Академический Документы
Профессиональный Документы
Культура Документы
net/publication/258834538
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:
All content following this page was uploaded by Nicky Dries on 27 May 2014.
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
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,
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)
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
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)
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
Table 3. Goodness‐of‐fit statistics for the four factors of the proposed model
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
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
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
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
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
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
* 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
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
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
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
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
a
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
a
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
a
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
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
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