and veto right (paper 2). An 
interesting point that emerged from study 1 was that the high potentials 
themselves  often  also  played  a  role  in  their  own  identification.  Often, 
these  employees  communicate  their  aspirations  and  ambitions  very 
clearly  to  the  HR  department  and  to  management.  For  many  of  them, 
obtaining  the  high  potential  label  is  perceived  as  a  ‘psychological 
reward’  they  feel  they  deserve  for  their  exceptional  hard  work  and 
achievements.  Overall,  our  findings  imply  that  high  potentials  have 
protean  (i.e.  self‐directed  and  personal  values‐driven)  careers  rather 
than  boundaryless  (i.e.  boundary‐crossing)  careers  (Briscoe  &  Hall, 
2006). We will revisit this latter issue at  a later point in this Discussion.   
 
High potential career outcomes 
 
Performance.  Our  findings  indicate  that  high  potentials  achieve 
significantly higher performance scores, both self‐rated (paper 1, paper 
3, paper 6) and supervisor‐rated (paper 4, paper 6).  
 
As for the self‐rated performance scores, we found that high potentials 
consistently  rate  their  own  performance  higher  than  non‐high 
potentials  do.  Although  one  could  criticize  the  use  of  self‐report 
measures  for  assessing  job  performance,  this  is  still  an  interesting 
finding.  Even  if  self‐enhancement  bias  (i.e.  the  predisposition  that 
individuals  have  to  distort  self‐appraisals  so  as  to  maintain  the  most 
favorable  self‐view,  Campbell  &  Lee,  1988)  did  play  a  role  in 
respondents’  self‐assessments  of  performance,  there  is  no  apparent 
reason why this type of bias would affect high potentials more than non‐
high  potentials.  Possibly,  respondents  based  their  self‐reports  of 
performance in part on the performance evaluation scores received by 
their  superiors,  and  on  the  knowledge  of  being  identified  as  a  high 
potential or not.   
 
Whereas  in  paper  4,  the  supervisor‐rated  scores  of  job  performance 
included in the research model were actually supervisor‐rated, paper 6 
used  respondents’  self‐reports  of  their  most  recently  received 
supervisor‐rated performance rating as a proxy. By including both self‐
rated and (a self‐rated proxy of) supervisor‐rated performance in paper 
6, we were able to draw comparisons. Interestingly, we found that high 
potentials  (but  not  non‐high  potentials)  rated  their  own  performance 
lower  than  their  supervisors  had.  A  possible  explanation  is  that  high 
308
potentials  are,  typically,  ‘overachievers’  who  can  be  quite  hard  on 
themselves  (Dubouloy, 2004). 
 
In  paper  6,  we  examined  possible  Pygmalion  effects  (i.e.  performance 
effects  occurring  when  individuals  or  groups  act  in  accordance  with 
what  other,  significant  parties  expect  of  them)  that  might  occur  when 
high  potentials  (and  non‐high  potentials)  are  openly  informed  of  their 
status  within  the  organization  (Kierein  &  Gold,  2000).  Contrary  to  our 
hypotheses, we did not find any effects of open communication on self‐
rated  or  supervisor‐rated  performance  scores.  However,  we  did  find 
that high potentials (but not non‐high potentials) rate their satisfaction 
with  their  performance  throughout  their  career  (a  subscale  of  the 
subjective  career  success  scale  developed  by  Dries,  Van  Loo  and 
Pepermans,  2009)  higher  under  conditions  of  open  communication.  In 
study  1,  several  high  potentials  admitted  to  being  quite  impatient  and 
driven  by  continuous  challenge.  This  might  explain  why  they  can  look 
back  on  their  careers  and  feel  quite  satisfied  with  what  they  have 
achieved so far, whilst rating their current performance rather low due 
to their inherent drive to ‘always want more’.   
 
As  regards  the  popular  belief  that  high  potentials  who  are  informed  of 
their status become complacent and ‘rest on their laurels’ (i.e. the crown 
prince syndrome, Göbel‐Kobialka, 1998), our findings indicate that quite 
the  opposite  is  true.  Receiving  the  high  potential  label  seems  to 
encourage  employees  to  work  even  harder.  Furthermore,  in  study  1, 
several  of  the  interviewed  high  potentials  reported  feeling  insecure  or 
stressed because they believed that failure from their side would cause 
their  organizations  to  remove  them  from  the  high  potential  list. 
However,  a  certain  degree  of  failure  at  some  point  in  their  careers  is 
likely. High potentials typically move through accelerated development 
tracks in their careers, leaving little space for socialization and induction 
at  each  level.  Similar  results  were  found  in  the  psycho‐analytical  study 
by  Dubouloy  (2004).  Hence,  rather  than  worrying  about  the  ‘crown 
prince syndrome’, organizations should beware of inducing burn‐out in 
their high potentials.  
 
Career commitment. Consistently, we found that the high potentials 
in  our  study  samples  were  very  committed  to  their  careers.  The 
different  papers  used  different  measures  of  career  commitment, 
however. Paper 1, building on interview data, found that high potentials 
act  quite  assertively  in  their  careers,  doing  everything  within  their 
possibilities  to  achieve  their  career  goals.  The  high  potentials  in  this 
sample  reported  to  consciously  initiate  activities  that  enhanced  their 
309
visibility  to  top  management.  In  general,  they  were  quite  frank  in 
communicating  their  career  aspirations  to  their  line  managers  or 
coaches.  Furthermore,  they  enrolled  in  high‐status  MBA  programs, 
volunteered  for  special  projects  and  task  forces,  and  engaged  in 
‘political’ (i.e. goal‐oriented) networking.  
 
Similar results were found in the survey studies we conducted. Paper 3 
only found significant differences between high potentials and non‐high 
potentials  for  one  item,  i.e.  “If  I  could  get  another  job  and  be  paid  the 
same  amount,  I  would  probably  take  it”,  which  high  potentials  rated 
lower.  However,  the  career  commitment  scale  that  was  used  in  this 
study  (that  of  Blau,  1989)  had  a  more  vocational  focus  (i.e.  a  focus  on 
occupational commitment). We used a different scale in paper 6, that is, 
the  career  commitment  scale  developed  by  Day  and  Allen  (2004).  This 
scale  measures  commitment  to  the  career  apart  from  organizational 
setting  or  occupational  choice.  We  found  that  high  potentials  scored 
significantly  higher  on  this  scale  than  non‐high  potentials,  both  under 
conditions of closed and open communication.  
 
Organizational  commitment.  None  of  the  high  potentials 
interviewed  in  study  1  reported  having  any  intention  of  leaving  their 
organization in the near future. Furthermore, only a few of them had a 
plan  B  ready  for  when  their  career  goals  could  not  be  achieved  within 
the  current  organization.  Study  6  replicated  these  findings  in  a  survey 
design;  high  potentials  were  found  to  score  significantly  higher  than 
non‐high  potentials  on  an  organizational  commitment  scale 
(Rottinghaus, Day & Borgen, 2005), both under conditions of closed and 
open  communication.  This  scale  operationalized  organizational 
commitment  as  being  proud  and  happy  to  work  for  the  current 
organization, and not wanting to work elsewhere. 
 
However,  the  findings  reported  in  paper  5  imply  that  we  must  apply 
nuance  in  interpreting  the  above  findings.  This  study,  examining 
psychological  contract  factors  in  high  potential  careers,  concluded  that 
although  high  potentials  do  expect  to  receive  job  security  from  their 
organizations (to a greater extent than non‐high potentials), they do not 
feel  obliged  to  reciprocate  by  offering  organizational  loyalty  in  return. 
Especially when they scored highly on self‐rated employability, they did 
not  perceive  themselves  as  obligated  to  stay  with  their  current 
employer.  On  the  other  hand,  we  found  that  the  high  potentials  in  this 
sample actively pursued external career opportunities to a lesser degree 
than non‐high potentials.  
 
310
Bringing  the  above  findings  together,  we  might  conclude  that  high 
potentials  are  quite  committed  to  their  organizations,  but  not  out  of 
loyalty motives per se. Rather, their intentions to stay with their current 
employer  appear  motivated  by  ‘career  promises’  made  by  their 
organizations.  The  high  potentials  in  study  1  admitted  that,  realizing 
their  high  potential  status  was  organization‐specific,  they  expected  to 
achieve  more  successful  careers  by  staying  with  their  current 
organization. However, as paper 5 implies, this does not mean that they 
will not leave when other, external, career opportunities come along.   
 
Intra­organizational  mobility.  Rapid  and  upward  intra‐
organizational mobility emerged from the study data as a key feature of 
high  potential  careers.  The  high  potentials  that  were  interviewed  for 
paper  1  stated  that  they  expected  to  receive  regular  upward 
promotions.  Interestingly,  their  promotion‐oriented  career  aspirations 
did  not  appear  to  originate  from  some  sort  of  long‐term  career  plan. 
Rather,  upward  promotions  seemed  to  be  considered  as  a  form  of 
reward.  
 
The  interviewed  organizational  representatives,  from  their  side,  stated 
that they still offer plenty of internal career opportunities to their best 
people,  especially  to  those  identified  as  high  potentials.  Establishing  a 
leadership  pipeline  was  one  of  the  principal  goals  within  their  talent 
management  programs.  The  organizational  representatives  generally 
believed    that  internal  successors  would  perform  better  than  external 
hires,  because  the  former  exhibit  fewer  adaptation  problems  and  fit  in 
with  management  easier.  The  quantitative  data  obtained  in  study  6, 
indeed,  demonstrated  that  the  high  potentials  in  the  sample  received 
more intra‐organizational promotions, and at a faster rate, than the non‐
high potentials.  
 
Inter­organizational  mobility.  The  inter‐organizational  mobility  of 
the high potentials in our study samples was rather low, both in terms of 
actual  mobility  patterns  and  in  terms  of  intention  to  pursue  outside 
opportunities.  In  study  1,  10  out  of  the  14  high  potentials  interviewed 
indicated that they had only worked for their current organization over 
the course of their careers. The other four high potentials had changed 
organizations  several  times  at  the  beginning  of  their  careers,  but  now 
felt they had ‘stabilized’. Study 4 obtained similar results. For an average 
work experience of over 12 years, both the high potentials and the non‐
high  potentials  in  the  sample  were  found  to  have  had  very  low 
‘institutional  variety’.  On  average,  respondents  had  worked  for  slightly 
under  two  organizations  during  their  entire  career.  No  significant 
311
differences were found between high potentials and non‐high potentials 
in this respect. 
  
Paper  6  examined  whether  high  potentials  (in  comparison  to  non‐high 
potentials) are more, or less, inclined to prefer stable employment with 
one  employer.  Results  indicated  that,  on  average,  the  high  potentials 
preferred  stable  employment  to  a  larger  extent  than  the  non‐high 
potentials  did.  However,  we  did  find  that  the  relative  openness  of 
organizational  communication  about  talent  management  moderated 
this relationship.  
 
Under  conditions  of  closed  communication  (i.e.  when  they  were  not 
informed  of  their  status),  high  potentials  were  more  likely  than  non‐
high  potentials  to  report  a  preference  for  inter‐organizational  career 
moves. Under conditions of open communication, they were less likely to 
prefer  this  type  of  mobility.  These  findings  imply  that  openly 
communicating  about  the  high  potential  label  possibly  encourages  the 
retention of high potentials. Another possible explanation is grounded in 
the  notion  of  person‐environment  fit  (Cable  &  Parsons,  2001). 
Organizations  with  a  closed  communication  culture  may  attract  more 
individualistic  people  who  are  less  concerned  with  organizational 
loyalty.  We  will  further  discuss  the  potential  effects  of  organizational 
culture on talent management variables in the section on organizational 
communication.  
 
Objective  career  success.  As  mentioned  earlier,  our  findings 
indicated  that  the  high  potentials  in  our  samples  obtained  more 
promotions, and at a faster rate, than the non‐high potentials. However, 
number  of  promotions  and  promotion  speed  were  not  the  only 
measures  of  objective  career  success  we  used  in  this  dissertation.  In 
order to achieve a full picture of high potentials’ objective career success 
as compared to that of non‐high potentials, we also assessed net month 
salary and net month salary increase over the course of working for the 
current organization (paper 6).  
 
We found that high potentials scored significantly higher than non‐high 
potentials  on  all  measures  of  objective  career  success,  although  the 
difference in terms of net month salary increase was non‐significant. A 
possible  explanation  is  that  organizations  are  finding  it  difficult  to  link 
pay  to  potential,  as  they  fear  that  employees  and  unions  alike  would 
perceive such an intervention as ‘distributive injustice’ (Bossuyt & Dries, 
2008).  
 
312
Furthermore,  we  found  a  rather  unexpected  interaction  effect  with 
organizational  communication.  High  potentials  received  a  lower  net 
month  salary  under  conditions  of  open  communication  then  under 
conditions  of  closed  communication.  A  possible  explanation  (also 
grounded in the findings of study 1) is that the high potential label acts 
as  a  form  of  ‘psychological  reward’  that  encourages  retention  (see  De 
Gieter,  De  Cooman,  Pepermans  &  Jegers,  2008).  Organizations 
communicating  openly  about  who  is  on  the  high  potential  list,  then, 
might  not  necessarily  need  to  use  financial  rewards  in  order  to  retain 
high potentials.  
 
Overall,  our  findings  seem  to  indicate  that  the  objective  career  success 
outcome of choice in high potential careers is rapid upward promotion, 
rather than pay.  
 
Subjective  career  success.  In  order  to  assess  subjective  career 
success,  we  used  a  self‐developed  multidimensional  measure  (Dries  et 
al.,  2009).  We  chose  to  develop  our  own  measure,  instead  of  using 
existing  measures  of  subjective  career  success,  as  these  are  widely 
criticized  for  being  overly  simplistic  (e.g.  Heslin,  2005).  The  measure 
was  adapted  from  our  model  of  subjective  career  success  presented  in 
paper  8  and  comprised  10  dimensions:  satisfaction  with  performance, 
satisfaction with advancement, satisfaction with personal development, 
satisfaction  with  creative  outlet,  satisfaction  with  degree  of  financial 
security, satisfaction with degree of job security, satisfaction with work‐
life  dynamics,  satisfaction  with  recognition,  satisfaction  with  inter‐
personal cooperation, and satisfaction with contribution.  
 
Study  6  found  that  high  potentials  scored  higher  than  non‐high 
potentials  on  all  subscales  of  subjective  career  success,  although  the 
differences in terms of satisfaction with degree of financial security and 
satisfaction with work‐life dynamics were not significant. Furthermore, 
we  uncovered  two  interaction  effects  with  organizational 
communication.  First,  under  conditions  of  open  communication,  high 
potentials (but not non‐high potentials) reported more satisfaction with 
performance  than  under  conditions  of  closed  communication.  Second, 
under  conditions  of  open  communication,  non‐high  potentials  (but  not 
high  potentials)  reported  less  satisfaction  with  degree  of  job  security 
than  under  conditions  of  closed  communication.  An  important 
conclusion  from  the  above  is  that  openly  communicating  about  who  is 
identified  as a high potential and who is not  has  some implications  for 
subjective career success.  
 
313
Organizational communication about talent management 
 
Only in paper 6, ‘organizational communication’ was explicitly included 
as a study variable. In all other papers except for paper 1, however, this 
variable  was  not  specifically  addressed,  although  it  most  likely  did 
influence the research context. In papers 3 and 5 (where high potentials 
participated  as  respondents),  the  participating  organizations  had 
adopted  ‘half‐open’  communication  policies,  meaning  that  high 
potentials  were  informally  notified  of  their  status,  whereas  non‐high 
potentials  were  not.  However,  it  is  not  unthinkable  that,  even  in  these 
organizations, the non‐high potentials had some idea of what was going 
on – unfortunately, due to the research designs, it was impossible for us 
to check for this (e.g. Fresina, 1987).  
 
Combining  the  findings  reported  in  paper  1  and  paper  6,  we  see  that 
there  are  some  general  differences  between  organizations  with  open, 
and organizations with closed communication policies. A first difference 
relates  to  the  parties  involved  in  high  potential  identification.  In 
organizations  with  closed  communication  policies,  high  potential 
identification is usually carried out  by special executive committees. In 
organizations  with  open  communication  policies,  direct  supervisors, 
career coaches (internal or external) and team members are also asked 
for  their  input.  It  appears  that  high  potential  programs  in  closed‐
communication  environments  are  more  formal,  involving  mostly 
member from upper management. Open‐communication environments, 
on  the  other  hand,  also  involve  lower‐level  employees  in  their  talent 
management procedures.  
 
A  second  difference  lies  in  the  criteria  used  to  identify  high  potentials. 
Organizations  with  open  communication  policies  stressed  the 
importance  of  interpersonal  skills  for  future  leaders,  whereas  closed‐
communication  organizations  emphasized  strategic  long‐term  thinking 
and  innovation.  We  must  thus  nuance  our  earlier  findings  about 
organizations  under‐emphasizing  interpersonal  skills  in  the  earlier 
stages of high potential careers. It appears that this might be an issue in 
organizations with closed communication policies only.  
 
A  third  difference  between  organizations  with  open  communication 
policies versus organizations with closed communication policies is the 
percentage  of  employees  that  is  identified  as  high  potentials.  Where 
closed‐communication  organizations  identify  between  10  and  20  per 
cent  of  their  workforce  as  high  potentials,  this  was  only  around  3  to  6 
per  cent  for  the  open‐communication  organizations.  A  potential 
314
explanation  lies  in  the  fact  that  openly  communicating  about  who  is 
identified as a high potential creates career expectations in employees, 
so that the label must be used sparingly.  
 
Study  6  uncovered  that  adopting  open  communication  policies  about 
talent  management  generated  positive  main  effects  on  self‐rated  job 
performance,  and  on  most  indicators  of    career  success,  for  all 
employees  in  the  sample  –  not  just  for  high  potentials.  Possibly, 
organizational  communication  climate  is  an  indicator  of  overall 
organizational  culture,  impacting  in  turn  on  employee  outcomes  and 
attitudes.  Closed  communication  policies  may  be  indicative  of  a  more 
performance‐oriented  and  competitive  culture.  Open  communication 
policies  may  point  at  a  culture  focused  on  interpersonal  relationships 
and teamwork. Further research needs to be done in order to be able to 
link  organizational  culture  to  talent  management  policies  more 
carefully.  
 
Furthermore, interaction effects were found between being identified as 
a high potential or not and organizational communication openness on a 
number  of  career  outcomes  (see  earlier  in  this  Discussion).  In  general, 
these interaction effects indicated that the high potential label acts as a 
form of psychological reward encouraging retention. These findings are 
reminiscent of symbolic action theory (Pfeffer, 1981), which states that 
the  mere  presence  of  certain  human  resource  management  initiatives 
are sometimes sufficient to influence employee attitudes.  
 
Non‐high potentials, on the other hand, experience less job security and 
a  greater  preference  for  inter‐organizational  mobility  when  they  are 
aware of the fact they are not identified as high potentials. Organizations 
with open communication policies indeed indicated that communicating 
to non‐high potentials is a delicate affair. However, in order to affect the 
turnover intentions of high potentials, an open communication appears 
essential.  By  adopting  closed  or  half‐open  communication  policies, 
organizations  run  the  risk  of  creating  distrust  between  employees  and 
management,  which  may  create  more  resentment  in  the  long  run  than 
having  open communication policies (Cope, 1998).  
 
The postmodern career context 
 
Protean  versus  boundaryless  careers.  Although  the  terms 
‘boundaryless’  and  ‘protean’  career  are  sometimes  used  as 
interchangeable  (i.e.  they  are  both  concrete  operationalizations  of  the 
new,  ‘postmodern’  career)  they  actually  refer  to  related  but  different 
315
constructs.  While  boundaryless  careers  highlight  different  levels  of 
physical  and  psychological  mobility  (across  organizational  and  other 
boundaries),  protean  careers  are  characterized  by  self‐direction, 
continuous learning, and staying true to one’s personal values (Briscoe 
& Hall, 2006). Putting together all of the above findings, it appears that 
high potentials have protean careers, but not boundaryless careers.  
 
According  to  our  study  findings,  high  potentials  still  have  traditional‐
organizational  careers,  characterized  by  stable  employment  in  one 
organization and rapid upward progression. Organizations still offer this 
type of career tracks to high potential employees, as they prefer internal 
successors  for  management  positions.  Furthermore,  they  report  an 
ongoing  need  for  a  stable  core  of  employees  who  are  familiar  with  the 
organization’s  history,  and  are  responsible  for  ensuring  strategic  and 
operational  continuity  (Tulgan,  2001).  However,  the  emergence  of  the 
postmodern  career  context  has  had  some  implications  for  talent 
management.  Organizations  today  apply  stricter  criteria  for  high 
potential  identification,  identifying  a  smaller  proportion  of  their 
workforce as high potentials than they did a decade ago. High potentials, 
from their side, rather than feeling loyal to their organizations, stay with 
their  organizations  as  long  as  they  feel  it  offers  them  the  best  possible 
career opportunities.  
 
In addition, study 7, although not explicitly dealing with the topic of high 
potentials, also provided some insights as to whether employees today 
are  having  more  boundaryless  careers.  It  found  a  relative  decrease  of 
traditional, single‐employer careers over the course of four generations 
in  terms  of  respondents’  actual  career  patterns.  However,  the  same 
respondents’ attitudinal data revealed that all generations greatly value 
job  security,  especially  the  oldest  and  the  youngest  generation.  So 
although  job  security  has  been  decreasing  in  actuality,  we  found  no 
support for the idea that “more and  more people are thinking like free 
agents” (Tulgan, 2001, p. 37).  
 
As  for  high  potentials  having  protean  careers,  the  evidence  was  more 
striking. In study 1, the high potentials reported on the many initiatives 
they  had  taken  throughout  their  careers  (e.g.  starting  up  new 
departments  or  product  lines,  improving  existing  procedures, 
mentoring  more  junior  staff),  often  setting  off  some  form  of  change 
within  their  organizations.  Overall,  our  findings  imply  that  self‐
management is a crucial feature of high potential careers. Several of the 
high  potentials  interviewed  indicated  that  they  were  identified  as  high 
316
potentials  partly  due  to  their  own  efforts  directed  at  making  their 
achievements and aspirations more visible to management.  
 
Furthermore, one of the conclusions that came out of paper 3 was that 
the EQ‐i subscales on which high potentials scored higher than non‐high 
potentials    (i.e.  assertiveness,  independence,  optimism,  and  flexibility) 
could all be linked, in one way or another, to the protean career concept.  
 
The  meaning  of  ‘career  success’.  Finally,  several  of  the  studies 
incorporated  in  this  dissertation  provided  some  input  about  the 
different  meanings  individual  career  actors  (high  potentials  as  well  as 
non‐high  potentials)  attach  to  the  ‘career  success’  construct.  Paper  7 
concluded  that  there  were  no  significant  differences  in  terms  of  how 
career  success  was  evaluated  between  generations,  nor  between  men 
and  women,  nor  between  employees  at  functional  levels,  nor  between 
employees  in  different  career  types.  All  respondents  agreed  that 
subjective career success (i.e. personal career satisfaction) is the single 
most  important  measure  of  career  success,  regardless  of  functional 
level,  (speed  of)  promotions,  and  salary.  However,  salary  was  a  close 
second, especially for the older generations.  
 
Paper  8,  subsequently,  developed  a  two‐dimensional  model  of 
subjective  career  success  based  on  in‐depth  qualitative  methods.  The 
horizontal  axis  of  the  model  was  labeled  ‘affect‐achievement’  (i.e. 
feelings and perceptions about career versus factual accomplishments). 
The  vertical  axis  was  labeled  ‘inter‐personal  –  intra‐personal’,  which 
referred  to  the  source  validating  the  career  (i.e.  the  external  world 
versus the career actor him or herself). In comparing our model to older 
ones,  we  found  that  previous  research  had  focused  mainly  on  inter‐
personal  affect  indicators  of  subjective  career  success.  The  opposite 
quadrant,  i.e.  the  intra‐personal  achievement  quadrant,  was  rarely 
mentioned at all in the older models.  
 
Some  of  the  statements  made  by  the  high  potentials  interviewed  in 
paper  1  can  be  linked  back  to  our  model  of  subjective  career  success. 
Several of the high potentials in this study indicated that they would feel 
less  satisfied  about  their  career  achievements  if  they  had  not  achieved 
them  by  their  own  initiative  (which  relates  to  the  ‘independence’ 
subscale in our career success model). This finding, again, points to high 
potentials    having  a  protean  career  orientation  (Briscoe  &  Hall,  2006). 
Furthermore,  the  interviewed  high  potentials  seemed  to  define  ‘career 
success’  primarily  as  related  to  extrinsic  rewards  (especially  to 
‘recognition’).  Especially  upward  promotion  (‘advancement’)  was 
317
equated  with  success  in  the  high  potentials’  discourse.  Furthermore,  a 
number  of  them  mentioned  pay  as  a  prime  indicator  of  success. 
Although  pay,  as  such,  is  not  a  subscale  in  the  model  of  career  success 
presented  in  paper  8,  it  can  be  linked  to  several  other  career  success 
constructs.  Pay  can  refer  to  ‘financial  security’,  to  ‘recognition’,  or  to 
‘advancement’, depending on the meaning attributed to it in the mind of 
the individual career actor.  
 
An important closing remark is that the high potentials in study 1 stated 
that  they  attached  so  much  importance  to  promotions  and  pay  raises, 
because  they  felt  that  these  are  career  rewards,  reserved  by  their 
organizations  for  the  best  only.  Hence,  for  high  potentials,  career 
success  is  an  outcome  that  is  rare  and  difficult  to  obtain.  Achieving 
‘advancement’  and  ‘recognition’  thus  appears  to  be  the  primary  career 
driver for high potentials.  
 
Limitations 
 
Cross­sectional designs 
 
All of the studies incorporated in this dissertation had a cross‐sectional 
design,  meaning  that  the  data  for  each  subject  for  each  variable  were 
collected at the same point in time. Although the data for study 6 were 
collected  at  two  different  time  points  (separated  by  a  few  weeks),  this 
intervention  was    aimed  mainly  at  reducing  common  method  variance 
(addressed  at  a  later  point  in  this  Discussion),  rather  than  achieving  a 
longitudinal research design. 
 
The major limitation of using cross‐sectional designs is that causality is 
impossible  to  demonstrate  (Podsakoff,  Mackenzie,  Lee  &  Podsakoff, 
2003).  Especially  in  the  survey  studies  where  high  potentials  and  non‐
high  potentials  were  compared  on  a  number  of  self‐reported  variables 
(i.e. study 3, study 5 and study 6), this issue was salient. In study 3, for 
instance,  we  could  not  conclude  with  certainty  whether  the  high 
potentials scored higher in terms of job performance due to the fact that 
those identified as high potentials are typically also high performers, or 
due  to  ‘Pygmalion’  effects  caused  by  respondents’  knowledge  of  their 
high potential status. The same goes for study 6. In study 5, as well, we 
could not prove that the psychological contract dynamics described for 
the  high  potentials  in  the  sample  were  actually  caused  by  their 
awareness  of  their  relative  bargaining  position  in  the  organization, 
although the findings do imply so.  
 
318
In the studies on the meanings of ‘career success’ (study 7 and study 8), 
as  well,  causality  could  not  be  demonstrated.  This  type  of  study  is 
sensitive  to  processes  of  cognitive  dissonance  reduction  or  post‐hoc 
rationalization  (Nicholson  &  de  Waal‐Andrews,  2005).  Put  simply,  it  is 
well possible that respondents (subconsciously) adapted their personal 
definitions  of  career  success  to  their  actual  career  experiences,  thus 
allowing themselves to  feel successful. This was  an issue in study 1,  as 
well. High potentials in this study expressed very traditional ideas as to 
what career success means to them (i.e. rapid upward progression and 
regular  pay  raises).  Due  to  the  cross‐sectional  design,  however,  it  is 
unclear  whether  these  notions  of  career  success  caused  them  to  be 
attracted  to  high  potential  career  tracks,  or  rather,  were  caused  by  the 
traditional  career  opportunity  structures  they  experienced  on  a  daily 
basis, being identified as high potentials.  
 
The  issue  of  obtaining  data  at  the  same  time,  from  the  same  location, 
through  the  same  medium  corresponds  to  one  of  the  four  causes  of 
common  method  variance  (CMV)  identified  by  Podsakoff  et  al.  (2003), 
i.e.  measurement  context  effects.  The  authors  describe  how  contextual 
factors  may  encourage  systematic  covariation  between  the  measures 
incorporated  in  the  study.  By  collecting  data  at  only  one  point  in  time, 
for  instance,  the  likelihood  of  respondents  constructing  an  implicit 
theory  based on the different  variables in the study increases. Another 
example of context effects is the influence of interviewer characteristics 
(which relates to ‘medium’) on interviewee statements.  
 
In  general,  cross‐sectional  studies  should  be  seen  as  representing  a 
‘snapshot’  of  reality.  This  means  that  the  timeframe  within  which  the 
studies  were  conducted  is  one  of  the  contextual  factors  potentially 
influencing  the  results.  An  important  remark  in  this  respect  is  that  the 
eight studies incorporated in this dissertation were not conducted in the 
order of presentation. Rather, we ordered them according to topicality. 
One  particular  time‐bound  development  within  our  research  context 
was  the  emergence  of  the  global  financial  crisis,  which  came  to  the 
forefront  of  the  worldwide  media  around  September  2008.  All  studies 
except study 5 had been completed by this time. As is discussed in paper 
5, the heavy mediatization of the crisis may have caused respondents to 
attach more value to job security and relational psychological contracts 
at the time of data collection.  
 
It  thus  appears  that,  in  order  to  advance  the  research  on  talent 
management and the meanings of  ‘career’, longitudinal designs are the 
way forward. Longitudinal studies could provide useful insights into the 
319
causality  of  the  relationships  hypothesized  in  this  dissertation.  On  the 
other  hand,  however,  researchers  must  beware  of  seeing  longitudinal 
research  as  the  panacea  to  all  research  limitations.  Even  with 
longitudinal  designs,  causality  is  difficult  to  demonstrate  as  the 
variables  investigated  in  the  study  will  not  be  the  only  variables 
changing over time (Taris & Kompier, 2003). Furthermore, considering 
the  methodological  difficulties  paired  with  studying  hard  to  reach 
populations  such  as  employees  (sometimes  unknowingly)  identified  as 
high  potentials,  the  adopted  research  designs  would  have  to  be  quite 
intricate, which means that research processes will be even more time‐ 
and  resource  consuming  than  those  of  the  cross‐sectional  designs 
described in this dissertation.  
 
Sampling strategies 
 
Due  to  the  fact  that  it  is  impossible  to  delineate  ‘the  high  potential 
population’, we had no other choice than to work with non‐probability 
samples.  The  general  consensus  in  the  literature  is  that  representative 
sampling  carries  more  reliability  and  validity.  Representative  sampling 
means that each member of the population under scrutiny has an equal 
chance of being selected into the research sample (Trochim & Donnelly, 
2007).  When  the  population  is  not  ‘known’  however,  obtaining  a 
representative  sample  is,  evidently,  unfeasible  (Austin,  Scherbaum  & 
Mahlman, 2002). Being identified as a high potential is not an outwardly 
manifested  variable  that  can  be  traced  throughout  the  working 
population.  Therefore,  composing  a  socio‐demographic  (or  other) 
profile  of  ‘the  average  (Belgian)  high  potential’  is  not  possible.  As  a 
result,  there  is  no  way  of  inferring  to  which  extent  the  samples  we 
obtained    were  representative  of  the  population  under  study.  One 
initiative that we did take was to ask each HR contact person composing 
the  study  sample  for  their  organization  to  aim  for  a  sample 
representative of the high potential population within the organization. 
However,  we  were  not  able  to  conduct  non‐response  analyses  as  the 
participating organizations were very strict in terms of the data we were 
allowed access to.   
 
Despite the limitations of non‐probability sampling, the methodological 
literature has suggested that, in the case of hard to reach populations, a 
series  of  studies  containing  non‐probability  samples  can  lead  to 
accumulation  of  knowledge.  When  similar  results  are  obtained 
repeatedly  with  many  different  non‐probability  samples  (as  was  the 
case  in  this  dissertation),  the  likelihood  that  those  results  apply  to  the 
population is greater than when only a single non‐probability sample is 
320
used (Fowler, 2002). Nevertheless, nuance must  be  applied in drawing 
generalizations  from  non‐probabilistic  data  sets  because  of  the 
possibility  of  hidden  systematic  bias  (Podsakoff  et  al.,  2003).  The 
following  questions  must  be  addressed  before  interpreting  findings 
based on non‐probability samples (Fowler, 2002):  
 
(1) What types of people were systematically excluded from the sample?  
In  our  case,  there  was  probably  an  under‐representation  of  high  potentials 
from  organizations  with  closed  talent  management  communication  policies, 
and  from  smaller  organizations  with  fewer  hierarchical  levels.  Over  the 
course  of  the  research  process,  we  were  often  contacted  by  organizations 
who were interested in our research; however, only a small portion of these 
organizations actually participated in one of the studies. The main drawback 
for  organizations  was  that  we  would  survey  their  high  potential  population 
and  compare  their  data  to  that  of  non‐high  potential  peers.  Organizations 
with  closed  communication  policies  typically  worried  that  participating  in 
the  research  would  create  some  sort  of  confidentiality  breach,  and  expose 
hidden variables from the organization’s talent management procedures. For 
other organizations, it was not possible to participate because they were not 
(yet) able to explicitly compose separate lists of high potential and non‐high 
potential respondents. 
   
(2) What types of people were over­represented in the sample? 
As follows logically from (1), the types of people that were over‐represented 
in  the  research  were  white‐collar,  highly  educated  employees  working  for 
large  (often  multinational)  organizations  with  mature  and  open  talent 
management  programs.  Furthermore,  in  study  8,  which  was  not  specifically 
aimed  at  high  potentials,  all  interviewees  were  managers.  This  means  that 
the  model  of  subjective  career  success  that  was  developed  might  not  be 
generalizable to other employee groups, such as blue‐collar workers.  
 
(3) Have the findings been replicated by different researchers using a variety of 
data­collection methods with different samples?  
As  we  discussed  elaborately  in  the  Introduction  as  well  as  in  the  different 
papers  included  in  this  dissertation,  there  have  hardly  been  any  empirical 
studies  on  the  topic  of  talent  management.  However,  when  comparing  the 
findings  of  our  studies  to  those  of  the  few  studies  there  have  been  (e.g. 
Dubouloy, 2004; Lombardo & Eichinger, 2000; Spreitzer, McCall & Mahoney, 
1997;  Viney,  Adamson  &  Doherty,  1997),  many  similarities  are  identified. 
Also  across  the  different  studies  included  in  this  dissertation  (which 
demonstrated  a  variety  of  designs  and  samples),  several  findings  were 
replicated.  
 
Another issue with our study samples was that they were usually quite 
small,  due  to  the  fact  that  obtaining  participation  from  organizations 
proved  a  very  difficult  process.  Furthermore,  in  each  study,  we  used 
data  from  high  potentials  (and  non‐high  potentials)  coming  from 
different  organizations.  By  doing  so  extra  heterogeneities  were 
introduced in the study samples, which have to be taken into account in 
generalizing  the  study  findings  (Trochim  &  Donnelly,  2007).  Study  6 
321
explicitly  addressed  differences  in  structural  and  talent  management 
program  characteristics  between  participating  organizations  in  a  table, 
whereas the other studies offer more general background variables per 
organization.  We  did  make  sure,  however,  that  all  organizations 
participating in the research drew on the exact same definitions of high 
potentials and non‐high potentials in composing the study samples.  
 
As  in  all  single‐standing  empirical  studies,  contextual  variables  have 
undoubtedly  impinged  on  some  of  our  findings.  Organizational  size, 
nationality, and culture, as well as broader cultural or language‐related 
variables  are  the  most  prevalent  contextual  factors  impacting  on  work 
and  organizational  psychology  research  (Fowler,  2002).  Steering  clear 
of this limitation, however, is a daunting challenge for every researcher. 
Obtaining  a  data  set  so  elaborate  that  all  of  these  contextual  factors 
could be controlled for would ask for a cross‐cultural research initiative 
where  data  is  collected  by  various  research  groups  around  the  world 
(e.g. Chudzikowski et al., 2009).  
 
 
Operationalizations 
 
As  was  elaborately  discussed  in  the  Introduction,  several  of  the  key 
concepts  in  this  dissertation  lack  conceptual  clarity  in  the  literature. 
‘Talent  management’  (Collings  &  Mellahi,  2009;  Lewis  &  Heckman, 
2006),  ‘high  potentials’  (Boudreau  &  Ramstad,  2005;  Garrow  &  Hirsch, 
2008),  ‘career’  (Adamson,  Doherty  &  Viney,  1998;  Evetts,  1992)  and 
‘career success’ (Poole, Langan‐Fox & Omodei, 1993; Heslin, 2005) have 
all been criticized for being ill‐defined in the literature. The definition of 
the latter two concepts was explicitly examined in this dissertation; this 
was  not  so  for  ‘talent  management’  and  ‘high  potentials’.  As  a  result, 
some conceptual tensions emerged within this dissertation, as well. 
 
Talent management. Although we mention in several papers that we 
consider talent  management to  be  broader than just encompassing the 
career  management  of  employees  identified  as  high  potentials,  our 
research  is,  in  fact,  focused  on  high  potentials  alone.  We  believe, 
however, that in order to be able to conduct quantitative studies on the 
topic  of  talent  management,  researchers  must  focus  on  specific  talent 
segments,  be  it  high  potentials,  key  experts,  key  project  workers,  or 
other. Studies that  aim to explore talent management in all its  breadth 
would  be  better  served  with  qualitative  methods,  such  as  individual 
interviews,  focus  groups  or  stakeholder  analyses  (e.g.  Bossuyt  &  Dries, 
2008).  
322
 
High potentials. Another  limitation affecting some of  our studies is 
that  there  is  no  external  criterion  to  evaluate  whether  employees 
identified as high potentials within their organizations actually are high 
potentials.  In  order  to  obtain  respondents  for  our  studies,  we  asked 
organizations to (directly or indirectly) put us into contact with “those 
individuals within the organization who are recognized, at that point in 
time,  as  the  organization’s  likely  future  leaders”  (Cope,  1998,  p.  15), 
which  was  our  working  definition  of  high  potentials.  Although  we 
always  performed  preliminary  interviews  to  ensure  that  the 
conceptualization  of  ‘high  potentials’  was  homogenous  among  the 
organizations  participating  in  our  studies,  we  are  not  in  a  position  to 
assess whether they identified the ‘right’ people as high potentials. This 
means  that  it  is  very  well  possible  that  the  non‐high  potential  samples 
contained  some  employees  who  did  in  fact  have  leadership  potential, 
whereas that the high potential samples may have contained employees 
with little potential. Nevertheless, and as we stated in the Introduction, 
‘high  potential’  is  a  label  more  than  it  is  an  attribute  (Vinkenburg  & 
Pepermans, 2005). As the high potential label represents the outcome of 
an  organization‐specific  assessment  (Boudreau  &  Ramstad,  2005),  it 
makes  sense  to  go  about  composing  samples  of  high  potentials  by 
drawing on lists of high potentials composed by organizations.  
 
We  are  aware  of  one  (unpublished)  study  that  did  apply  an  external 
criterion for composing its high potential sample (Abrams et al., 2008). 
In  this  study,  a  large  sample  of  random  participants  were  asked  to  fill 
out  a  survey  that  contained  scales  measuring  creativity,  leadership 
skills,  learning  agility,  and  autonomy.  As  these  are,  according  to  the 
authors,  typical  characteristics  of  high  potentials,  then  high  potentials 
are  those  respondents  scoring  high  on  these  scales.  This  is  a  different 
way  of  operationalizing  the  ‘high  potential’  variable  –  and  one  with 
different limitations. The main limitation of this approach is that it relies 
on  self‐ratings  of  respondents  to  determine  whether  they  are  high 
potentials  or  not.  However,  in  order  for  the  ‘high  potential’  variable  to 
explicitly  shape  an  individual’s  career,  other  parties  involved  in  this 
person’s  career  (i.e.  supervisors,  co‐workers,  clients)  would  have  to 
agree that he or she is in fact a high potential (Altman, 1997).  
 
Measures 
 
With  the  exception  of  the  high  potential  identification  criteria  model 
administered  as  a  scale  in  study  2,  the  subjective  career  success  scale 
(based on study 8) included in study 6, and the career success vignettes 
323
in study 7, all measures that were used in this dissertation were existing 
measures  that  had  been  validated  in  previous  research.  We  developed 
these new measures as no other measures were available (in the case of 
the  identification  criteria  and  the  career  success  vignettes),  or  because 
existing  measures  were  heavily  criticized  (in  the  case  of  the  subjective 
career  success  scale,  see  Heslin,  2005).  Obviously,  further  research  is 
necessary  in  order  to  (cross)  validate  these  new  measures  (Trochim  & 
Donnelly, 2007).  
 
Several  of  the  existing  measures  that  we  used,  although  extensively 
validated, have not been without critique in the literature. For instance, 
the EQ‐i (Bar‐On, 1997) has been criticized for adopting a trait approach 
to  emotional  intelligence  (e.g.  McEnrue  &  Groves,  2006);  and  the 
Psychological  Contract  Inventory  (PCI)  (Rousseau,  2000)  has  been 
condemned  for  its  scalar  instability  (e.g.  Conway  &  Briner,  2005).  We 
will  not  go  into  each  of  these  critiques  again,  as  they  were  already 
addressed  in  the  separate  papers.  Furthermore  (and  regrettably  so), 
there  are  hardly  any  measures  in  the  work  and  organizational 
psychology  literature  that  are  without  critique  (Furnham,  2004).  What 
we wish to focus on in this section is our use of self‐report measures.  
 
Self‐report measures are particularly prone to CMV, i.e. “variance that is 
attributable  to  the  measurement  method  rather  than  to  the  constructs 
the  measures  represent”  (Podsakoff  et  al.,  2003,  p.  879),  as  the 
respondent  rating  the  independent  and  the  dependent  variables  is  the 
same  person.  This  may  cause  both  common  rater  effects  and  item 
context effects (Howitt & Cramer, 2005). 
 
Common  rater  effects  might  occur  when  respondents  (subconsciously) 
strive  for  consistency  in  their  responses,  or  when  they  adapt  their 
responses to match socially desirable attitudes (Podsakoff et al., 2003). 
In most of our studies, we used both attitudinal and (proxies of) factual 
variables (i.e. being identified as a high potential or not, organizational 
communication policies about talent management, supervisor‐rated job 
performance,  organizational  tenure,  salary  variables,  promotions), 
which decreases the odds of common rater effects (Schwarz, Rizzuto & 
Schwarz, 2008). Social desirability bias (Krueger, 1998) from the side of 
the  individual  employees  may  have  played  a  role  especially  in  study  1 
and  in  study  8,  in  which  non‐anonymous  respondents  were  asked  to 
provide  their  personal  career  experiences  in  face‐to‐face  interviews. 
However,  in  these  studies,  we  highlighted  our  interest  in  idiosyncratic 
career  experiences,  and  instructed  respondents  explicitly  to  respond 
according  to  their  own  opinions.  Possibly,  social  desirability  bias 
324
impacted  more  on  the  data  collected  from  organizational 
representatives.  In  study  1,  we  explicitly  collected  data  from 
organizations; in each of the other studies, we collected (as a minimum) 
some  background  variables  about  organization’s  career  structures  and 
talent  management  programs.  We  noticed  during  the  prescreening 
interviews  prior  to  each  study  that  organizations  only  sparingly 
provided  information  about  their  talent  management  procedures.  In 
general, they were hesitant to admit that certain constraints (typically in 
terms  of  time,  structure,  resources  and  management  involvement) 
hindered an optimal implementation of talent management. It was only 
through  asking  unambiguous  questions  and  imposing  strict  criteria  for 
research  participation  that  we  succeeded  in  obtaining  correct 
information about the organization’s talent management policies.  
 
Item context effects, on the other hand, arise when items are presented 
per  scale,  in  thematic  ‘blocks’.  When  items  are  presented  like  this,  it 
encourages respondents to answer consistently across items rather than 
score  each  item  separately  (Schwarz  et  al.,  2008).  We  countered  this 
type of CMV effects by randomizing the order of the items in each of our 
studies.  Furthermore,  most  of  our  focal  variables  were  measured  by 
rather long, multidimensional scales. Hence, ‘consistency motif’ was less 
of a concern in our studies (Podsakoff et al., 2003). 
 
Another  type  of  bias  that  might  have  influenced  our  findings  was 
selective  recollection  bias  (Robinson  &  Taylor,  1998).  Especially  in 
paper 1 and paper 8, where respondents were asked to tell the stories of 
their  careers,  selective  recollection  may  have  played  a  role.  However, 
according to constructivist career theory, narratives should be regarded 
merely  as  entrance  points  into  interviewees’  internal  worlds,  rather 
than as factual accounts of what really happened (e.g. Bourne & Jenkins, 
2005). A related question is whether respondents are able to report on 
the  number  of  promotions  they  have  obtained  throughout  their  entire 
career, or recall correctly the first net month salary they received from 
their current employer. Although it would have been better to combine 
archival data with attitudinal data in some of the studies (e.g. study 6), 
this would have further hampered our data collection processes.  
 
Although  researchers  generally  agree  that  CMV,  potentially,  affects  the 
results  of  any  mono‐method  study,  there  is  no  consensus  about  the 
relative  impact  CMV  might  have  on  research  outcomes  (Crampton  & 
Wagner,  1994).  Some  authors  state  that  the  effects  of  CMV  are 
overestimated.  Spector  (2006),  for  instance,  asserts  that  using  self‐
reports  does  not  at  all  guarantee  finding  significant  results,  and  that 
325
mono‐method  correlations  are  not  necessarily  higher  than  multi‐
method correlations. He concludes that “the popular position suggesting 
common  method  variance  automatically  affects  variables  measured 
with the same method is a distortion and oversimplification of the true 
state of affairs” (p. 221).  
 
Another  issue  altogether  was  a  lack  of  outcome  measures  appropriate 
for  assessing  the  effectiveness  of  different  talent  management 
interventions.  For  instance,  although  we  were  able  to  come  to  a 
multidimensional  model  of  the  criteria  considered,  by  talent 
management  experts,  as  essential  for  identifying  high  potentials,  we 
could not assess whether employees possessing these characteristics, in 
fact,  develop  into  effective  leaders  several  years  after  identification. 
Furthermore,  we  could  also  not  get  access  to  actual  turnover  numbers 
for employees identified as high potentials. However, as became clear in 
study  1,  organizations  themselves  struggle  with  measuring  talent 
management outcomes, as they find it difficult to define and consolidate 
indicators  of  return  on  investment  (ROI)  (e.g.    Boudreau  &  Ramstad, 
2006).  
 
Debriefing 
 
In the studies where we compared data coming from high potentials to 
data  coming  from  non‐high  potentials  (i.e.  studies  3,  5  and  6),  full 
debriefing of the respondents was not possible. Not disclosing the ‘talent 
category’ variable at any point was a demand made by the organizations 
participating in the studies. It remains unclear to  what extent  the non‐
disclosure of important career variables by organizations to employees 
is unethical (Van Buren, 2003).  
 
As for our responsibility as researchers, we found some input in the APA 
Ethical Principles of Psychologists and Code of Conduct (2002): 
 
(1) Psychologists should always debrief study participants about the nature, 
results,  and  conclusions  of  the  study,  unless  there  are  scientific  reasons  for 
delaying  or  withholding  certain  information.  In  such  case,  they  must  take 
reasonable measures to reduce the risk of harm (Article 8.08);  
 
(2) Psychologists may conduct research without fully informed consent only 
when  the  research  design  involves  anonymous  questionnaires,  or  other 
methods  for  which  disclosure  of  the  results  does  not  harm  participants’ 
anonymity, reputation, or employability (Article 8.05); 
 
(3)  Psychologists  should  not  use  deceptive  information,  or  omit  relevant 
information,  in  their  study  designs  unless  alternative,  ‘non‐deceptive’ 
procedures are unavailable (Article 8.07). 
326
 
Although we would have personally  preferred to fully debrief all study 
participants,  this  would  have  resulted  in  exclusion  of  high  potentials 
working  for  organizations  with  closed  communication  policies,  which 
would have introduced systematic bias in our findings (Fowler, 2002).   
 
Directions for future research 
 
“The more I read, the more I meditate; and the more I acquire, the more I am enabled to 
affirm that I know nothing” (Voltaire). 
 
Talent management 
 
High  potential  identification.  As  for  high  potential  identification, 
several  directions  for  future  research  can  be  laid  out.  First  of  all,  it 
would  be  interesting  to  further  examine  to  what  extent  ‘generic’  or 
‘universal’  criteria  of  leadership  potential  are  workable,  and  to  what 
extent leadership potential identification criteria need to be customized 
to  certain  elements  of  organizational  strategy  (Boudreau  &  Ramstad, 
2006).  An  interesting  paper,  guiding  the  way  forward,  is  that  of 
Hollenbeck,  McCall  and  Silzer  (2006).  The  authors  discuss  elaborately 
how  any  model  of  leadership  should  include  not  only 
(meta)competencies, but also situations (e.g. job content, interpersonal 
dynamics,  team  influences,  organizational  cultures,  national  cultures) 
and outcomes (i.e. “what do we expect of an effective leader?”), and the 
interactions between them. Qualitative designs (e.g. focus groups) seem 
particularly well suited for research endeavors aimed at exploring these 
issues.  
 
Second, in order to find out which criteria organizations actually use to 
identify high potentials and by which processes, we need to collect more 
factual data (e.g. evaluation reports, assessment center data) or engage 
in  participant  observations  of  some  sort.  In  study  2,  we  attempted  to 
find out which criteria organizations use to identify high potentials, and 
detect  possible  conflicts  of  interest  between  the  different  parties 
involved.  However,  we  found  that  all  77  criteria  included  in  the  model 
were  rated  as  ‘essential’,  and  that  there  were  no  conflicts  of  interest 
between HR managers, line managers and top managers concerning the 
identification of high potentials. These findings seemed to be affected by 
social  desirability  bias  (Krueger,  1998),  especially  since  the 
organizational  representatives  in  study  1  did  acknowledge  conflicts  of 
interest between the different parties.  
 
327
Third,  apart  from  describing  the  criteria  used  for  identifying  high 
potentials, it would also be interesting to study their predictive value as 
concerns  the  effectiveness  of  employees’  leadership  behavior  x  years 
after their identification as high potentials. Longitudinal studies linking 
assessment center data to performance criteria later on in participants’ 
careers may provide helpful input for setting up such research designs 
(e.g. Jansen & Vinkenburg, 2006). It would also be interesting to look at 
factors  leading  to  career  derailment,  that  is,  those  factors  that  lead  to 
removal  from  the  high  potential  list,  or  ineffective  leadership  behavior 
(Van  Velsor  &  Leslie,  1995).  It  would  be  interesting  to  replicate  the 
findings  from  paper  3  –  about  the  relative  importance  of  performance 
versus  interpersonal  skills  at  different  functional  levels  –  within  a 
longitudinal design.  
 
Organizational culture. Another interesting idea for future research 
is to link the literature on organizational culture to talent management. 
Elaborating  further  on  our  findings  pertaining  to  organizational 
communication,  this  might  help  us  identify  the  different  contextual 
factors  impacting  on  the  way  talent  management  is  defined  and 
implemented  in  organizations.  Inspiration  might  be  found  in  the 
organizational  culture  typology  developed  by  Cameron  and  Quinn 
(1999).  
 
It might also be possible to link organizational culture to specific career 
success values. It has been suggested that organizations, by encouraging 
or  rewarding  certain  types  of  career  mobility,  influence  the  career 
success  definitions  of  their  members  (Evetts,  1992;  Duxbury,  Dyke  & 
Lam,  1999),  and  attract  employees  with  similar  values  (e.g.  Cable  & 
Parsons, 2001). 
 
The postmodern career 
 
Career  patterns.  The  postmodern  careers  literature  has  been 
criticized  for  developing  (new)  dogmas,  rather  than  building  on 
empirical  data  to  describe  contemporary  careers  (e.g.  Dyer  & 
Humphries,  2002).  In  order  to  be  able  to  draw  inferences  about  the 
career  patterns  of  employees  today,  large‐scale  longitudinal  survey 
studies must be conducted. One interesting suggestion is to administer 
annual follow‐up surveys from high school, college or university alumni 
(e.g. Abele & Spurk, 2009). Such designs could help researchers separate 
cause  from  effect.  Furthermore,  they  might  provide  insights  into  the 
dynamics  of  postmodern  careers:  are  career  actors,  indeed, 
experiencing more boundaryless and/or protean careers? And if so, did 
328
they  choose  to  pursue  this  type  of  career  or  did  circumstances  require 
them  to?  Longitudinal  designs  could  also  clarify  the  two‐way  causal 
relationship  between  career  success  definition  and  actual  career  type 
(Nicholson & de Waal‐Andrews, 2005).  
 
Career  success.  In  order  to  develop  a  nomological  network  for  our 
model  of  subjective  career  success  (presented  in  paper  8),  we  need  to 
examine  its  interrelations  with  existing  models  of  career  success 
(Cronbach  &  Meehl,  1955).  First  of  all,  we  know  that  measures  of 
objective  and  subjective  career  success  should  correlate,  but  only 
moderately  so  (e.g.  Heslin,  2005).  Second,  it  would  be  interesting  to 
scrutinize  the  interrelations  of  subjective  career  success  with  other 
career  concepts,  such  as  career  orientation  (e.g.  Derr,  1986),  career 
anchors (Schein, 1978), job satisfaction (Hackman & Oldham, 1980) and 
life  satisfaction  (Diener,  Emmons,  Larsen  &  Griffin,  1985).  Obviously, 
surveys  combining  these  different  variables  would  have  to  be  split  up 
and  administered  at  multiple  time  points  to  minimize  the  potential 
effects  of  CMV,  whilst  controlling  for  mood  and  affect  variables 
(Podsakoff et al., 2003).  
 
Furthermore,  it  would  be  interesting  to  compile  an  overview  of  the 
different  measures  available  for  assessing  career  success,  matching 
these to research questions, designs, resources and constraints. We are 
currently  working  on  a  ‘decision  tree’  type  model  aimed  at  helping 
careers  researchers  determine  which  measure  of  career  success  fits 
their  research  best.  Possible  decisions  to  be  made  are:  measuring 
objective  and/or  subjective  career  success;  using  factual  (e.g.  archival) 
data  or  self‐report  data;  using  long,  multidimensional  measures  or 
short,  unidimensional  measures;  using  self‐referent  or  other‐referent 
measures; and using absolute or relative measures.  
 
Finally,  there  is  a  significant  need  and  opportunity  for  careers 
researchers  to  explore  the  career  success  definitions  of  less  privileged 
workforce  segments,  such  as  blue‐collar  workers,  low‐skilled 
immigrants, or people with disabilities (e.g. Lucas & Buzzanell, 2004). In 
general,  more  research  needs  to  be  done  examining  the  effects  of 
contextual factors (e.g. cultural, life‐cycle, and generational differences) 
on  individual  career  actors’  definitions  of  career  success  (Adamson  et 
al., 1998). 
 
 
 
 
329
Implications for practice 
 
A  first  implication  for  practice  is  that  organizations  engaging  in  talent 
management  need  to  explicitly  define  what  talent  management  means 
to  them  and  which  employee  groups  they  consider  to  be  ‘talent’ 
(Collings  &  Mellahi,  2009;  Garrow  &  Hirsch,  2008;  Lewis  &  Heckman, 
2006). Failing to do so reduces talent management to a hype rather than 
a strategic tool (Creelman, 2004). Furthermore, organizations must take 
into  account  a  number  of  strategic  considerations:  which  employee 
groups  they  consider  as  contributing  most  to  the  mission  of  the 
organization;  which  criteria  they  will  use  for  the  identification  of  high 
potentials  (and/or  key  experts,  and/or  key  project  workers);  which 
career opportunities they are able and willing to offer to each employee 
group;  and  whether  to  adopt  open  or  a  closed  communication  policies 
about  talent  management,  taking  into  account  that  these  policies  must 
accord  with  general  organizational  culture.  It  logically  follows  that 
strategic  talent  management  policies  in  large  organizations  with 
established  career  structures  will  look  different  from  those  in  smaller 
organizations with fewer hierarchical levels (Baruch & Peiperl, 1997). A 
possible  benefit  for  smaller,  leaner  organizations  is  that  their  career 
structures  may  be  more  accommodating  to  extremely  self‐directed, 
protean employees (Briscoe & Hall, 2006). 
 
A second implication that emerged quite clearly from this dissertation is 
that  organizations  must  engage  in  expectations  management  if  they 
want  to  retain  their  high  potential  staff  (Granrose  &  Baccili,  2006). 
Implementing talent management into an organization, and identifying a 
proportion  of  the  workforce  as  high  potentials,  creates  certain 
expectations in employees (Cope, 1998). Our findings indicated that the 
high  potential  label  creates  expectations  in  terms  of  regular  upward 
career  moves,  regular  salary  increases,  receiving  job  security,  being 
assigned  a  broad  job  content,  and  receiving  plenty  of  organizational 
career  support.  Since  we  also  found  that  these  perceived  promises, 
rather  than  organizational  loyalty,  are  at  the  heart  of  high  potentials’ 
intentions  to  stay  with  their  current  employer,  it  becomes  clear  that 
frustrating  the  above  expectations  would  be  detrimental  for  retention 
(Baruch  &  Peiperl,  1997;  Granrose  &  Baccili,  2006).  A  well‐organized 
succession  planning  (Garrow  &  Hirsch,  2008)  and  a  lucid 
communication  strategy  (Eisenberg,  1984)  are  two  crucial  elements  in 
establishing  a  realistic  expectations  framework  by  organizations 
towards  their  employees.  In  many  organizations,  there  is  no  adequate 
follow‐up  or  consolidation  of  the  data  obtained  from  their  high 
potentials  at  different  times  and  through  various  channels  (e.g. 
330
performance  appraisal  interviews,  employee  satisfaction  surveys, 
personal  development  plans,  conversations  with  career  coaches  or 
mentors),  so  that  by  the  time  organizations  become  aware  of  a  high 
potential’s dissatisfaction, it is often too late (Tulgan, 2001). By setting 
up  a  database  aimed  specifically  at  collecting  all  possible  data  about 
high  potentials  in  the  organization  (and  other  groups  in  the 
organization,  for  that  matter),  better  follow‐up  becomes  possible 
(Boudreau  &  Ramstad,  2006).  Furthermore,  as  high  potentials  tend  to 
know  exactly  what  they  want,  regular  meetings  with  a  career  coach  or 
mentor  seem  appropriate  in  order  to  guard  the  match  between 
individual  and  organizational  goals.  Career  management  approaches 
aimed at developing programs for large employee groups, whilst leaving 
space  for  individual  customization  have  been  referred  to  in  the  recent 
literature as ‘mass career customization’ (Benko & Weisberg, 2007).   
 
A third implication is that organizations must beware of causing ‘career 
reification’,  in  other  words,  of  imposing  certain  career  success 
frameworks  on  their  employees  by  rewarding  or  encouraging  certain 
types of career behavior (Evetts, 1992). Career reification brings with it 
several undesirable side‐effects. First of all, our data demonstrated that 
too great a focus on rapid upward advancement in early high potential 
careers might cause an over‐emphasis on performance at the cost of the 
development  of  interpersonal  skills.  Interpersonal  skills,  however,  are 
crucial in the later stages of high potential careers, when they must take 
up  leadership  roles  and  manage  and  inspire  others  (Dulewicz  et  al., 
2005).  Organizations  focusing  too  much  on  rapid  career  advancement 
may unintentionally hamper the learning processes of high potentials in 
general.  Previous  research  has  demonstrated  that  performance 
objectives  may  override  learning  new  skills  and  behaviors  when 
management  is  very  focused  on  short‐term  performance  (e.g. 
Pepermans  et  al.,  2003).  Second,  through  the  systematic  depreciation 
and  inadequate  rewarding  of  employees  on  untraditional  career  tracks 
(e.g. Lucas, Liu & Buzzanell, 2006), organizations are partly responsible 
for creating career success frameworks frustrating a large part of their 
workforce, since they cannot offer regular promotions and pay raises to 
every  employee  who  want  to  achieve  ‘a  successful  career’  (Granrose  & 
Baccili,  2006).  Instead  of  promoting  only  one  notion  of  career  success, 
organizations  might  try  to  build  more  inclusive  frameworks  of  career 
success  based  on  their  values  and  mission  statement.  Personnel 
development could be then linked to this broader career success model 
and, consequently, to different aspects of the organization’s goals. Such a 
‘diversity’  approach  to  talent  management  would  possibly  cause  more 
employees to feel successful (Bossuyt & Dries, 2008).  
331
 
A fourth, related, implication is that organizations may need  to expand 
their  notions  of  ‘talent’  in  order  to  survive  in  the  war  for  talent 
(Michaels, Handfield‐Jones & Axelrod, 2001). As ‘traditional’ talent pools 
(i.e.  young,  highly  educated,  highly  motivated  ‘workaholics’)  are 
becoming smaller due to demographic (Tucker, Kao & Verma, 2005) and 
psychological contract drivers (Sullivan, 1999), organizations will have 
to reflect about the potential value of investing in non‐traditional talent 
(e.g. career immigrants), thus adopting a ‘developmental’ as opposed to 
a ‘selection’ approach to talent management (e.g. Briscoe & Hall, 1999). 
In order to be able to do so, organizations will have to figure out which 
(leadership)  competencies  are  innate,  and  which  can  be  developed 
(Boyatzis & Saatcioglu, 2008).  
 
Individual  career  actors,  from  their  side,  should  try  to  establish  a 
personal and dynamic framework of what career success means to them 
personally.  Our  research  indicates  that  employees  identified  as  high 
potentials typically act as protean career actors, implying that they self‐
manage  their  careers  departing  from  their  personal  values  framework 
(Briscoe  &  Hall,  2006).  For  other  employees,  as  well,  awareness  of  the 
different meanings career success can take on might help them in their 
careers, if only by instilling in them the relative value of ‘career success’. 
Furthermore,  previous  research  has  indicated  that  the  more  satisfied 
employees  are  with  their  career,  the  fewer  initiatives  they  take  to 
enhance their external employability and the less interest they show in 
external career counseling (Verbruggen, Sels & Forrier, 2007). However, 
individual  career  actors  need  to  be  realistic  about  the  career  promises 
organizations  today  are  willing  and  able  to  make  considering  the 
current  economic  climate  (Granrose  &  Baccili,  2006).  As  organizations 
can  no  longer  guarantee  lifelong  careers,  developing  personal 
employability is an asset in the labor market (Van Buren, 2003). 
 
Summary and contributions 
 
A  ‘talent  management  paradox’  was  identified  based  on  opposing 
statements  made  about  talent  management  in  the  postmodern  careers 
literature  and  in  the  management  literature,  respectively.  Put  briefly, 
the talent management paradox describes how: 
 
(1) talent management is simultaneously depicted as utterly outdated, and as 
more pivotal than ever for the competitive advantage of organizations; and  
 
(2) talent management practices seem to be aimed at retaining exactly those 
employees who are most likely to leave. 
332
 
The  current  dissertation  presented  eight  papers  laid  out  over  four 
chapters,  each  of  them  responding  to  calls  in  the  literature  for  more 
research  about  talent  management  and  the  different  meanings 
attributed  to  ‘career  success’.  Qualitative  and  quantitative  approaches 
were combined, and different levels of analysis (i.e. organizational‐level, 
individual‐level,  or  both)  were  adopted.  The  postmodern  career  was 
studied  both  as  focus  and  context  of  the  research.  Hence,  the 
management  literature  was  integrated  with  the  postmodern  careers 
literature. Overall, this dissertation’s findings indicated that: 
 
(1)  Talent  management  is  considered  a  crucial  HRM  challenge  by  the 
majority  of  contemporary  organizations.  However,  in  many  of  these 
organizations, talent management programs are not yet in a mature stage of 
implementation.  Hence,  rather  than  being  outdated,  talent  management 
emerges as an ongoing challenge faced by organizations today. Organizations 
do  admit  that  there  are  fewer  opportunities  to  ‘climb  the  corporate  ladder’ 
today  than  there  were  a  decade  ago.  However,  these  dynamics  have  not 
marked  the  end  of  talent  management  programs.  Organizations  still  prefer 
internal  successors  for  top  management  positions  so  that  some  form  of 
continuity  within  the  organization  can  be  guaranteed.  As  a  result,  however, 
the criteria employees have to meet in order to enter into talent management 
programs  have  become  stricter.  In  the  future,  a  broader  approach  to  talent 
management may become necessary for organizations to be able to compete 
in the ‘war for talent’.  
 
(2) On average, employees identified as high potentials are found to be more 
committed to their organizations than employees who were not identified as 
high potentials. However, high potentials’ organizational commitment stems 
not  so  much  out  of  loyalty,  than  it  is  attributable  to  their  perceptions  of 
‘career  promises’  associated  with  their  high  potential  label.  These  findings 
imply  that  expectations  management  by  organizations  is  crucial  for  the 
retention of high potentials.  
 
References 
 
Abele,  A.E.  &  Spurk,  D.  (2009).  How  do  objective  and  subjective  career  success  interrelate  over 
time?  Journal  of  Occupational  and  Organizational  Psychology,  early  online,  corrected 
proof.  
Abrams, J., Castermans, S., Cools, H., Michielsen, M., Moeyaert, B., Van Meeuwen, N. & Van Nooten, L. 
(2008).  Leren  en  talent  management:  factoren  die  het  behouden  van  (talentvolle) 
werknemers beïnvloeden [Learning and talent management: Factors affecting the retention 
of (talented) employees]. KULeuven, Belgium: Unpublised research report.  
Adamson,  S.J.,  Doherty,  N.  &  Viney,  C.  (1998).  The  meanings  of  career  revisited:  Implications  for 
theory and practice. British Journal of Management, 9, 251‐259. 
Altman,  Y.  (1997).  The  high  potential  fast‐flying  achiever:  Themes  from  the  English  language 
literature 1976‐1995. Career Development International, 2 (7), 324‐330. 
APA  Ethical  Principles  of  Psychologists  and  Code  of  Conduct  (2002).  Retrieved  on  September  20th, 
2009 from http://www.apa.org/ethics/code2002.html. 
Austin, J.T., Scherbaum, C.A. & Mahlman, R.A. (2002). History of research methods in industrial and 
organizational  psychology:  Measurement,  design,  analysis.  In  S.G.  Rogelberg  (Ed.), 
Handbook  of  research  methods  in  industrial  and  organizational  psychology  (pp.  3‐33). 
Malden, MA: Blackwell Publishing.  
333
Bar‐On, R. (1997). Emotional Quotient Inventory: Technical Manual. Toronto, Canada: Multi‐Health 
Systems.  
Baruch,  Y.  &  Peiperl,  M.  (1997).  High‐flyers:  Glorious  past,  gloomy  present,  any  future?  Career 
Development International, 2 (7), 354‐358. 
Benko, C.  & Weisberg, A.C. (2007).  Mass career customization:  Aligning the workplace with today's 
nontraditional workforce. Boston, MA: Harvard Business School Publishing.  
Blau,  G.  J.  (1989).  Testing  the  generalizability  of  a  career  commitment  measure  and  its  impact  on 
employee turnover. Journal of Vocational Behavior, 35 (1), 88‐103. 
Bossuyt, T. & Dries, N. (2008). Talent management en flexibele loopbaanpaden voor de werknemers 
van  morgen  [Talent  management  and  flexible  career  tracks  for  tomorrow's  workforce]. 
In  H.  Calmeyn,  K.  De  Witte,  &  J.  Weverbergh  (Eds.),  Licht  op  leren  2008:  Leren  en 
ontwikkelen in een talentgerichte maatschappij [Spotlight on  learning 2008: Learning and 
developing in a talent­oriented society] (pp. 55‐90). Leuven, Belgium: LannooCampus.  
Boudreau,  J.W.  &  Ramstad,  P.M.  (2005).  Where’s  your  pivotal  talent?  Harvard  Business  Review,  83 
(4), 23‐24. 
Boudreau, J.W. & Ramstad, P.M. (2006, March). Talentship and human resource measurement and 
analysis: From ROI to strategic organizational change. Human Resource Planning. 
Bourne, H. & Jenkins, M. (2005) Eliciting managers’ personal values: An adaptation of the laddering 
interview method. Organizational Research Methods, 8 (4), 410‐428.  
Boyatzis, R.E. & Saatcioglu, A. (2008). A twenty‐year review of trying to develop emotional, social, 
and  cognitive  intelligence  competencies  in  graduate  management  education.  Journal  of 
Management Development, 27 (3), 92‐108. 
Briscoe, J.P.  & Hall, D.T. (1999).  Grooming and picking leaders using  competency frameworks: Do 
they  work?  An  alternative  approach  and  new  guidelines  for  practice.  Organizational 
Dynamics, 28 (2), 37‐52.  
Briscoe, J.P. &  Hall, D.T.  (2006).  The interplay of boundaryless  and protean careers: combinations 
and implications. Journal of Vocational Behavior, 69, 1‐18.  
Cable,  D.M.  &  Parsons,  C.K.  (2001).  Socialization  tactics  and  person‐organization  fit.  Personnel 
Psychology, 54 (1), 1‐23.  
Campbell, D. & Lee, C. (1988). Self appraisal in performance evaluation: Development vs. evaluation. 
Academy of Management Review, 13 (2), 302‐314. 
Cameron, K.S., & Quinn, R.E. (1999). Diagnosing and changing organizational culture. Reading, MA: 
Addison‐Wesley. 
Chudzikowski, K., Demel, B., Mayrhofer, W., Briscoe, J.P., Unite, J., Bogicevic Milikic, B., Hall, D.T., Las 
Heras, M.  & Zikic, J. (2009). Career transitions and their causes: A country‐comparative 
perspective.  Journal  of  Occupational  and  organizational  psychology,  early  online, 
corrected proof.  
Collings,  D.G.  &  Mellahi,  K.  (2009).  Strategic  talent  management:  A  review  and  research  agenda. 
Human Resource Management Review, early online, corrected proof.  
Conway, N. & Briner, R. (2005). Understanding psychological contracts at work. Oxford, UK: Oxford 
University Press.  
Cope, F. (1998). Current issues in selecting high potentials. Human Resource Planning, 21 (3), 15‐17. 
Crampton,  S.M.  &  Wagner,  J.A.  (1994).  Percept‐percept  inflation  in  micro‐organizational  research: 
An investigation of prevalence and effect. Journal of Applied Psychology, 79, 67‐76.  
Creelman, D. (2004, September). Return on investment in talent management:  Measures you can put 
to work right now. Washington, DC: Human capital institute research report.  
Cronbach, L.J., &  Meehl, P.E. (1955). Construct validity in  psychological tests. Psychological Bulletin, 
52, 281‐302. 
Day,  R.  &  Allen,  T.D.  (2004).  The  relationship  between  career  motivation  and  self‐efficacy  with 
protégé career success. Journal of Vocational Behavior, 64 (1), 72‐91.  
De Gieter, S., De Cooman, R., Pepermans, R., & Jegers, M. (2008). Manage through rewards, not only 
through  pay:  establishing  the  Psychological  Reward  Satisfaction  Scale  (PReSS).  In  M. 
Vartiainen, C. Antoni, X. Baeten, & R. Lucas (Eds.), Reward Management: Facts and Trends 
in Europe (pp. 97‐117). Lengerich, Germany: Pabst Science Publishers.  
Derr,  C.B.  (1986).  Managing  the  new  careerists:  The  diverse  career  success  orientations  of  today's 
workers. San Francisco, CA: Jossey‐Bass. 
Diener, E., Emmons, R.A., Larsen, R.J., & Griffin, S. (1985). The satisfaction with life scale. Journal of 
Personality Assessment, 49, 71‐75. 
334
Dries,  N.,  Van  Loo,  S.  &  Pepermans,  R.  (2009).  Towards  a  truly  subjective  conception  of  career 
success.  Paper  presented  at  the  25th  annual  conference  of  the  European  Group  for 
Organization Studies in Barcelona, Spain.  
Dubouloy, M. (2004). The transitional space and self‐recovery: A psychoanalytical approach to high 
potential managers’ training. Human Relations, 57 (4), 467‐496.  
Dulewicz,  C.,  Young,  M.  &  Dulewicz,  V.  (2005).  The  relevance  of  emotional  intelligence  for 
leadership performance. Journal of General Management, 30 (3), 71‐86. 
Duxbury, L.,  Dyke, L., & Lam, N. (1999). Career development in the federal public service: Building a 
world­class workforce. Ottawa, Canada: Treasury Board of Canada Secretariat. 
Dyer,  S.  &  Humphries,  M.  (2002).  Normalising  workplace  change  through  contemporary  career 
discourse.  In  L.  Morrow,  I.  Verins  &  E.  Willis  (Eds.),  Mental  health  and  work:  Issues  and 
perspectives (pp. 48‐62). Bedford Park, Asutralia: Auseinet.  
Eisenberg,  E.M.  (1984).  Ambiguity  as  strategy  in  organizational  communication.  Communication 
Monographs, 51, 227‐242.  
Evetts, J. (1992). Dimensions of career: Avoiding reification in the analysis of change. Sociology, 26, 
1‐21. 
Fields, G. S. (2002). Predicting potential for promotion: How the data in human resource information 
systems  can  be  used  to  help  organizations  gain  competitive  advantage  (CAHRS  Working 
Paper  #02‐14).  Ithaca,  NY:  Cornell  University,  School  of  Industrial  and  Labor  Relations, 
Center for Advanced Human Resource Studies. 
Fowler, F.J. (2002). Survey research methods. London, UK: Sage Publishers.  
Fresina,  T.  (1987).  The  identification  &  development  of  high  potential  managers:  A  benchmarking 
report on midsized & Fortune 500 companies in all major industries. Palatine, IL: Executive 
Knowledgeworks.  
Furnham,  A.  (2004).  The  future  (and  past)  of  work  psychology  and  organizational  behaviour:  A 
personal view. Management Revue, 15 (4), 420‐436.  
Garrow,  V.  &  Hirsch,  W.  (2008).  Talent  management:  Issues  of  focus  and  fit.  Public  Personnel 
Management, 37 (4), 389‐402.  
Göbel‐Kobialka,  S.  (1998).  Reaching  business  excellence  through  sound  people  management. 
European Journal of Work and Organizational Psychology, 7 (4), 549‐556.  
Granrose,  C.S.  &  Baccili,  P.A.  (2006).  Do  psychological  contracts  include  boundaryless  or  protean 
careers? Career Development International, 11 (2), 163‐182.  
Hackman, J. R., & Oldham, G. R. (1980), Work redesign. Reading, MA: Addison‐Wesley.  
Heslin,  P.A.  (2005).  Conceptualizing  and  evaluating  career  success.  Journal  of  Organizational 
Behavior, 26 (2), 113‐136. 
Hollenbeck, G.P., McCall, M.W. & Silzer, R.F. (2006). Theoretical and practitioner letters: Leadership 
competency models. The Leadership Quarterly, 17, 398‐413.  
Howitt,  D.  &  Cramer,  D.  (2005).  Introduction  to  Research  Methods  in  Psychology.  Harlow,  UK: 
Pearson Education. 
Iles,  P.  (1997).  Sustainable  high  potential  career  development:  A  resource‐based  view.  Career 
Development International, 2 (7), 347‐353.  
Jansen, P.G.W., & Vinkenburg, C.J. (2006). Predicting management career success from assessment 
center data: a longitudinal study. Journal of Vocational Behavior, 68, 253‐266.  
Kierein,  N.M.  &  Gold,  M.A.  (2000).  Pygmalion  in  work  organizations:  A  meta‐analysis.  Journal  of 
Organizational Behavior, 21, 913‐928.  
Krueger,  J.  (1998).  Enhancement  bias  in  descriptions  of  self  and  others.  Personality  and  Social 
Psychology Bulletin, 24 (5), 505‐516.  
Lewis,  R.E.  &  Heckman,  R.J.  (2006).  Talent  management:  A  critical  review.  Human  Resources 
Management Review, 16 (2), 139‐154. 
Lombardo,  M.M.  &  Eichinger,  R.W.  (2000).  High  potentials  as  high  learners.  Human  Resource 
Management, 39 (4), 321‐330.  
Lucas, K., & Buzzanell, P. M. (2004). Blue‐collar work, career, and success: Occupational narratives 
of sisu. Journal of Applied Communication Research, 32, 273‐292. 
Lucas,  K.,  Liu,  M.,  &  Buzzanell,  P.  M.  (2006).  No  limits  careers:  A  critical  examination  of  career 
discourse in the U.S. and China. In M. Orbe, B. J. Allen, & L. A. Flores (Eds.), International 
and intercultural communication annual 28. Thousand Oaks, CA: Sage.  
McEnrue,  M.P.  &  Groves,  K.  (2006).  Choosing  among  tests  of  emotional  intelligence:  What  is  the 
evidence? Human Resource Development Quarterly, 17 (1), 9‐42. 
Michaels,  E.,  Handfield‐Jones,  H.,  &  Axelrod,  B.  (2001).  The  war  for  talent.  Boston,  MA:  Harvard 
Business School Press. 
335
Nicholson, N. & de Waal‐Andrews, W. (2005). Playing to win: Biological imperatives, self‐regulation, 
and trade‐offs in  the game of career success. Journal  of Organizational  Behavior,  26  (2), 
137‐154.  
Pepermans,  R.,  Vloeberghs,  D.  &  Perkisas,  B.  (2003).  High  potential  identification  policies:  An 
empirical  study  among  Belgian  companies.  Journal  of  Management  Development,  22  (8), 
660‐678. 
Pfeffer, J. (1981). Management as symbolic action: The creation and maintenance of organizational 
paradigms. In L.L. Cummings & B.M. Staw (Eds.), Research in organizational behavior (1, 
pp. 1‐52). Greenwich, CT: JAI Press. 
Podsakoff,  P.M.,  MacKenzie,  S.B.,  Lee,  J.Y.,  &  Podsakoff,  N.P.  (2003).  Common  method  biases  in 
behavioral  research:  A  critical  review  of  the  literature  and  recommended  remedies. 
Journal of Applied Psychology, 88, 879‐903. 
Poole  M.E.,  Langan‐Fox  J.,  &  Omodei  M.  (1993).  Contrasting  subjective  and  objective  criteria  as 
determinants  of  perceived  career  success.  Journal  of  Occupational  and  Organizational 
Psychology. 66 (1), 39‐54. 
Robinson,  J.A.,  &  Taylor,  L.R.  (1998).  Autobiographical  memory  and  self‐narratives:  A  tale  of  two 
stories.  In  C.P.  Thompson,  D.J.  Herrmann,  D.  Bruce,  J.D.  Read,  D.G.  Payne  &  M.  P.  Toglia 
(Eds.),  Autobiographical  memory:  Theoretical  and  Applied  Perspectives.  Mahwah,  N.J.: 
Erlbaum. 
Rothwell,  W.  &  Poduch,  S.  (2004).  Introducing  technical  (not  managerial)  succession  planning. 
Public Personnel Management, 33(4), 405–420. 
Rottinghaus,  P.  J.,  Day,  S.  X.,  &  Borgen,  F.  H.  (2005).  The  Career  Futures  Inventory:  A  measure  of 
career‐related adaptability and optimism. Journal of Career Assessment, 13 (1), 3‐24. 
Rousseau,  D.M.  (2000).  Psychological  Contract  Inventory  technical  report.  Carnegie  Mellon 
University, Pittsburgh, PA.  
Roussillon,  S.  &  Bournois,  F.  (2002).  Identifying  and  developing  future  leaders  in  France.  In  C.B. 
Derr,  S.  Roussillon  &  F.  Bournois  (Eds.),  Cross­cultural  approaches  to  leadership 
development (pp. 51‐60). Westport, CT: Quorum Books.  
Schein,  E.H.  (1978).  Career  dynamics:  Matching  individual  and  organizational  needs.  Reading,  MA: 
Addison‐Wesley.  
Schwarz, A., Rizzuto, T., & Schwarz, C. (2008). Urban legend of common method bias: nine common 
errors  and  their  impacts.  Paper  presented  at  the  Hawaii  International  Conference  on 
System Sciences, 7‐10 January, Manoa, HI. 
Segalla, M., Rouziès, D.  & Flory, M. (2001). Culture and career advancement in Europe: Promoting 
team players vs. fast trackers. European Management Journal, 19 (1), 44‐57. 
Sels  L.,  Van  Woensel  A.  &  Herremans,  W.  (2008).  Over  rode,  oranje  en  groene  lichten  in  het 
eindeloopbaanbeleid [About  red, orange and green  lights in the ‘career ending policies’]. 
Over.Werk. Tijdschrift van het Steunpunt WSE, 1, 8 ‐33. 
Spector,  P.E.  (2006).  Method  variance  in  organizational  research:  truth  or  urban  legend? 
Organizational Research Methods, 9 (2), 221‐232.  
Spreitzer, G.M., McCall, M.W. & Mahoney, J.D. (1997). Early identification of international  executive 
potential. Journal of Applied Psychology, 82 (1), 6‐29. 
Sullivan,  S.E.  (1999).  The  changing  nature  of  careers:  a  review  and  research  agenda.  Journal  of 
Management, 25 (3), 457‐484. 
Taris,  T.W.  &  Kompier,  M.A.J.  (2003).  Challenges  in  longitudinal  research  designs  in  occupational 
health psychology. Scandinavian Journal of Work, Environment and Health, 29 (1), 1‐4.  
Trochim, W. & Donnelly, J.P. (2007). The Research Methods Knowledge Base. Cincinnati, OH: Atomic 
Dog Publishers. 
Tucker,  E.,  Kao,  T.  &  Verma,  N.  (2005,  July/August).  Next‐generation  talent  management:  Insights 
on  how  workforce  trends  are  changing  the  face  of  talent  management.  Business  Credit, 
107, 20‐27.  
Tulgan, B. (2001). Winning the talent wars. Employment Relations Today, 23 (1­2), 37‐51.  
Van  Buren,  H.J.  III  (2003).  Boundaryless  careers  and  employability  obligations.  Business  Ethics 
Quarterly, 13 (2), 131‐149. 
Van  Velsor,  E.  &  Leslie,  J.B.  (1995).  Why  executives  derail:  perspectives  across  time  and  cultures. 
Academy of Management Executive, 9 (4), 62‐72. 
Verbruggen,  M.,  Sels,  L.  &  Forrier,  A.  (2007).  Unraveling  the  relationship  between  organizational 
career  management  and  the  need  for  external  career  counselling.  Journal  of  Vocational 
Behavior, 71 (1), 69 ‐ 83. 
336
Viney, C., Adamson, S. & Doherty, N. (1997). Paradoxes of fast‐track career management. Personnel 
Review, 26 (3), 174‐186. 
Vinkenburg,  C.  &  Pepermans,  R.  (2005).  Top  potentials  in  organisaties  –  Een  inleiding  [Top 
potentials in organizations – An introduction]. In C. Vinkenburg  & R. Pepermans (Eds.), 
Top  potentials  in  organisaties:  Identificeren,  ontwikkelen,  binden  [Top  potentials  in 
organizations:  Identification,  development,  engagement]  (pp.  3‐16).  Assen,  The 
Netherlands: van Gorcum.  
 
Young people today are increasingly for instance, people tend to look up
focused on themselves, on their to high-potential co-workers see

Different ladder, different story?


work-life balance. They are less
prepared to make sacrifices even Different them as role models rather than
be jealous of them. It is standard
so, people are still very sensitive procedure to start discriminating

Different ladder, different story?


Dissecting the talent management paradox within the
framework of the postmodern career
about being promoted or not In this
l a d d e r , between employees some years after
kind of organization, when you’re entry, and everyone is aware of this

Putting the literature on talent management alongside postmodern


not promoted or you don’t get a pay
rise, it means they probably think
different There’s really no use being secretive
about who’s on the list and who isn’t.
careers literature, a twofold ‘talent management paradox’ seems to arise. you’re not that good at what you s t o r y ? People will talk to each other anyway
First of all, talent management is simultaneously depicted as utterly do lateral moves can definitely be they’ll figure it out themselves. A
outdated, and as more pivotal than ever for the competitive advantage of challenging. It takes up a lot of time decade ago, about 10 to 15 percent
D i s s e c t i n g
organizations. Second, talent management practices seem to be aimed at and effort to take on this whole new of our employees were considered
retaining exactly those employees who are most likely to leave. role in the organization and I’m sure t h e high potentials. There were more
it’s gratifying when you succeed in opportunities to climb the ladder
The current dissertation presents four chapters covering eight papers,
a transition like that But you don’t t a l e n t back then and we figured, if some of
each approaching the topic of talent management within the context
really feel successful unless you get them didn’t meet expectations we
of the postmodern career from a different vantage point. It aims to
a real promotion or a rise and then
management could always take them off the list.
contribute to the field in a threefold manner: first, by integrating the
careers literature with the talent management literature, thus creating
there’s the fact that a lot of people
paradox But today we’re much stricter about
are involved in project work all the who gets on the list in the first place
a comprehensive framework; second, by building on empirical data time, so every time you get started it’s about 2 percent worldwide. For
collected from different talent management stakeholders (i.e. general on a new project, that’s already a w i t h i n t h e instance, we’ll probably only pay
managers, HRM professionals, and individual employees identified as lateral move So it doesn’t really mean for an expensive training program
high potentials); and third, by examining cross-level effects, in that anything unless it’s an upward move.
framework when a high potential is asking.

Nicky Dries
organization-level variables (i.e. the classification of employees into It’s dangerous when the system is all o f t h e Talented individuals can get stuck
different ‘talent categories’) are related to individual-level variables (i.e. about status. Then people start to
postmodern in the leadership pipeline when
career antecedents and outcomes). think that you have to be on the list to organizations do not succeed

career
be somebody The problem with that in developing them properly.
kind of mindset is that high potentials Organizations usually have a two
Nicky Dries

usually make up only a really to three-year vision, but then their


www.vubpress.be small portion of an organization’s development activities reach a
population, so that would imply plateau Young people today are
that everyone else is frustrated. All increasingly focused on themselves,
N i c k y D r i e s
employees want to see extra effort on their work-life balance. They are
translated into extra rewards and Advisor: less prepared to make sacrifices even
9 789054 876700 recognition but when communication Prof. Dr. Roland so, people are still very sensitive

ISBN 978 90 5487 670 0 [about workforce segmentation] is Pepermans about being promoted or not In this
too explicit, resentment between kind of organization, when you’re
colleagues will probably arise. This not promoted or you don’t get a pay
Dissertation submitted to obtain the degree of Doctor in Psychological Sciences
applies especially to the Belgian rise, it means they probably think
Financial support: Research Foundation – Flanders (FWO TM490)
culture. I get the impression that you’re not that good at what you
Vrije Universiteit Brussel, 2009
in America for instance, people do lateral moves can definitely be
tend to look up to high-potential challenging. It takes up a lot of time
View publication stats

Вам также может понравиться