Вы находитесь на странице: 1из 482


Series Editors: Wilfred J. Zerbe, Charmine E. J. Härtel
and Neal M. Ashkanasy
Recent Volumes:
Volume 2: Individual and Organizational Perspectives on Emotion
Management and Display Edited by Wilfred J. Zerbe,
Neal M. Ashkanasy and Charmine E. J. Härtel
Volume 3: Functionality, Intentionality and Morality Edited by
Wilfred J. Zerbe, Neal M. Ashkanasy and
Charmine E. J. Härtel
Volume 4: Emotions, Ethics and Decision-Making Edited by Wilfred
J. Zerbe, Charmine E. J. Härtel and Neal M. Ashkanasy
Volume 5: Emotions in Groups, Organizations and Cultures Edited by
Charmine E. J. Härtel, Neal M. Ashkanasy and
Wilfred J. Zerbe
Volume 6: Emotions and Organizational Dynamism Edited by
Wilfred J. Zerbe, Charmine E. J. Härtel and
Neal M. Ashkanasy
Volume 7: What Have We Learned? Ten Years On Edited by
Charmine E. J. Härtel, Neal M. Ashkanasy and
Wilfred J. Zerbe
Volume 8: Experiencing and Managing Emotions in the Workplace
Edited by Neal M. Ashkanasy, Charmine E. J. Härtel and
Wilfred J. Zerbe
Volume 9: Individual Sources, Dynamics, and Expressions of Emotion
Edited by Wilfred J. Zerbe, Neal M. Ashkanasy and
Charmine E. J. Härtel
Volume 10: Emotions and the Organizational Fabric Edited by
Neal M. Ashkanasy, Wilfred J. Zerbe and
Charmine E. J. Härtel

University of Queensland, Queensland, Australia

Memorial University of Newfoundland,
Newfoundland, Canada

University of Queensland, Queensland, Australia

United Kingdom North America Japan

India Malaysia China
Emerald Group Publishing Limited
Howard House, Wagon Lane, Bingley BD16 1WA, UK

First edition 2015

Copyright r 2015 Emerald Group Publishing Limited

Reprints and permissions service

Contact: permissions@emeraldinsight.com

No part of this book may be reproduced, stored in a retrieval system, transmitted in

any form or by any means electronic, mechanical, photocopying, recording or
otherwise without either the prior written permission of the publisher or a licence
permitting restricted copying issued in the UK by The Copyright Licensing Agency
and in the USA by The Copyright Clearance Center. Any opinions expressed in the
chapters are those of the authors. Whilst Emerald makes every effort to ensure the
quality and accuracy of its content, Emerald makes no representation implied or
otherwise, as to the chapters’ suitability and application and disclaims any warranties,
express or implied, to their use.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN: 978-1-78560-221-4
ISSN: 1746-9791 (Series)

ISOQAR certified
Management System,
awarded to Emerald
for adherence to
ISO 14001:2004.

Certificate Number 1985

ISO 14001
I dedicate this book to the greatest love. To Günter, Jameson and Jasmin
who are always unwavering in their love to me and who are my greatest
loves. To my dear friends whose full embrace of all that I am remains
steadfast no matter how our roads have travelled over time. And to my
beloved sister Sharron, lost to cancer a year ago, with whom I shared the
greatest love. And with the deepest gratitude to my brother-in-law Tom,
who showed the greatest love, never failing to safeguard her dignity and
wellbeing as cancer ravaged her body and she slipped away. And to Pat,
who showed the greatest love as a lifelong friend to her and to me, not
shying away from our pain and suffering, and travelling the long distance
to be by my side to grieve and to share what is the greatest love. Although
our beloved Sharron’s body is gone, she continues giving the greatest love
to us, and always will have our greatest love. This poem remembers her,
celebrating the greatest love.

The Greatest Love

To my sister Sharron
Who gave the greatest love,
Tending to my wounds
Shining a beacon of hope
Loving me unwaveringly.
To my brother-in-law Tom
Who showed the greatest love
To my sister Sharron
Tending to her wounds
Loving her unwaveringly
Holding her in his arms as she slipped away.
To my sister Sharron
The greatest love.

C. E. J. H.
This page intentionally left blank








Suzanne J. Peterson, Christopher S. Reina, 3
David A. Waldman and William J. Becker


Michael J. Gill 29


Marilena Antoniadou, Peter John Sandiford, 51
Gillian Wright and Linda Patricia Alker


Markus Plate 81



Dirk Lindebaum 109


Patricia L. Baratta and Jeffrey R. Spence 139




Olof Brunninge and Anders Melander 175


Magdalena Markowska, Charmine E. J. Härtel, 215
Ethel Brundin and Amanda Roan


Sandra Kiffin-Petersen 241


Jim A. McCleskey 271
Contents ix




Nuno Da Camara, Victor Dulewicz and Malcolm Higgs 297


Sanjeewa Perera and Carol T. Kulik 341


Kathryn Moura, Ashlea C. Troth and Peter J. Jordan 369


Avina J. Mendonca, Nidhi Mishra and Sanket S. Dash 397


Jason J. Dahling, Sophie A. Kay and Nickolas F. Vargovic 423


Charmine E. J. Härtel and Jennifer M. O’Connor 443


This page intentionally left blank

Linda Patricia Alker Manchester Metropolitan University

Business School, Manchester Metropolitan
University, Manchester, UK
Marilena Antoniadou Manchester Metropolitan University
Business School, Manchester Metropolitan
University, Manchester, UK
Neal M. Ashkanasy UQ Business School, The University of
Queensland, Brisbane, Queensland,
Patricia L. Baratta Department of Psychology, University of
Guelph, Guelph, Ontario, Canada
William J. Becker Department of Management, Neeley
School of Business, Texas Christian
University, Fort Worth, TX, USA
Ethel Brundin Jönköping International Business School,
Jönköping University, Jönköping, Sweden
Olof Brunninge Jönköping International Business School,
Jönköping University, Jönköping, Sweden
Nuno Da Camara Southampton Business School, University
of Southampton, Southampton, UK
Jason J. Dahling Department of Psychology, The College of
New Jersey, Ewing, NJ, USA
Sanket S. Dash Indian Institute of Management
Ahmedabad, India
Victor Dulewicz Henley Business School, University of
Reading, Henley-on-Thames, UK


Michael J. Gill School of Management, University of

Bath, Bath, UK
Charmine E. J. Härtel UQ Business School, The University of
Queensland, Brisbane, Queensland,
Malcolm Higgs Southampton Business School, University
of Southampton, Southampton, UK
Peter J. Jordan Griffith Business School, Griffith
University, Brisbane, Queensland,
Sophie A. Kay Department of Psychology, The College of
New Jersey, Ewing, NJ, USA
Sandra Kiffin-Petersen Business School, University of Western
Australia, Crawley, Western Australia,
Carol T. Kulik School of Management, City West
Campus, University of South Australia,
Adelaide, South Australia, Australia
Dirk Lindebaum University of Liverpool Management
School, University of Liverpool,
Liverpool, UK
Magdalena Markowska Jönköping International Business School,
Jönköping University, Jönköping, Sweden
Jim A. McCleskey North American University, Houston, TX,
Anders Melander Jönköping International Business School,
Jönköping University, Jönköping, Sweden
Avina J. Mendonca Indian Institute of Management
Ahmedabad, India
Nidhi Mishra Indian Institute of Management
Ahmedabad, India
List of Contributors xiii

Kathryn Moura Griffith Business School, Griffith

University, Brisbane, Queensland,
Jennifer M. O’Connor UQ Business School ,The University of
Queensland, Brisbane, Queensland,
Sanjeewa Perera School of Management, City West
Campus, University of South
Australia, Adelaide, South Australia,
Suzanne J. Peterson Department of Management, W.P. Carey
School of Business, Arizona State
University, Tempe, AZ, USA
Markus Plate Jönköping International Business School,
Jönköping University, Jönköping,
Christopher S. Reina Department of Management, W.P. Carey
School of Business, Arizona State
University, Tempe, AZ, USA
Amanda Roan UQ Business School, The University of
Queensland, Brisbane, Queensland,
Peter John Sandiford Business School, Faculty of the
Professions, The University of Adelaide,
Adelaide, South Australia, Australia
Jeffrey R. Spence Department of Psychology, University of
Guelph, Guelph, Ontario, Canada
Ashlea C. Troth Griffith Business School, Griffith
University, Brisbane, Queensland,
Nickolas F. Vargovic Department of Psychology, The College of
New Jersey, Ewing, NJ, USA

David A. Waldman Department of Management, W.P. Carey

School of Business, Arizona State
University, Tempe, AZ, USA
Gillian Wright Manchester Metropolitan University
Business School, Manchester Metropolitan
University, Manchester, UK
Wilfred J. Zerbe Faculty of Business Administration,
Memorial University of Newfoundland, St.
John’s, Newfoundland, Canada

Charmine E. J. Härtel is Full Professor of Management at The University

of Queensland Business School. She is Life Fellow and Past President of
the Australian and New Zealand Academy of Management (ANZAM),
Fellow of the Australian Institute of Management (AIM), and Fellow of
the Australian Human Resources Institute (AHRI). She has extensive
experience in senior management roles and management consulting and is
recognized internationally as one of the originators of the study of emotion
in organizations and a leading expert in the strategies, systems, and prac-
tices underpinning positive organizations. Professor Härtel is recipient of
numerous national and international awards, including five awards for
innovation in organizational practice and seven nominations and awards
for excellence in research. She is recipient of nearly $3 million in Australian
Research Council funding as well as numerous research contracts from the
private and public sectors. Her publications appear in leading international
journals such as Academy of Management Review, Journal of Management,
British Journal of Industrial Relations, Human Resource Management
Review, Journal of Applied Psychology, and Leadership Quarterly. She is
Series Editor for Research on Emotion in Organizations and author of the
textbook Human Resource Management (Pearson), which emphasizes
understanding the employment relationship from a human well-being
Wilfred J. Zerbe is Professor of Organizational Behavior and Dean of the
Faculty of Business Administration at Memorial University of
Newfoundland. His research interests focus on emotions in organizations,
organizational research methods, service sector management, business
ethics, and leadership. His publications have appeared in books and jour-
nals including The Academy of Management Review, Industrial and Labor
Relations Review, Canadian Journal of Administrative Sciences, Journal of
Business Research, Journal of Psychology, Journal of Services Marketing,
and Journal of Research in Higher Education. He is Series Co-Editor of
Research on Emotion in Organizations.


Neal M. Ashkanasy is Professor of Management in the UQ Business School

at the University of Queensland, where he received his PhD in social/
organizational psychology. His research in leadership, organizational cul-
ture, ethics, and emotions in organizations has been continuously funded
by the Australian Research Council since 1996. His work is published in
leading journals such as the Academy of Management Journal and Review,
the Journal of Organizational Behavior, The Leadership Quarterly, the
Journal of Applied Psychology, the Journal of Personality and Social
Psychology, and the Journal of Management. He serves on several editorial
boards including the Journal of Applied Psychology and the Journal of
Management. He is Associate Editor for Emotion Review, and Series Editor
for Research on Emotion in Organizations, and has previously served as
Editor-in-Chief of the Journal of Organizational Behavior, and Associate
Editor for the Academy of Management Review and Academy of
Management Learning and Education. Professor Ashkanasy is a Fellow of
the Academy for the Social Sciences in the UK (AcSS) and Australia
(ASSA); the Association for Psychological Science (APS); the US Society
for Industrial and Organizational Psychology (SIOP); and the Australia
and New Zealand Academy of Management (ANZAM), where he served
as President in 2010.

Research on emotions in organizations has organically evolved through

academic scholarship. Today, as the field matures, there is growing atten-
tion to integrating the study of emotion in organizations within mainstream
theories and approaches across the sub-domains of organizational research.
This broadening of perspective is inspiring novel theoretical insights and
methodological approaches for studying emotions and worklife. The com-
plex nature of emotions and unique challenges posed in measuring emotion
provides fertile ground to craft new ways of researching the dimensions,
underlying phenomenon, causes, and the significant role that emotion plays
in organizations. The choice of theme of this volume reflects these new
ways of studying emotions in organizations, offering both practitioners and
academics a rich vein to mine for planning future endeavors. Chapters in
this volume showcase scholarly advances in empirical research and theory
development relating to existing notions of emotions in worklife, as well as
innovations in terms of the methods, theory, and contexts to which emo-
tions research is being applied.


Chapters in this volume are drawn from the best contributions to the
Ninth International Conference on Emotions and Worklife: “Emonet
IX” held in Philadelphia, USA, 2014. This popular scholarly bi-annual
conference, founded by the editors of this volume, was first held in 1998
and is fondly referred to as the “Emonet” conference, after the email
discussion listserv. The Emonet email group and conference were estab-
lished to provide a forum for the international scholarly network of
emotions researchers committed to understanding the roles of emotions
in organizational settings. The conference, and this book series, provide
a leading outlet for presentation of significant empirical and theoretical

research that advances knowledge and practice regarding emotion in

organizational life.
The chapters presented in this volume represent a selection of best
papers (as determined by blind peer review process) from the 2014 confer-
ence contributions, complimented with invited chapters to reflect outstand-
ing innovations relating to new ways of studying emotions and worklife.
We are especially grateful to the Emonet conference paper reviewers and
wish to acknowledge their assistance in the review process (see list of con-
ference reviewers). The next volume will be based on a selection of papers
accepted from the 2016 “Emonet X” conference, to be held in Rome, Italy.
Readers interested in learning more about the conferences or the Emonet
list should visit the Emonet website http://www.emotionsnet.org, where
they will find the conference program and paper abstracts.


The chapters in this volume are organized into three sections. Authors of the
first group of chapters in Section I, showcase methodological approaches
that offer innovative ways of studying and understanding emotions and
worklife. In Section II, the authors highlight contextualization developments
in studying emotions in organizations. The final set of chapters in
Section III, introduce novel areas for empirical investigation exemplifying
the role and underlying phenomena of emotions in organizational contexts.



The six chapters in this section utilize a diverse range of methodological

approaches previously underused or not applied to understanding and
assessing the role that emotion plays in organizational contexts. These
methodological approaches encompass a multi-method approach utilizing
psychological and physiological measures (Chapter 1), a phenomenological
methodology known as Interpretative Phenomenological Analysis (IPA)
(Chapter 2), a qualitative interpretive study drawing upon in-depth inter-
views (Chapter 3), a TV documentary case study providing insight in the
use of visual/TV material (Chapter 4), and a qualitative phenomenological
Introduction xix

study based on an inductive research approach (Chapter 5). The final chap-
ter in Section I (Chapter 6) utilizes an integrated analytical framework to
test the multidimensional constructs of boredom, which poses unique mea-
surement challenges, and results in support for a multidimensional model
that incorporates five dimensions.
In Chapter 1, Suzanne J. Peterson, Christopher S. Reina, David A.
Waldman, and William J. Becker propose a multi-method approach that
utilizes both psychological and physiological measures for understanding
and assessing emotions and affect in organizations. In so doing, they tackle
measurement issues commonly raised in assessing emotions. Moreover,
they review and discuss the ethical and practical implications associated
with various measurement forms, providing an important guide for practi-
tioners and academics considering the measurement of emotion in organi-
zational settings.
Chapter 2, by Michael J. Gill, reveals new and valuable insights demon-
strated through a novel phenomenological approach called Interpretative
Phenomenological Analysis (IPA). The aim of IPA is to capture contextua-
lized and rich accounts of how individuals make sense of their experiences.
In the case of emotions research, IPA can be used to explore individuals’
emotional experiences of organizational events and processes. The chapter
outlines the foundations of IPA which make it particularly suited to emo-
tions research, provides examples of how such research is conducted, and
gives due consideration to the criticisms and limitations of the approach.
In Chapter 3, the authors Marilena Antoniadou, Peter John Sandiford,
Gillian Wright, and Linda Patricia Alker describe an innovative interpre-
tive study of how university lecturers experience and manage fear. Drawing
upon in-depth interviews with nineteen lecturers in Cypriot universities,
Antoniadou and her colleagues found that fear is a major factor in the
working environment of Cypriot universities. Foci included fear of job loss,
fear of failure, fear of loss of status and reputation, and even fear of expos-
ing weaknesses in the classroom. On the more positive side, respondents
were more able to adapt to these threats if they were able to adopt mechan-
isms that gave them a sense of autonomy and confidence at work. To
enable this, however, the lecturers’ managers needed to ensure a supportive
working environment, positive mentoring, and to encourage them to speak
up whenever they encountered threatening situations.
In Chapter 4 Markus Plate applies a novel methodology to examine
shame, a powerful but under-studied emotion in organizational settings.
When an individual experiences shame, it means that face has been lost.
Using a case study approach involving a TV show, this chapter reveals how

shame can lead to the derailing of a leader through the negative feedback
loop put into motion by the shame experience of a leader.
Motivated by an effort to understand the dynamics and processes that
underlie emotional intelligence (EI), Dirk Lindebaum presents in Chapter 5
the results of a qualitative study that examines the lived experiences of indi-
viduals. As a result of this investigation, Lindebaum problematizes our
understanding of EI. For example, he concludes that courage and honesty
are necessary precursors to the ability to process emotional information.
This challenges the argument that EI unequivocally leads to emotional
growth. Lindebaum also concludes that the sequential and hierarchical
logic underlying conceptions of EI may not be supported by the thought
processes of individuals. Similarly, Lindebaum concludes that the way that
emotional challenges were processed was better described by the emotion
regulation literature than that of EI, and also that the dynamic of emo-
tional information processing may be curvilinear rather than linear as
assumed by models of EI. He relates these discrepancies to methodological
critiques of the MSCEIT and concludes that efforts to rethink the construct
of EI and in particular its underlying processes would advance theory and
Chapter 6 is the last chapter in this section. In it, Patricia L. Baratta and
Jeffrey R. Spence outline a study where they sought to understand the nat-
ure and the structure of boredom; and conclude that boredom is best mod-
eled as a multidimensional construct. In two studies involving
undergraduate psychology students, the authors compared different config-
urations of boredom, including superordinate and multivariate multidimen-
sional models and a unidimensional parallel model. Their results support a
multidimensional model based on five dimensions: disengagement, low
arousal, high arousal, inattention, and time perception. They conclude
however that, while the multivariate conceptualization of boredom is the
best fit to their data, it is nonetheless acceptable to view boredom as an
overarching construct that incorporates the five dimensions which they
refer to as a “superordinate construct.”



Section II comprises four chapters that exemplify contextualization develop-

ments to advance our understanding of the significant role emotion plays in
Introduction xxi

organizational processes, further influencing the changing nature of

studying emotions and worklife. In particular, these chapters provide
insight into the interplay between causes of emotion affects and underlying
phenomena that influence outcomes in organizational behavior and context.
The first chapter of Section II (Chapter 7), by Olof Brunninge and
Anders Melander, presents a processual study aimed at understanding the
longevity of family firms by focusing on the management of resources from
the perspective of financial and socioemotional wealth rationalities. The
novel perspective in this chapter pertains to the processual methodology
used and the use of a socioemotional wealth perspective (SEW), which
takes the view that family firms do not only try to maximize their financial
wealth, but also their socioemotional wealth. Both perspectives have much
to offer the study of all forms of organizations from an emotions lens.
In Chapter 8, Magdalena Markowska, Charmine E. J. Härtel, Ethel
Brundin, Amanda Roan introduce us to emotions in the entrepreneurial
setting. They conceptually build a model to illustrate how emotions play a
role for the crafting of the entrepreneurial identity. The authors suggest
that emotions play a significant role in this process where the interplay
between the context and the enactment of the role of being an entrepreneur
takes place. The model suggests that the drivers behind an individual’s deci-
sion to become an entrepreneur and the affective responses and their signif-
icant emotional experiences during the process of developing their
entrepreneurial role progresses toward an enactment or a “dis-enactment”
cycle of the entrepreneurial identity. Thus, emotions can be the decisive
impetus that makes or breaks the entrepreneurial identity process.
Work design, defined as “the content and organization of one’s work
tasks, activities, relationships, and responsibilities” (Parker, 2014, p. 662),
has largely overlooked the complex cognitive-emotional interaction in
understanding employee motivation and satisfaction. In Chapter 9, Sandra
Kiffin-Petersen reviews what we know about this interaction from research
on emotions and recent studies in neuroscience together with traditional
and emergent work design perspectives. Examining emotions and work
design is temporally relevant because of accumulating evidence supporting
the stress buffering effects of positive emotions and the ubiquitous presence
of emotional and interpersonal tasks in service jobs. The chapter introduces
the Self-Referential Emotional Regulatory Model (SERM) of work design
and concludes with the suggestion that a focal point of new research into
emotions in organizations should be on studying how to combine tasks,
activities, and relationships so as to promote positive affective states and
minimize the impact of negative emotions on employees’ well-being.

Emotional Intelligence (EI) has been a central topic in the emotions in

organizations literature since its rapid growth beginning 30 years ago. As
Jim A. McCleskey points out in Chapter 10, EI research and theorizing
have been limited by a plurality of conceptual definitions and measurement
methods. For example, EI has been conceptualized as an ability, a compe-
tency, or a trait and some theories of EI mix these conceptualizations.
Moreover, methods of assessing EI have also differed and have included
self-report and ability measures. McCleskey advocates for use of perfor-
mance based or “Stream One Ability Model” (SOAM) measures of EI,
such as the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT
v2.0; Mayer, Salovey, & Caruso, 2002). The MSCEIT has been criticized,
however, for weak construct validity. McCleskey therefore argues for
expanded use of situational judgment tests (SJTs) in studying EI within
SOAM. Chapter 10 reviews alternate SJTs and concludes with recommen-
dations for the future of SOAM EI research, in particular a revision of the



The six chapters in Section III represent novel areas of empirical investiga-
tion and new dimensions of researching emotion in organizational contexts.
In the following chapters, the authors cover a range of dynamic topics
including organizational-level emotional intelligence (OEI) (Chapter 11),
expertise as a firewall for emotion work requirements (Chapter 12),
employee experiences of anger (Chapter 13), dynamics of the flow experi-
ence among academicians (Chapter 14), academic self-control based in per-
sonality systems interaction theory (PSI) (Chapter 15), and the important
role of emotions in the neglected context of volunteerism (Chapter 16).
In the first chapter in this section, Chapter 11, authors Nuno Da
Camara, Victor Dulewicz, and Malcolm Higgs describe a field study where
they examined the effects of a recently developed measure of
organizational-level emotional intelligence (OEI), which they pitch as a cli-
mate construct involving a set of shared norms and practices. Based on sur-
vey data collected from 173 employees of a charitable organization in the
United Kingdom, Da Camira and his colleagues find a relationship
between OEI and intention to quit and show further that this is mediated
by employees’ job satisfaction and affective commitment.
Introduction xxiii

The authors of Chapter 12, Sanjeewa Perera and Carol T. Kulik exam-
ine the emotion work requirements of customer service employees in high
emotion work environments. The study, undertaken in the context of
upmarket hotels, found that novices experienced high emotion workload,
whereas experts experienced low workload. The differentiating factor iden-
tified was that experts had developed additional strategies to the novices,
including following interaction level versus organizational level, rules.
In Chapter 13, authors Kathryn Moura, Ashlea C. Troth and Peter J.
Jordan describe a qualitative study where they investigated employee experi-
ences of anger. In the study, which was conducted immediately prior to
instigation of an anger management intervention program, the authors
interviewed 20 employees with a view to identifying anger triggers and to
see what actions employees were taking to deal with their anger. Results
showed that the principal trigger for interviewees’ anger was a perception of
unfair treatment by management that employees felt powerless to deal with,
resulting in angry outbursts. Employees reported that the only strategy they
had to deal with the anger source was to try to “walk away,” although they
also acknowledged that this provided only temporary respite. Moura and
her colleagues conclude with a discussion of possible remedies that might be
taken to deal with employees’ anger responses to perceptions of injustice.
In Chapter 14, Avina J. Mendonca, Nidhi Mishra, and Sanket S. Dash
examine the dynamics of the flow experience among academicians.
Drawing from a review of the literature on flow, and using a structured
interview method, the authors asked professors about the extent to which
predefined constructs increased their sense of involvement, and the effect of
student characteristics on the duration of feelings. They also asked intervie-
wees about feelings in the research process and how collaboration and
intrinsic and extrinsic motivators influenced flow, as well as querying per-
ceptions of the change in flow over time and its consequences. Based on
their results, the authors conclude that different psychological needs lead to
different flow experiences and that characteristics of the situation (such as
interactions with students or collaborators) affected the experience of flow.
In Chapter 15, authors Jason J. Dahling, Sophie A. Kay, and Nickolas F.
Vargovic outline a weekly diary study of academic self-control based in per-
sonality systems interaction theory (PSI), which includes a dichotomy of an
individual’s action versus state orientation (ASO). In this theory, action-
oriented individuals are more likely than their state-oriented counterparts to
adjust to demanding situations and to self-regulate (i.e., control their
emotions and behaviors). Dahling and his associates hypothesized that
action-oriented individuals’ propensity to self-regulate would be mediated by

their affective responses to events. To test this idea, the authors collected
weekly reports of negative affect and self-regulation from 39 undergraduate
students over the period of one academic semester. Results supported the
expected effect of ASO but, unexpectedly, there was no mediating effect of
trait affect.
Chapter 16 is the last in the volume. In it, the authors Charmine E. J.
Härtel and Jennifer M. O’Connor call attention to the paucity of emotions
research in the context of volunteerism. After reviewing the small number
of studies examining emotions and volunteering, she concludes that the
available evidence indicates that emotions play a particularly central role in
the attraction, retention, and engagement of volunteers. The chapter con-
cludes with a call for research into this important societal role.
In conclusion, the chapters in this volume collectively illustrate a range
of different perspectives, approaches, and opportunities that inform new
ways of studying and understanding emotions in worklife. We hope the
present volume encourages further momentum to inspire scholarly contribu-
tions for advancing theoretical and empirical developments to the field of
research of emotions in organizations. To close this introduction, we take
the opportunity to acknowledge contributions of those involved in enabling
this volume to be produced, including the Emonet conference organizers,
participants, presenters, reviewers, and authors, and the assistants involved
in supporting this volume. We would especially like to express our deep
appreciation to the editorial and production staff at Emerald.

Charmine E. J. Härtel
Wilfred J. Zerbe
Neal M. Ashkanasy


Mayer, J. D., Salovey, P., & Caruso, D. R. (2002). Mayer-Salovey-Caruso emotional intelli-
gence test. North Tonawanda, NY: Multi-Health Systems, Inc.
Parker, S. K. (2014). Beyond motivation: Job and work design for development, health, ambi-
dexterity, and more. Annual Review of Psychology, 65, 661 691.
This page intentionally left blank



Suzanne J. Peterson, Christopher S. Reina,

David A. Waldman and William J. Becker


The application of physiological methods to the study of psychological

phenomena has garnered considerable interest in recent years. These
methods have proved especially useful to the study of emotions, since evi-
dence suggests that validly measuring a person’s emotional state using
traditional, psychometric methods such as surveys or observation is con-
siderably more difficult than once thought. The present chapter reviews
the challenges associated with measuring emotions from a purely psycho-
logical perspective, and suggests that the study of emotions in organiza-
tions can benefit from the use of physiological measurement to
complement traditional assessment methods. We review more established
approaches to physiological measurement, including those related to
hormone secretion, cardiovascular activity, and skin conductance. We
then highlight somewhat more recent attempts to use neurological scan-
ning. A theme of this chapter is that both psychological and physiological
measures are relevant to understanding and assessing emotions in

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 3 27
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011002

organizations. Accordingly, we propose a multi-method approach invol-

ving both types of assessment. Finally, we discuss the practical and
ethical implications of employing various forms of physiological
measurement in the study of emotions, specifically in the context of
Keywords: Emotions; physiology; measurement; neuroscience


From a layperson perspective, emotions should be fairly easy to measure.

Emotions are often seen as synonymous for feelings so simply asking a
person how he or she feels should be adequate. Another common way to
measure the emotions of others is to watch and infer based on how they
behave or act. This includes facial expression, vocal expression and body
language. Increasingly, however, researchers are recognizing that self-
report and observational methods do not capture the full picture of
emotions (Davidson, 2003; Mauss & Robinson, 2009). For instance, an
individual might answer that he or she feels calm, but that person’s heart is
beating rapidly. Someone else might smile when another person walks into
a room, but report not feeling anything. Yet another individual might
report feeling disgust toward another, yet will hug him or her when they
meet, perhaps for appearances sake. In short, traditional psychological
methods are important, yet limited. They are best suited to evaluate an
individual’s subjective experience of emotion or others’ subjective interpre-
tation of behaviors that they believe indicate emotion. What they cannot
ensure is accuracy that is whether individuals are really feeling what they
are reporting or displaying.
The use of physiological methods in the study of emotion at work can
address this limitation by providing a more ecologically valid way of cap-
turing emotions that employees are experiencing. For instance, self-report
measures of emotion focus on how someone reports to feel. Although this
is important, we also know that certain physiological changes occur in our
bodies when we experience emotions (e.g., when we feel anxious our heart
rate increases and sweat glands activate). Furthermore, we may believe that
we are accurately capturing excitement by asking people whether they feel
excited or by asking others if they believe someone is excited based on their
behavior. However, can we be sure that what we see is excitement, rather
than nervousness? Individual differences in physiological activity can allow
Using Physiological Methods to Study Emotions in Organizations 5

us to draw more valid conclusions regarding the differences between these

two related, yet different, emotions. Quite simply, physiological measure-
ment can increase our confidence that someone is truly feeling what he/she
is reporting to feel or what others believe to be observing that one feels.
Second, and perhaps more importantly, physiological methods allow
researchers to answer more complex questions regarding emotions that
should facilitate the development of stronger theories. For instance, are
emotions really all in the head? Psychological theories (e.g., Lazarus, 1991;
Scherer, 1984) suggest that they are, while neurophysiologists suggest that
there is a strong visceral component to emotion (Adolphs, Tranel, &
Damasio, 2003). Another question is whether emotions are always
conscious feeling states? Neurophysiological measures of brain activity such
as EEG and fMRI allow researchers to record rapid, immediate changes in
emotional responses that would otherwise be impossible to assess without
interrupting an individual’s participation in an experiment, or simply
impractical through self-report because these processes are not available in
the conscious mind.
In summary, a complete assessment of emotions should take into
account all levels of analysis, ranging from the feelings and behaviors asso-
ciated with emotion to how they are measured at the physiological and
neural level. Therefore, the purpose of this chapter is to inform researchers
in the applied psychology and organizational behavior fields on how
physiological methods can complement and advance the study of emotions
at work. We are not suggesting a “throw the baby out with the bathwater
approach” when it comes to psychological methods. Rather, we argue that
psychological measurement needs a supporting cast to fully capture or
complement (Becker & Cropanzano, 2010) the complexity associated with
emotion measurement. We also seek to better inform those who may not
be fully aware of the increasing accessibility of physiological methods. In
our experience, many researchers in the organizational sciences assume
these methods to be too complex namely, they require deep-level exper-
tise to understand and operate, invasive to participants, and expensive to
employ. We seek to dispel those largely faulty perceptions to some degree
by highlighting the advent of new methods. Our hope is that this chapter
encourages more widespread use of these methods in organizational
In the section “Psychological Approach to the Study and Measurement
of Emotions,” we describe how we are defining emotion for the purposes of
this chapter. We then provide an overview of the most commonly used
psychology-based measures of emotion and highlight why emotion should

not be studied from a purely psychological perspective. In the section

“Connecting the Physiological and Neurological Measures,” we describe
the most readily available or practical physiological (also known as biologi-
cal) as well as neurological methods, and emphasize how they might help
to overcome some of the problems found in traditional, psychological
methods. In the section “Ethical and Practical Considerations,” we point
out which conceptual areas related to emotions at work can most benefit
from physiological methods. Finally, we conclude with a discussion of the
ethical and practical considerations surrounding the use of these methods
in organizations.



A common challenge facing psychologists engaged in emotion research

relates to the blurred definition between emotion and affect. Although
some researchers have suggested that emotion refers to the unconscious
experience of emotion and affect refers to the conscious experience of emo-
tion (Damasio, 1999), others use the terms interchangeably (Davidson,
2003, Panksepp, 2000). For the purposes of clarity and consistency in this
chapter, we use emotion as an umbrella term for all the behavioral expres-
sive, cognitive, and physiological changes that occur in individuals
(Panksepp, 2000). This includes moods, discrete emotions, and general
affectivity. Our goal is not to make a statement regarding construct defini-
tions in the realm of emotion and affect, but rather to provide an overview
of physiological methods that can be used to study a variety of emotional
and/or affective phenomena in organizations.
In organizational research, self-report and observer methods are by far
the most commonly used methods to capture emotional phenomena at
work. Although they offer researchers important insight into people’s per-
ceptions of emotion and are easy to administer, important limitations
should be recognized. We detail these methods and their associated limita-
tions below.


Self-report methods can take various forms, but generally consist of asking
people to describe the nature of their emotional experience. Participants
Using Physiological Methods to Study Emotions in Organizations 7

may be given a list of emotion terms (e.g., angry, alert, enthusiastic, and so
forth) and are asked to choose the term to best describe their emotional
experiences, rate the intensity of emotion, or state how long that emotion
has been experienced. Alternatively, participants may be provided with
traditional Likert-style questionnaires regarding their emotional states.
Asking participants to evaluate their level of agreement regarding felt emo-
tion (e.g., strongly agree) exemplifies this approach.
Self-report methods of emotions are beneficial and popular primarily
because they are efficient and easy to administer. Yet, several challenges
remain. Self-report methods assume that people are consciously aware of
their emotions and willing to report them. Similarly, self-report methods are
subject to response bias such that several alternatives may bias the indi-
vidual to choose them. Finally, people not accurately report what they feel
due to social desirability concerns (Lopatovska & Arapakis, 2011; Mauss &
Robinson, 2009). In addition, self-report measures are vulnerable to time
effects (Robinson & Clore, 2002). Specifically, reports concerning one’s
past, future, or trait-related experiences (i.e., how one feels “in general”)
have been found to be less valid than reports that assess how someone feels
“right now,” that day, or immediately following an emotional event
(Mitchell, Thompson, Peterson, & Cronk, 1997).


Observer or behavior measures of emotions involve inferring a person’s

emotional state from their vocal characteristics, facial expressions, and
body language (Mauss & Robinson, 2009). For example, when people
cringe, we infer disgust, if they raise their voice and yell we infer anger, and
so forth. These methods are qualitative in nature. They involve directly
watching and coding the facial, vocal, and gesture reactions of people when
they are exposed to emotional stimuli. Like self-report methods, observer
methods are popular because they are relatively unobtrusive. However,
they are limited because how people react behaviorally is highly dependent
on environmental conditions (e.g., light, noise, temperature). For example,
if a person is folding her arms across her chest, is this a sign of boredom,
anger, disgust, or defensive posture, or is she just cold? Moreover, some
responses can be faked (e.g., the quintessential fake smile) and importantly,
the presence of emotional expression cannot be equated with emotions. Just
because one expresses happiness for a friend’s success does not mean that
he/she is not feeling jealous, for example.

In summary, if a study is mainly interested in people’s perceptions and

explanations of felt emotions, self-report may suffice. However, for more
complex questions such as whether someone is really feeling what they
claim to feel or to predict emotional responding, different methods may be
necessary. Moreover, both self-report and observational methods may be
adequate if research questions are focused on identifying specific emotions
(happy vs. sad), capturing the valence of emotions (positive vs. negative),
or interpreting arousal (anger implies high arousal and quiet implies low
arousal). However, for more in-depth questions, more sophisticated metho-
dology that assesses changes in the activity of the body, including the brain,
are required to ensure emotions are being adequately captured.

Physiological Measures of Emotion

Physiological measures relate to automatic responses triggered by the brain

but manifested in the body. Even subtle, implicit emotions send ripples
through the nervous system that prepare the body for action. Some of the
methods described next will be at least passingly familiar to many research-
ers. In fact, it was not that long ago that many of these methods were more
widely used in organizational research than they are today (Austin,
Scherbaum, & Mahlman, 2002). However, physiological measures are on
the cusp of resurgence due to improved technology and a renewed interest
in investigating and understanding the underpinnings of emotion and how
emotions influence behavior (Becker & Menges, 2013). Here, we will dis-
cuss a few of the most promising methods, including: (1) endocrinology,
(2) cardiovascular, and (3) electrodermal measures. Each has been used
recently to investigate emotions in organizationally relevant contexts.

The levels of neurotransmitters such as cortisol, testosterone, and oxytocin
can now be measured by relatively unobtrusive means and these levels have
been shown to be reliable indicators of emotional arousal and response
(Akinola, 2010). Cortisol is widely recognized as an indicator of stress
response and negative emotions. There is a growing understanding of
exactly how cortisol affects emotional and cognitive brain functioning.
Recently, this new understanding has been used to show that, for example,
in demanding situations, elevated cortisol produces short-term benefits to
decision-making (Akinola & Mendes, 2012). Elevated cortisol levels have
also been shown to influence risk preferences of financial traders
Using Physiological Methods to Study Emotions in Organizations 9

(Kandasamy et al., 2014). At the same time, chronic elevation of cortisol

levels has harmful effects on employee health and job performance
(Lundberg, 2005; Melamed et al., 1999). In these studies, cortisol levels are
typically measured before and after an emotion-eliciting event, and the dif-
ference in the levels is operationalized as a measure of the stress experi-
enced during the event.
More recently, researchers have begun to artificially elevate levels of
hormones in attempts to moderate behavior in adaptive ways. In these
studies, hormone levels are manipulated using a nasal mist. For example,
research has shown that elevating oxytocin can lead to increased trust
(Kosfeld, Heinrichs, Zak, Fischbacher, & Fehr, 2005) and conformity to
group norms (Stallen, De Dreu, Shalvi, Smidts, & Sanfey, 2012). Other
studies have shown that elevated oxytocin can have potentially harmful
effects, particularly for out-group members (De Dreu, 2011; Radke &
De Bruijn, 2012). Thus far, oxytocin levels have not been directly investi-
gated in organizational settings. However, even this brief overview suggests
a number of important implications that might be investigated.

Cardiovascular measures represent even older and seemingly pedestrian
methods for examining human physiology that have experienced a theoreti-
cal and practical renaissance. Equipment is now available that allows
researchers to measure heart rate, blood pressure, and even cardiac activity
(EKG) in naturalistic settings. Heart rate measures provide a direct assess-
ment of rate and volume of heart activity, which has been shown to be an
indicator of arousal. A variety of blood pressure measures, including arter-
ial, systolic, and diastolic, can also be produced that can be used alone or
in combination as a measure of autonomic nervous system response. EKG
measures are generally transformed to provide a more readily usable
measure of cardiac reactivity (Akinola, 2010). New analysis methods allow
the combination of measures to identify emotional states and differentiate
between hindrance and threat stress appraisal and response (Tomaka,
Blascovich, Kibler, & Ernst, 1997).
Combining a variety of cardiovascular measures also allows a reason-
able amount of differentiation between discrete emotions (Herrald &
Tomaka, 2002; Kreibig, 2010). These measures have been used in the
laboratory to link discrete emotions with cognitive appraisals during
ongoing real-life emotional episodes (Herrald & Tomaka, 2002).
Cardiovascular measures have been paired with cognitive appraisals to
differentiate between challenge and threat stress that is induced by the

presence of unknown others during the performance of familiar and

unfamiliar tasks (Blascovich, Mendes, Hunter, & Salomon, 1999). More
recently, cardiovascular measures were used to evaluate mentee emotional
and cognitive responses to two different methods of coaching in a simu-
lated organizational setting (Passarelli, 2014). The continued development
of cardiovascular methods is particularly promising because they are rela-
tively simple to administer.

Another type of physiological methods includes those that measure electro-
dermal response. Despite arousing memories of lie detector machines,
modern technology has produced electrodermal equipment that is
portable and highly sensitive. This new equipment has proven extremely
adaptable to the investigation of emotional responses in a variety of
settings. Most emotions produce increased electrodermal activity to sup-
port the action tendencies associated with the particular emotion (Kreibig,
2010; Sequeira, Hot, Silvert, & Delplanque, 2009). Modern electrodermal
equipment provides a measure of change in skin conductance that is extre-
mely sensitive in terms of level and time, yet readily incorporated into data
analyses. Skin response has been used to investigate cognitive and
emotional responses when experiencing and reflecting on negative experi-
ences (Nikula, 1991; Pennebaker, Hughes, & O’Heeron, 1987). Even more
promising, electrodermal methods have been used to investigate implicit
and anticipatory emotional responses that do not enter conscious aware-
ness (Denburg, Recknor, Bechara, & Tranel, 2006). One drawback of these
measures is their lack of discrete emotion specificity (Kreibig, 2010).
Nonetheless, their relative ease of use and adaptability to field applications
make them a valuable tool for investigating general levels of emotional
arousal at the conscious and nonconscious level (Akinola, 2010).

Neurological Measures of Emotion

Neurologically based methods represent a related, yet distinct, family of

methods that has also seen an increase in popularity in the organizational
sciences in recent years (e.g., Becker & Cropanzano, 2010; Hannah,
Balthazard, Waldman, Jennings, & Thatcher, 2013; Reynolds, 2006;
Reynolds, Leavitt, & DeCelles, 2010; Waldman, Balthazard, & Peterson,
2011a), including considerations of how neuroscience could potentially be
applied to the study of emotions in organizations (see Waldman,
Using Physiological Methods to Study Emotions in Organizations 11

Balthazard, & Peterson, 2011b). Nevertheless, the widespread use of neuro-

methods have been relatively limited. We believe that this lack of actual
application is due in part to a lack of awareness regarding the rapidly evol-
ving availability and accessibility of neurologically based methods, as well
as limited understanding of how best to apply these methods to questions
relevant to emotions in organizational life. A complete consideration of
neuroscience applications to the study of emotions in organizations would
need to include theoretical issues pertaining to the neurological location of
brain activity relevant to specific emotions, and how to operationalize that
activity in terms of specific variables. Considerations of these issues can
indeed be found in recent work (e.g., Waldman et al., 2011b; Lindquist,
Wager, Kober, Bliss-Moreau, & Barrett, 2012). Our goal here is to outline
two key issues that are relevant to the neurological assessment of emotions
in organizational research: (1) state versus trait assessment and (2) choice
of a scanning technology.

State versus Trait

The stimulation of emotional states represent a common way of consider-
ing how emotions are evoked in organizational life. For example, negative
feedback from a supervisor could evoke fear and anger from an employee.
This framing of emotions also parallels a common view of the brain as
reflective of dynamic environmental stimuli (Raichle & Snyder, 2007). In a
series of studies by Greene and colleagues (e.g., Greene et al., 2009),
researchers demonstrated how portions of the brain associated with the
ventromedial prefrontal cortex were commonly stimulated when individuals
were shown pictures depicting moral dilemmas. Interestingly, this area of
the brain is also associated with emotional processing, leading Greene et al.
to consider connections between moral reasoning and emotions.
The term mirroring has been used when the stimulation of the brain is
specifically based on an imitation or reflection of another person’s emotions
(Iacoboni, 2009). Thus, groups of neurons in Person A may become acti-
vated on the part of Person B when the latter observes the emotional
experience of the former (e.g., pain, excitement, and so forth). It follows
that neurological mirroring may be largely responsible for empathy
(Decety & Meyer, 2008), as well as such processes as emotional contagion
(Pugh, 2001).
Neuroscience researchers have increasingly recognized that the brain is
not simply in a reactive state based on stimuli that are presented to it.
Rather, it can also be characterized in terms of a resting state that is rela-
tively stable over time (Raichle, 2010; Raichle & Snyder, 2007). Such a

resting state is indicative of what has been termed the intrinsic brain, which
has not been activated through information processing or stimuli
(Waldman et al., 2011b). With that said, this resting state does not imply
an inactive brain. Cacioppo et al. (2003) suggested that the brain in a rest-
ing state is indeed engaged in measurable activities. Further, intrinsic brain
functioning differs across individuals, making the intrinsic brain analogous
to the concept of individual trait differences. Increasingly, organizational
research is showing the relevance of the intrinsic brain in understanding
organizational phenomena and outcomes (Balthazard, Waldman,
Thatcher, & Hannah, 2012; Hannah et al., 2013; Waldman et al., 2011a).
In short, it is possible to conceive of neural assessment of emotional
phenomena in terms of dynamic or changing states, as well as more
trait-like aspects of emotions. For example, using the intrinsic brain
approach, a researcher could attempt to identify areas of the brain asso-
ciated with emotion-based traits (often referred to as affect) that are rela-
tively stable over time, thus reflecting the individual’s enduring personality
or temperament (Izard, 2009). Alternatively, and depending on the research
question, certain stimuli that are designed to elicit discreet emotional reac-
tion could be linked to particular areas of the brain. Together, these neuro-
logically based approaches could help answer questions regarding the role
of emotions in organizational behavior.

Scanning Technologies
But with all of that said, researchers may be unsure about the second key
issue mentioned above, specifically the choice of a scanning technology.
Simply stated, brain scanning forms the foundation of neuroscience assess-
ment methods, much like tools such as surveys form the foundation of
psychometrics. A number of scanning techniques are available. However,
we consider here the two most popular to date in the social sciences:
(1) functional magnetic resonance imaging or fMRI and (2) quantitative
electroencephalogram or qEEG. “fMRI relies on the paramagnetic proper-
ties of oxygenated and deoxygenated hemoglobin to see images of changing
blood flow in the brain associated with neural activity” (Waldman et al.,
2011b, p. 1096). The most common metric produced by fMRI is known as
blood oxygen level dependence (BOLD), which can be readily subjected to
statistical analyses and matched with other types of data, such as those pro-
duced by psychometrics. In addition, colorful images reflecting BOLD
activity, especially when an individual’s brain is stimulated, are commonly
associated with fMRI. For example, an individual could be presented with
different images or sounds, and potentially engage in actions with limited
Using Physiological Methods to Study Emotions in Organizations 13

physical movement (e.g., press a button). The neural result (i.e., BOLD)
can be used to better understand brain structures and processes associated
with perception of stimuli, or action that is taken based on stimuli.
qEEG represents a second neuroimaging technique that has grown in
popularity in recent years (e.g., Hannah et al., 2013; Waldman et al.,
2011a). It uses advanced signal processing techniques to reveal electrical
energy data about the brain through the scalp and skull (Niedermeyer &
Silva, 1995). Two general categories of qEEG variables are commonly
examined across delta (1 4 Hz), theta (4 8 Hz), alpha (8 12 Hz), beta
(12 30 Hz), and gamma (30 100 Hz) frequency bands: (1) power or ampli-
tude measures and (2) network connection measures such as coherence. As
with fMRI, variables formed through qEEG can be used in statistical ana-
lysis programs.
fMRI and qEEG are both conducive to the reactive/reflexive and intrin-
sic approaches discussed above, although fMRI is probably more known
for reactive/reflexive assessment, while qEEG has been applied more to
intrinsic assessment (Waldman et al., 2011b). Table 1 shows a comparison
between fMRI and qEEG techniques in terms of other issues. A potential
advantage of fMRI is its spacial resolution, or precision. However, as dis-
cussed by Balthazard et al. (2012), the precision of fMRI (especially MRI)
may be more important for things such as medical purposes (e.g., location
and nature of tumors), rather than understanding the neural origins of
aspects of organizational behavior. Perhaps more importantly, qEEG is
superior in terms of temporal resolution in that electrical data are recorded
immediately in relation to the concomitant brain activity. In contrast,
BOLD has been shown to represent a delayed response to actual neuronal
activity, often taking several seconds to materialize after that behavior.
Accordingly, it may be challenging to distinguish distinct BOLD data with
regard to events that occur within a short timeframe.
Other advantages are evident for qEEG. As compared to fMRI, qEEG
is much more practical, perhaps especially for studies involving

Table 1. A Comparison of fMRI and qEEG Technologies.


Spatial resolution High degree of resolution Lower degree of resolution

Temporal resolution Lower degree of resolution High degree of resolution
Practicality Restricted use to laboratory Highly portable; can be used in
settings naturalistic settings
Cost-effectiveness Highly expensive Relatively low costs

organizational participants. fMRI is not portable, thus requiring partici-

pants to go where the equipment is located (i.e., a university-based labora-
tory). Moreover, fMRI requires participants to remain highly immobilized
in a confined space (i.e., within a confined tube). In contrast, qEEG is
portable and can be completed while people are comfortably seated and
even engaged in various tasks, such as decision-making and group-based
conversations (e.g., see Waldman et al., 2013).
Finally, and perhaps most importantly, the high cost and lack of
accessibility of fMRI scanners simply makes fMRI a cost-prohibitive
technique for most organizational researchers. Scanners represent multi-
million dollar devices, staffed by highly trained technicians, and are
largely located at medical facilities or clinical research laboratories. Due
in part to high cost and low accessibility, the number of participants in
fMRI studies is characteristically small (i.e., less than 20; see Lieberman,
Berkman, & Wager, 2009). As noted by Lindebaum and Jordan (2014),
such N-sizes are too small for the types of statistical analyses in which
organizational researchers typically engage. In contrast, cost, accessibility,
and capability to process large numbers of research participants make
qEEG much more feasible for organizational researchers. Accordingly,
we suggest that, as compared to fMRI, qEEG applications are better
suited for immediate adoption across a wide range of questions
(e.g., those pertaining to emotions) and settings. While fMRI remains a
powerful tool, its growth in organizational research is likely to progress
more slowly.



We believe that a systematic application of multiple methods, including phy-

siological and brain imaging technologies, will be necessary to more
fully understand emotional phenomena in organizations. When considering
the above methodologies, two key issues come to mind. First, most of the
methods surveyed here are best suited for the assessment of emotional reac-
tions or relatively brief states. Examples include cardiovascular and fMRI-
based assessments to stimuli. But with that said, we have also summarized
how qEEG could be used to assess what has been referred to as the “intrinsic”
brain (Waldman et al., 2011b), which may be able to tap the physiological
basis of more trait-like or enduring emotional characteristics.
Using Physiological Methods to Study Emotions in Organizations 15

Second, some promising technology is emerging that is capable of asses-

sing neurological and other physiological measures simultaneously. For
example, Waldman et al. (2013) used equipment developed by Advanced
Brain Monitoring, Inc. (advancedbrainmonitoring.com) that has simulta-
neous qEEG and qECG (i.e., electrocardiogram) capabilities. In other
words, the brain and heart can be assessed simultaneously in real time.
Waldman et al. (2013) demonstrated how the wireless nature of this tech-
nology could be used to provide physiologically assessment in a natural set-
ting on a millisecond-by-millisecond basis. For example, with this
technology, physiologically based assessment of individual- and team-level
emotions could be made throughout a team problem-solving process.

Specific Applications of Physiological Measurement

The use of physiological and neurological measures have important impli-

cations for advancing the study of three commonly discussed areas within
the emotional realm of management research. Specifically, we discuss emo-
tional regulation, emotional expression, and emotional transfer or conta-
gion as three broad areas that are especially ripe to utilize the alternative
techniques considered above in order to facilitate a deeper level of under-
standing beyond that which psychological measurement can provide.

Emotional Regulation
Emotional regulation refers broadly to individuals’ ability to manage their
expressions and feelings (Grandey, 2000). Emotional labor, defined as the
management of feeling and emotion in order to produce a certain public
display (Hochschild, 1983), represents an often-studied aspect of emotional
regulation. Employees in a broad range of industries ranging from call
centers, to restaurants, to sales organizations are required to interact with
customers in a certain way which has been termed “service with a smile”
(Pugh, 2001). The overt act of smiling to customers, no matter what may
be going on within an employee’s life, is an example of surface acting
defined as regulating an outward display of emotion in order to meet the
expectations of the work environment (Hochschild, 1983). A second type of
emotional labor called deep acting involves employees actually inducing
certain emotions so that they can genuinely experience these emotions
rather than only outwardly simulating the experience of such emotions, as
is the case with surface acting (Hochschild, 1983).

Recent meta-analytic evidence suggests that surface acting, but not deep
acting, exhibited positives relationships with emotional exhaustion, psycho-
logical strain, psychosomatic complaints and negative relationships with
job satisfaction and organizational attachment (Hülsheger & Schewe,
2011). Given the accumulation of research suggesting the negative implica-
tions of employee surface acting, researchers recommend that employees
utilize deep acting instead. However, assessing the extent to which an indi-
vidual is utilizing surface or deep acting emotional labor is difficult to
ascertain via psychological measurement techniques given that individuals
may not be fully aware of which strategy they are using, may exhibit biases
related to social desirability, or may not be able to separate how they feel
internally regarding their emotions from what they are attempting to feel
via emotional regulation. In other words, for any number of reasons,
employees may not be aware of which method of emotional regulation they
are using, and subsequently may be utilizing surface acting, rather than
deep acting. This clearly has important implications for the health and
well-being of employees and highlights the need to have an ecologically
valid way of assessing the underlying emotions that employees are
Better measurement, including physiological measures, may ultimately
help researchers design more effective training in order to help employees
avoid the negative implications associated with emotional labor.
Physiological measures could be utilized to help employees better regulate
their emotions by providing real-time feedback regarding the emotions that
they are physiologically exhibiting, versus the emotions that they are out-
wardly trying to exhibit. In short, self-report measures of emotional regula-
tion may give us a general sense of how successful employees are in
regulating their emotions to match the needs of the situation, but more eco-
logically valid methods will allow researchers to identify when there are
gaps between what employees say and actually feel internally. This will
facilitate an understanding of the trust “cost” of the emotional labor
A second area under the umbrella of emotional regulation is emotional
intelligence. Despite a level of confusion associated with the construct and
the various ways to conceptualize and measure emotional intelligence,
Chernis (2010) suggests that researchers commonly agree on the common
core of emotional intelligence which is “the ability to perceive and express
emotion, assimilate emotion in thought, understand and reason with emo-
tion, and regulate emotion in the self and others” (Mayer, Caruso, &
Salovey, 2000, p. 396). Of the various conceptualizations of emotional
Using Physiological Methods to Study Emotions in Organizations 17

intelligence, the Mayer and Salovey (1997) approach has generally garnered
the most attention due to its conceptualization of emotional intelligence as an
ability and its close alignment with the common core of the construct shared
by emotional intelligence researchers (Ashkanasy & Daus, 2005; Walter,
Cole, & Humphrey, 2011). This approach has also provided researchers with
two methods to measure emotional intelligence: (1) an ability-based test that
captures individual’s performance in solving emotional problems, and (2) self
and/or other-report, perception-based measures that ask individuals to reflect
on their own or others’ levels of emotional intelligence.
While researchers disagree as to the ultimate utility of emotional intelli-
gence (see Antonakis, Ashkanasy, & Dasborough, 2009), the basic idea
that individuals differ in their ability to process emotional information is
simple (Ashkanasy & Humphrey, 2011). Researchers are already beginning
to utilize neuroscience to understand how the brains of individuals who
exhibit high emotional intelligence differ from those individuals who exhi-
bit lower emotional intelligence (see Timoshanko, Desmond, Camfield,
Downey, & Stough, 2014). We suggest that diverse measures in the realm
of physiology or neurology will allow researchers to ask and answer
research questions to further unlock our understanding of emotional intelli-
gence. For example, neurological methods may help to shed light on the
debate between researchers, on one side advocating that general intelligence
is the only form of intelligence that matters, versus other researchers who
suggest that there are multiple types of intelligence (see Antonakis et al.,
2009 for a debate).
Studying emotional intelligence with neuroscience methodologies will
further clarify which of the current models most closely represents how
emotional intelligence functions in relation to other personality variables
and which models most accurately account for differences in performance
on emotional tasks. The stream of emotional intelligence research that is
commonly referred to as trait emotional intelligence (Petrides & Furnham,
2003) has been criticized for encompassing everything except cognitive abil-
ity (Ferris, Perrewé, & Douglas, 2002). A neuroscientific approach to
studying emotional intelligence could help reveal whether this is the case by
parsing out the variance in outcomes accounted for by various individual
differences such as personality. In sum, we suggest that alternative
approaches to studying emotional intelligence will advance research in the
management field by more clearly defining the construct and determining
the similarities and differences in the various conceptualizations of
the construct so that we can more accurately understand how emotional
intelligence functions.

Emotional Expression
We suggest that physiological measurement would also be useful to
further our understanding of the expression of emotions in the workplace.
Emotional expression is considered a subcomponent of emotion regulation
and refers to the observable verbal and nonverbal behaviors that commu-
nicate emotional states (Gross, 1999). Examples are smiling or crying.
Emotional expression is especially important because of the social nature
of work and given the common knowledge that employees leave man-
agers/leaders rather than companies. Ashkanasy and Humphrey (2011)
provide a useful framework to understand how the expression of emo-
tional phenomena in the workplace affects others. They suggest that indi-
vidual’s moods, emotions, attitudes, emotional traits and states such as
neuroticism and emotional intelligence affect interpersonal relationships
and trust via leadership, and ultimately affect the emotional climate and
performance of an organization. We suggest that physiological methods
would be especially useful to understand the origins of these traits and
states in order to help leaders be more in tune with the emotions that they
are feeling inside and how these are linked to the behaviors they exhibit,
which ultimately impact how they are perceived by others. For example,
leaders who may come off cold or not empathetic to followers might lead
a team that reports low engagement or motivation. Alternatively, leaders
who overreact in fits of rage might be able to gain insight about how they
process information through “deeper” emotion-based measures that go
beyond self-report.
In short, using physiological methods, researchers will be better able
to understand how the brain and body respond to various situations at work
and better understand the link between what goes on inside an individual
and how they outwardly behave. Researchers can use physiological data
such as heart rate, hormone secretion, blood pressure, skin conductance, and
qEEG brain data to triangulate the causes of leader behavior and in turn,
train leaders to be more aware of these physiological changes and how they
relate to their behaviors. Recent work suggests the strong link between our
neuroanatomy, our behaviors, and our interpersonal relationships (see
Cozolino, 2014). We suggest that a broad range of physiological methods
can similarly help researchers understand the linkages between what goes on
inside us with what goes on between us in the workplace. Management topics
such as aggressive and abusive leader behaviors toward employees, are ideal
areas to seek a more complete understanding of emotional expression, given
the harmful effects these negative behaviors have in the workplace (see
Tepper, 2007 for a review).
Using Physiological Methods to Study Emotions in Organizations 19

Emotional Contagion
Emotional contagion, or the idea that individuals “catch” the emotions
and moods from those with whom they interact (Hatfield, Cacioppo, &
Rapson, 1994), represents a third and highly interrelated area that has
already begun to benefit from the use of physiological methods (Decety &
Meyer, 2008; Iacoboni, 2009). Organizational scholars have long recog-
nized that emotions and moods seem to spread between individuals, almost
as if one individual’s emotional system copies that of another individual via
social learning (Bandura, 1971). More recently, neuroscientists have studied
the neurological foundations of emotional contagion, suggesting that this
copying behavior actually begins in the brain when an individual’s neurons
copy or mirror the same neurological activity patterns of another indivi-
dual who is observed or with whom the individual comes into contact
(Gallese, Fadiga, Fogassi, & Rizzolatti, 1996). These neurons fire in such a
way that even though an individual is not engaging in the same behavior
pattern as the person who is observed, the individual’s mirror neurons are
in essence “playing along.” This explains how followers may ultimately end
up feeling the same emotion or mood as their leader despite not being con-
sciously aware of this occurring. Using fMRI technology, researchers have
supported this process of emotional transfer through the replication of
others’ facial expressions and subsequent activation of the limbic system to
induce an emotion that is consistent with those facial expressions (Carr,
Iacoboni, Dubeau, Mazziotta, & Lenzi, 2003).
Although mirror neurons reproduce similar patterns of neuronal firing
in the brain of one individual in the presence of another, it is important
and interesting to note that the patterns are not identical. Researchers have
demonstrated that mirror neurons fire in accordance with the intention of
the action rather than simply the action by itself, which suggests that mir-
ror neuron firing is a complex process that takes a lot of information into
account (Fogassi et al., 2005). This is important because it provides an
explanation for how emotions and moods spread between individuals, not
just based on what is observed between two individuals, but also based on
the interpretations and attributions individuals make regarding the intention
or purpose of another’s actions. Thus, observing high levels of positive
emotion will serve as one input into the emotional contagion process at the
neurological level, but so will the assumed intention of why this high level
of positive emotion exists.
Organizational research on topics such as leader and follower relation-
ship quality, team composition, vision articulation and vision sharing,
organizational culture, helping behaviors, and perceptions of justice and

trust only begin to scratch the surface of the many topics that could benefit
from a more in-depth understanding of how emotions and moods become
shared between individuals and how this complex process unfolds.
Physiological methods hold much promise in allowing researchers to
further understand how the various levels of emotions articulated by
Ashkanasy and Humphrey (2011) reinforce each other and cascade from
one level to another, and in which directions.
While physiological and neurological measures have important implica-
tions for furthering our understanding of many different topics within the
management literature, we have discussed three particular areas in which
these measurement techniques may especially hold promise in allowing us
to better understanding emotional phenomena in the workplace. Emotional
regulation, emotional expression, and emotional transfer via contagion are
important and interrelated topics that provide an important emotional
base from which other management topics build, and we suggest that these
three areas can greatly benefit from increased attention via physiological/
neurological measurement perspectives.


Despite the positives to be garnered from the use of physiological methods

to advance the study of emotions at work, researchers must account for
certain ethical and practical concerns before considering their use. Similar
to psychological methods, researchers employing physiological methods in
organizational research must address concerns related to privacy and confi-
dentiality. However, because of the biological and even medical nature of
the data, these concerns may be exacerbated. Employees will naturally be
concerned with how their data are handled, stored, and used. For example,
although researchers may be interested in measuring physiology only
to assess employees’ emotional states, what if employees’ physiology
(e.g., heart rate) is symptomatic of a medical issue (e.g., serious cardiovas-
cular issue)? How should researchers handle this information with employ-
ees as well as employers? Although researchers can certainly guarantee
confidentiality of the data, what moral or legal obligations do they have to
report health problems and what implications might these problems have
on employment decisions or healthcare claims?
Another potential limitation relates to the invasiveness of the methods.
In order to gather physiological or neurological data, employees need to
Using Physiological Methods to Study Emotions in Organizations 21

wear or be attached to specialized equipment. Moreover, they may be

required to come to a laboratory or undergo experiments. Whether organi-
zations and their employees will be willing to accommodate and endorse
these methods remains to be seen. There are practical concerns related to
the disruption of work. For example, will employees be willing to wear
heart rate monitors during meetings or electrodes while operating machin-
ery? Employers would have to weigh any requests that could interfere with
employee productivity or motivation before advocating participation to
their employees. Moreover, human resource and legal departments may be
naturally more wary to approve projects that they believe are either inva-
sive to employee privacy or that have medical implications. Such concerns
can surely be overcome, but will take patience, candor, and trust between
researchers and organizations to ensure employee safety and confidential-
ity, as well as to convince organizations of the benefit of physiological-
based research. The good news is that the methods described in this chapter
are increasingly becoming less invasive, thereby allowing employees to
wear equipment that is virtually undetectable. For example, wireless heart
rate monitors are now available that can clip to the earlobe or finger, and
wearable monitors that measure cardiovascular resistance can be worn
under the participants’ normal clothing (Pantelopoulos & Bourbakis,
From the perspective of researchers, there are other hurdles to consider.
First, attempts to link physiology and emotions are inherently interdisci-
plinary in nature. It may simply not be practical for researchers who are
specialized in the organizational sciences to feel confident that they have
the necessary technical expertise or knowledge of the associated literature
to effectively utilize these methods (Waldman et al., 2011b). Applying a
physiological or neurological lens to the study of emotions in organizations
will require a diverse research team. Such collaborations can be cumber-
some given the diversity of training and academic backgrounds.
Furthermore, even when interdisciplinary teams are successfully formed,
the product is not always rewarded or encouraged.
Second, as noted by Waldman et al. (2011b), academic departments with
stringent tenure requirements are often resistant to rewarding publications
in journals (even premier ones) if they are not in their specific discipline.
Similarly, journals encourage interdisciplinary work on the one hand yet
do not often publish it on the other. This is due in part to the difficulty of
finding reviewers qualified to evaluate the work. For example, it is difficult
for editors of a management journal to have a neurophysiologist at their
disposal. Without such an expert, editors cannot feel confident that the

work is of the caliber to warrant publication. Although these hurdles are

somewhat understandable, the bottom line is that such behavior dis-
courages faculty (especially untenured faculty) from engaging in interdisci-
plinary research topics and methods. The hope is that over time, journals
in the organizational sciences will continue to build a diverse reviewer base
and will publish more of this type of work. Special issues of journals are a
common way to begin this journey.
Another researcher concern relates to the resource requirements of
this type of research. Researchers in the organizational sciences are used
to the relatively minimal expenses involved in survey research (Waldman
et al., 2011b). However, as discussed earlier, technologies representative
of the physiological or neurological methods typically carry a higher
financial burden. Although such methods are less costly than they once
were, they still require substantially more investment than traditional
psychometric methods. Moreover, much of this work requires the
presence of a lab, which carries another financial cost. Research that
carries a larger cost is often dependent on external grant money.
However, researchers in the organizational sciences are not as accus-
tomed to writing and submitting proposals and winning the grant, at
least when compared to those in the hard sciences. Therefore, the devel-
opment of strong research networks are particularly important in this
area of research so that labs and other such resources can be shared
across research teams. The result is that publications of this type may
have large author groups ranging from five to ten authors. This is similar
to what we see in the medical fields.
Despite these obstacles, they are not insurmountable by any means.
Researchers with the interest and expertise are able to use physiological
and neurological methods to study emotions with greater ease than ever
before. As more researchers join this arena, surely the path will become
easier for others.

The goal of this chapter was to provide researchers with an introduction to
the most commonly used physiological methods that can be used to study
emotion in organizations. We argue that the study of emotion can best be
moved forward by the use of diverse and multiple methods. As such, we
advocated for the use of physiological methods as a complement (rather
Using Physiological Methods to Study Emotions in Organizations 23

than replacement) to traditional, psychological methods. All the reviewed

methods have pros and cons, and should be chosen based on the study
goals. For example, studies that are interested in employee perceptions and
explanations of felt emotions should utilize self-report methods. Studies
that are interested in how individuals react to emotional stimuli from a
systems perspective should consider physiological methods. Moreover, phy-
siological and neurological measures could be used in conjunction with
self-report or observation measures to ensure the accuracy and reliability,
and create a more complete picture of employees’ emotional states and

Adolphs, R., Tranel, D., & Damasio, A. R. (2003). Dissociable neural systems for recognizing
emotions. Brain and Cognition, 52, 61 69.
Akinola, M. (2010). Measuring the pulse of an organization: Integrating physiological mea-
sures into the organizational scholar’s toolbox. Research in Organizational Behavior, 30,
203 223.
Akinola, M., & Mendes, W. B. (2012). Stress-induced cortisol facilitates threat-related decision
making among police officers. Behavioral Neuroscience, 126(1), 167 174.
Antonakis, J., Ashkanasy, N. M., & Dasborough, M. T. (2009). Does leadership need emo-
tional intelligence? The Leadership Quarterly, 20, 247 261.
Ashkanasy, N. M., & Daus, C. S. (2005). Rumors of the death of emotional intelligence in
organizational behavior are vastly exaggerated. Journal of Organizational Behavior, 26,
441 452.
Ashkanasy, N. M., & Humphrey, R. H. (2011). Current emotion research in organizational
behavior. Emotion Review, 3, 214 224.
Austin, J. T., Scherbaum, C. A., & Mahlman, R. A. (2002). History of research methods in
industrial and organizational psychology: Measurement, design, analysis. In S. G.
Rogelberg (Ed.), Handbook of research methods in industrial and organizational
psychology (pp. 3 33). Oxford: Blackwell.
Balthazard, P., Waldman, D. A., Thatcher, R. W., & Hannah, S. T. (2012). Differentiating
transformational and non-transformational leaders on the basis of neurological ima-
ging. The Leadership Quarterly, 23, 244 258.
Bandura, A. (1971). Social learning theory. New York, NY: General Learning Press.
Becker, W. J., & Cropanzano, R. (2010). Organizational neuroscience: The promise and pro-
spects of an emerging discipline. Journal of Organizational Behavior, 31, 1055 1059.
Becker, W. J., & Menges, J. I. (2013). Biological implicit measures in HRM and OB: A ques-
tion of how not if. Human Resource Management Review, 23, 219 228.
Blascovich, J., Mendes, W. B., Hunter, S. B., & Salomon, K. (1999). Social “facilitation” as
challenge and threat. Journal of Personality and Social Psychology, 77(1), 68 77.
Cacioppo, J. T., Berntson, G. G., Lorig, T. S., Norris, C. J., Rickett, E., & Nusbaum, H.
(2003). Just because you’re imaging the brain doesn’t mean you can stop using your

head: A primer and set of first principles. Journal of Personality and Social Psychology,
85, 650 661.
Carr, L., Iacoboni, M., Dubeau, M. C., Mazziotta, J. C., & Lenzi, G. L. (2003). Neural
mechanisms of empathy in humans: A relay from neural systems for imitation to limbic
areas. Proceedings of the National Academy of Sciences, 100, 5497 5502.
Chernis, C. (2010). Emotional intelligence: Toward clarification of a concept. Industrial and
Organizational Psychology, 3, 110 126.
Cozolino, L. (2014). The neuroscience of human relationships: Attachment and the developing
social brain (2nd ed.). New York, NY: W. W. Norton and Company.
Damasio, A. (1999). The feeling of what happens: Body emotion and the making of conscious-
ness. London: Vintage.
Davidson, R. J. (2003). Seven sins in the study of emotion: Correctives from affective neu-
roscience. Brain and Cognition, 52, 129 132.
De Dreu, C. K. W. (2011). Oxytocin modulates cooperation within and competition between
groups: An integrative review and research agenda. Hormones and Behavior, 61(3),
419 428.
Decety, J., & Meyer, M. (2008). From emotion resonance to empathic understanding: A social
developmental neuroscience account. Development and Psychopathology, 20,
1053 1080.
Denburg, N. L., Recknor, E. C., Bechara, A., & Tranel, D. (2006). Psychophysiological antici-
pation of positive outcomes promotes advantageous decision-making in older persons.
International Journal of Psychophysiology, 61(1), 19 25.
Ferris, G. R., Perrewé, P. L., & Douglas, C. (2002). Social effectiveness in organizations:
Construct validity and research directions. Journal of Leadership & Organizational
Studies, 9, 49 63.
Fogassi, L., Ferrari, P. F., Gesierich, B., Rozzi, S., Chersi, F., & Rizzolatti, G. (2005). Parietal
lobe: From action organization to intention understanding. Science, 308, 662 667.
Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action recognition in the premo-
tor cortex. Brain, 119(Pt. 2), 593 609.
Grandey, A. A. (2000). Emotional regulation in the workplace: A new way to conceptualize
emotional labor. Journal of Occupational Health Psychology, 5, 95 110.
Greene, J. D., Cushman, F. A., Stewart, L. E., Lowenberg, K., Nystrom, L. E., & Cohen, J. D.
(2009). Pushing moral buttons: The interaction between personal force and intention in
moral judgment. Cognition, 111, 364 371.
Gross, J. J. (1999). Emotion regulation: Past, present, future. Cognition and Emotion, 13,
551 573. doi:10.1080/026999399379186
Hannah, S. T., Balthazard, P. A., Waldman, D. A., Jennings, P., & Thatcher, R. (2013). The
psychological and neurological bases of leader self-complexity and effects on adaptive
decision-making. Journal of Applied Psychology, 98, 393 411.
Hatfield, E., Cacioppo, J., & Rapson, R. L. (1994). Emotional contagion. Paris: Cambridge
University Press.
Herrald, M. M., & Tomaka, J. (2002). Patterns of emotion-specific appraisal, coping, and car-
diovascular reactivity during an ongoing emotional episode. Journal of Personality and
Social Psychology, 83(2), 434.
Hochschild, A. R. (1983). The managed heart: Commercialization of human feeling. Berkeley,
CA: University of California Press.
Using Physiological Methods to Study Emotions in Organizations 25

Hülsheger, U. R., & Schewe, A. F. (2011). On the costs and benefits of emotional labor: A
meta-analysis of three decades of research. Journal of Occupational Health Psychology,
16(3), 361 389.
Iacoboni, M. (2009). Imitation, empathy, and mirror neurons. Annual Review of Psychology,
60, 653 670.
Izard, C. E. (2009). Emotion theory and research: Highlights, unanswered questions, and
emerging issues. Annual Review of Psychology, 60, 1 25.
Kandasamy, N., Hardy, B., Page, L., Schaffner, M., Graggaber, J., Powlson, A. S., &
Coates, J. (2014). Cortisol shifts financial risk preferences. Proceedings of the National
Academy of Sciences, 111(9), 3608 3613.
Kosfeld, M., Heinrichs, M., Zak, P. J., Fischbacher, U., & Fehr, E. (2005). Oxytocin increases
trust in humans. Nature, 435(7042), 673 676.
Kreibig, S. D. (2010). Autonomic nervous system activity in emotion: A review. Biological
Psychology, 84(3), 394 421.
Lazarus, R. S. (1991). Emotion and adaptation. New York, NY: Oxford University Press.
Lieberman, M. D., Berkman, E. T., & Wager, T. D. (2009). Correlations in social neuroscience
aren’t voodoo. Perspectives in Psychological Science, 4, 300 310.
Lindebaum, D., & Jordan, P. J. (2014). A critique on neuroscientific methodologies in organi-
zational behavior and management studies. Journal of Organizational Behavior, 35,
898 908.
Lindquist, K. A., Wager, T. D., Kober, H., Bliss-Moreau, E., & Barrett, L. F. (2012). The
brain basis of emotion: A meta-analytic review. Behavioral and Brain Sciences, 35,
121 202.
Lopatovska, I., & Arapakis, I. (2011). Theories, methods and current research on emotions in
library and information science, information retrieval and human-computer
Interaction. Information Processing and Management. doi:10.1016/j.ipm.2010.09.001
Lundberg, U. (2005). Stress hormones in health and illness: The roles of work and gender.
Psychoneuroendocrinology, 30(10), 1017 1021.
Mauss, I. B., & Robinson, M. D. (2009). Measures of emotion: A review. Cognition and
Emotion, 23, 209 237.
Mayer, J. D., Caruso, D. R., & Salovey, P. (2000). Selecting a measure of emotional
intelligence: The case for ability scales. In R. Bar-On & J. D. A. Parker (Eds.),
The handbook of emotional intelligence (pp. 320 342). New York, NY: Jossey-Bass.
Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey &
D. J. Sluyter (Eds.), Emotional development and emotional intelligence (pp. 267 298).
New York, NY: Basic Books.
Melamed, S., Ugarten, U., Shirom, A., Kahana, L., Lerman, Y., & Froom, P. (1999). Chronic
burnout, somatic arousal and elevated salivary cortisol levels. Journal of Psychosomatic
Research, 46(6), 591 598.
Mitchell, T. R., Thompson, L., Peterson, E., & Cronk, R. (1997). Temporal adjustments in the
evaluation of events: The “rosy view”. Journal of Experimental Social Psychology,
33(4), 421 448.
Niedermeyer, E., & Silva, F. L. (1995). Electroencephalography: Basic principles, clinical appli-
cations and related fields. Baltimore: Williams & Wilkin.
Nikula, R. (1991). Psychological correlates of nonspecific skin conductance responses.
Psychophysiology, 28(1), 86 90.

Panksepp, J. (2000). Affective consciousness and the instinctual motor system: The neural sources
of sadness and joy. The caldron of consciousness: Motivation, affect and self-organization,
advances in consciousness research. Amsterdam: John Benjamins Pub. Co.
Pantelopoulos, A., & Bourbakis, N. G. (2010). A survey on wearable sensor-based systems for
health monitoring and prognosis. Systems, Man, and Cybernetics, Part C: IEEE
Transactions on Applications and Reviews, 40, 1 12.
Passarelli, A. M. (2014). The heart of helping: Psychological and physiological effects of con-
trasting coaching interactions. Doctoral dissertation, Case Western Reserve University,
Cleveland, OH.
Pennebaker, J. W., Hughes, C. F., & O’Heeron, R. C. (1987). The psychophysiology of confes-
sion: Linking inhibitory and psychosomatic processes. Journal of Personality and Social
Psychology, 52(4), 781 793.
Petrides, K. V., & Furnham, A. (2003). Trait emotional intelligence: Behavioural validation in
two studies of emotion recognition and reactivity to mood induction. European Journal
of Personality, 17, 39 57.
Pugh, D. S. (2001). Service with a smile: Emotional contagion in the service encounter.
Academy of Management Journal, 44, 1018 1027.
Radke, S., & De Bruijn, E. R. A. (2012). The other side of the coin: Oxytocin decreases the
adherence to fairness norms. Frontiers in Human Neuroscience, 6, 193.
Raichle, M. E. (2010). Two views of brain function. Trends in Cognitive Sciences, 14, 180 190.
Raichle, M. E., & Snyder, A. Z. (2007). A default mode of brain function: A brief history of
an evolving idea. Neuroimage, 37, 1083 1090.
Reynolds, S. J. (2006). A neurocognitive model of the ethical decision-making process:
Implications for study and practice. Journal of Applied Psychology, 91, 737 748.
Reynolds, S. J., Leavitt, K., & DeCelles, K. (2010). Automatic ethics: The effects of implicit
assumptions and contextual cues on moral behavior. Journal of Applied Psychology, 95,
752 760.
Robinson, M. D., & Clore, G. L. (2002). Episodic and semantic knowledge in emotional self-
report: Evidence for two judgment processes. Journal of Personality and Social
Psychology, 83(1), 198 215.
Scherer, K. R. (1984). Emotion as a multicomponent process: A model and some cross cultural
data. Review of Personality and Social Psychology, 5, 37 63.
Sequeira, H., Hot, P., Silvert, L., & Delplanque, S. (2009). Electrical autonomic correlates of
emotion. International Journal of Psychophysiology, 71(1), 50 56.
Stallen, M., de Dreu, C. K. W., Shalvi, S., Smidts, A., & Sanfey, A. G. (2012). The herding
hormone oxytocin stimulates in-group conformity. Psychological Science, 23(11),
1288 1292.
Tepper, B. J. (2007). Abusive supervision in work organizations: Review, synthesis, and
research agenda. Journal of Management, 33(3), 261 289.
Timoshanko, A., Desmond, P., Camfield, D. A., Downey, L. A., & Stough, C. (2014).
A magnetic resonance spectroscopy (1H MRS) investigation into brain metabolite
correlates of ability emotional intelligence. Personality and Individual Differences, 65,
69 74.
Tomaka, J., Blascovich, J., Kibler, J., & Ernst, J. M. (1997). Cognitive and physiological ante-
cedents of threat and challenge appraisal. Journal of Personality and Social Psychology,
73(1), 63 72.
Using Physiological Methods to Study Emotions in Organizations 27

Waldman, D. A., Balthazard, P. A., & Peterson, S. J. (2011a). The neuroscience of leadership:
Can we revolutionize the way that leaders are identified and developed? Academy of
Management Perspectives, 25, 60 74.
Waldman, D. A., Balthazard, P., & Peterson, S. J. (2011b). Social cognitive neuroscience and
leadership. The Leadership Quarterly, 22, 1092 1106.
Waldman, D. A., Wang, D., Stikic, M., Berka, C., Balthazard, P. A., Richardson, T., …
Maak, T. (2013). Emergent leadership and team engagement: An application of
neuroscience technology and methods. Academy of Management Best Paper
Walter, F., Cole, M. S., & Humphrey, R. H. (2011). Emotional intelligence: Sine qua non of
leadership or folderol? Academy of Management Perspectives, 25, 45 59.
This page intentionally left blank


Michael J. Gill


This chapter outlines the potential of phenomenology to illuminate how

individuals experience the emotions replete within organizations. It
employs one particular type of phenomenological approach known as
Interpretative Phenomenological Analysis (IPA). The chapter considers
how the hermeneutic and phenomenological foundations of this approach
lend themselves to the study of affect. The chapter then clarifies and
develops established IPA guidelines to render them more appropriate for
research on emotions. In doing so, the chapter demonstrates how IPA
can produce contextualized accounts that explore the role of emotions in
individuals’ experiences of organizational events and processes.
Keywords: Emotion; interpretative phenomenological analysis;
phenomenology; hermeneutics; qualitative research

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 29 50
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011003


Phenomenology is a broad term that encapsulates both a philosophical

movement and a variety of qualitative research approaches that examine
individuals’ subjective experiences. Phenomenological philosophers have
emphasized that emotions are intimately involved in individuals’ experi-
ences of the world as they shape and are shaped by individuals’ self-
consciousness (Solomon, 2006). Such a perspective has led some scholars to
argue that “phenomenology has done more than any other school of
thought for bringing emotions to the forefront of philosophical enquiry”
(Hatzimoysis, 2010, p. 215). The prominence of emotions in phenomenolo-
gical thought may help to explain why phenomenologically inspired con-
cepts in organizational studies like sense-making acknowledge explicitly the
role of emotion (Weick, Sutcliffe, & Obstfeld, 2005).
Proponents of phenomenological philosophy have called for application
of phenomenology to the study of emotion (Solomon, 2006). At the same
time, many scholars have called for the investigation of emotional experi-
ences and their integration into organizational theories (e.g., Voronov,
2014; Voronov & Vince, 2012). However, whilst phenomenological research
approaches have become increasingly popular across a range of social
sciences (Smith, 2011), there is little specific guidance for researchers seek-
ing to employ phenomenology to analyze, describe or examine the role of
emotions. As such, many scholars remain uncertain how a phenomenologi-
cal approach could illuminate the emotional experiences replete within
The objective of this chapter is to explore how researchers could employ
one phenomenological approach to examine the role of emotions in organi-
zations. Given that a variety of phenomenological methodologies exist,
each with their own aims and assumptions, it is important to explain the
selection of one particular type (see Gill, 2014). This chapter considers the
approach known as Interpretative Phenomenological Analysis (IPA).1 It
does so for two reasons. First, a core assumption of IPA is that “emotions
are absolutely central to our human understanding of experience” (Smith,
Flowers, & Larkin, 2009, p. 199) and this approach is therefore attuned to
the examination of emotions. Second, although IPA studies have only
recently begun to emerge within organizational settings, they have demon-
strated the potential of IPA to provide new insights into organizational
processes (Cope, 2011; Fitzgerald & Howe-Walsh, 2008; Gill, 2015;
Murtagh, Lopes, & Lyons, 2011; Rehman & Roomi, 2012; Tomkins &
Eatough, 2014; Wise & Millward, 2005). As such, this chapter considers
A Phenomenology of Feeling 31

how an IPA approach could be used to examine the experience of emotions

in organizations, defining an emotion as subjective and transient feeling
state (Ashforth & Humphrey, 1995; Fineman & Sturdy, 1999).
Five sections structure this chapter. The section “Foundations of
Interpretative Phenomenological Analysis” reviews the phenomenological
and hermeneutic foundations of IPA that inform its approach to the study
of emotions. The section “Recommendations for Employing IPA to
Examine Emotions” builds on and develops established IPA guidelines for
researchers, to render them more appropriate for inquiry into emotions
within organizations. The section “Comparison with Other Qualitative
Approaches” contrasts IPA with several other qualitative research
approaches to highlight its distinctive features. The section “The Utility of
IPA Studies Within Organizational Research” discusses IPA’s potential to
illuminate the shades of emotions intertwined in individuals’ experiences,
explicate the role of emotions in social phenomena and enrich organiza-
tional theory across a range of subject areas. The section “Limitations”
considers the criticisms and limitations of the approach before concluding.


Initially conceived by Jonathan Smith and applied to health psychology

(Smith, 1996), IPA is a qualitative research methodology that examines
how individuals understand and make sense of particular experiences. IPA
is a perspective from which to approach qualitative research as opposed to
a distinct “method” (Larkin, Watts, & Clifton, 2006). This approach has
become popular in the field of psychology and its influence continues to
grow in other disciplines including education and nursing. Whilst few orga-
nizational research textbooks mention IPA and those that do make only
brief references (Cassell & Symon, 2004), there is a wealth of guidance for
scholars who wish to use this approach (Chapman & Smith, 2002; Reid,
Flowers, & Larkin, 2005; Smith & Osborn, 2003). IPA is now widely taught
across a range of social science programs, with its proponents producing
dedicated textbooks (Smith et al., 2009) and hundreds of published empiri-
cal journal chapters (Smith, 2011). As such, this chapter does not seek to
repeat the many other publications that detail the aspects and procedures
of IPA. Instead, it aims to build on these guides by considering how the
IPA research process could be used specifically to examine the experience

of emotions in organizations. To do so, it is important to consider briefly

IPA’s foundations in the traditions of phenomenology and hermeneutics
that inform its approach to the study of emotions.
IPA rests on the principle that individuals subjectively experience the
social world. This notion can be traced back to the work of Edmund
Husserl, the founder of phenomenological philosophy (Wrathall, 2007).
Husserl stressed that phenomenology is a descriptive endeavor that aims to
capture the subjective perspective of the individual. The term phenomenol-
ogy refers to the description of phenomena, where a “phenomenon” is any-
thing that appears to someone in their consciousness (Moran, 2000).
Husserl posited that any experience involves emotions, as any “thing”
appears from the very beginning as likable or not, useful or not, pleasur-
able or not (Drummond, 2006). Although IPA departs from much of
Husserl’s work in favor of subsequent phenomenologists’ ideas, particularly
those of Martin Heidegger, it shares his emphasis on understanding an
individual’s subjective experience through their own account. It also shares
his belief that emotions are an intrinsic aspect of experience.
The phenomenological philosophy of Martin Heidegger builds on the
work of Husserl but emphasizes the role of context and relationships in
shaping interpretations. This is fundamental to IPA, which “concurs with
Heidegger that phenomenological inquiry is from the outset an interpreta-
tive process” (Smith et al., 2009, p. 32). Heidegger disputed Husserl’s trans-
cendental phenomenology and its attempts to reduce the understanding of
phenomena to the essential elements that transcend a particular context
(Stapleton, 1983). In contrast, Heidegger argued that meaning operates in a
referential totality, or the “historically learned practices and background
understandings we have of the world as a holistic web of interrelated
things” (Berglund, 2007, p. 79). That is, any understanding is always in
relation to something else and depends on context. Heidegger, together
with his student Hans-Georg Gadamer, fused hermeneutics and phenomen-
ology to develop the hermeneutic phenomenology perspective. They argued
that interpretation is not an additional or isolated activity but, instead, the
basic structure of experience (Moustakas, 1994). An individual’s experience
is already an interpretation that rests upon pre-conceived ideas or preju-
dices (Heidegger, 1996). Prejudice, in this sense however, is not necessarily
negative as it enables individuals to understand their experiences.
Nonetheless, detached or objective interpretation is limited, as individuals
cannot fully deny their own context, history, or moods.
Heidegger, more so than Husserl (Solomon, 2006), asserts that our emo-
tions and moods are involved in our perceptual and cognitive engagement
A Phenomenology of Feeling 33

with the world, present in even our most abstract intellectual endeavors
(Hatzimoysis, 2010).2 Indeed, in his seminal text, Being and Time,
Heidegger criticized prior philosophers for their neglect of emotion, sug-
gesting there has been little progress since the work of Aristotle (Heidegger,
1996). Heidegger (1996, p. 131) charged many philosophers as believing
that “affects and feelings fall thematically under the psychic phenomena
[…] They sink to the level of accompanying phenomena.” Heidegger is sug-
gesting that philosophers have tended to consider emotions as mere
responses or as peripheral to an understanding of how individuals engage
with the world (Ratcliffe, 2002). In contrast, Heidegger (1996, p. 131)
argues that emotions help individuals to make meaning and understand
their engagement with the world. As Heidegger (1996, p. 143) noted
“understanding is always attuned,” meaning that the emotions or feelings
of an interpreter inform their interpretations.
By drawing on the ideas of hermeneutic phenomenology, IPA views
individuals as always emotionally engaged with the world and interpreting
their experiences through their unique historical and social context. Thus,
IPA is an interpretive approach to social science committed to learning
about particular individuals in context and examining how they interpret
and understand their experiences.


Although phenomenological philosophers emphasized the importance of
emotions, they rarely offered instruction for the application of their ideas
to research. Furthermore, approaches like IPA, which have begun to trans-
late philosophical phenomenology into practical research approaches,
rarely specify how to examine emotions. As such, this section builds on the
existing IPA guidelines to develop recommendations for the investigation
of emotions. These recommendations consider the research process from
the development of research questions, sampling, and collection of data
through to the analysis of data and evaluation of quality. This section
develops these recommendations by revisiting several phenomenological
and hermeneutic concepts, which underpin IPA, to elicit their implications
for the study of emotions. These hermeneutic or phenomenological con-
cepts include; intentionality, Dasein, the life-world, the hermeneutic cycle,
and the double hermeneutic. By connecting each concept to a key stage of

the research process, this chapter offers a phenomenological approach to

the study of emotions that is firmly rooted in its philosophical heritage.

Intentionality: Research Questions that Focus on Experience

Intentionality is a central concept in phenomenological inquiry, employed

by Husserl to refer to consciousness always being consciousness of some-
thing. For example, “every act of loving is a loving of something, every act
of seeing is a seeing of something” (Moran, 2000, p. 16). In this way, indivi-
duals cannot separate an experience from what is experienced there is an
intentional relationship between an object and an individual’s awareness of
it. Heidegger also emphasized the importance of intentionality but argued,
unlike Husserl, that intentional experiences reside in individuals’ relation-
ships and not in their own inner sphere. As such, to Heidegger, the experi-
ence of an intentional relationship is likely to vary across different
contexts. Nonetheless, this notion of intentionality remains important for
the study of emotions as it posits that emotions cannot exist independently
of individuals’ experiences of an object. As Gunther (2004, p. 47) describes,
“take the case of the grief one experiences upon learning that a loved one
has passed away. It isn’t as though one experiences the feeling indepen-
dently of what one is grieving about.” In this way, one of the central attri-
butes of experiencing an emotion is intentionality: an emotion has an
object (Solomon, 2006). To phenomenologists, then, individuals cannot
separate the emotions that are involved in their experiences from what is
A researcher therefore begins by investigating an individual’s experience
of an organizational event or process. This primary focus allows a
researcher to explore how an emotion or set of emotions may feature in
that experience subsequently. As Heidegger (1996) pointed out, individuals’
emotions and moods intertwine their experiences or interpretations of
events. Phenomenological approaches thereby assume that “our feelings
are not senseless states of consciousness or psychic facts, but modes to
detect the signification of situations, to know what is savory, disgusting,
alarming, distressing, lovely etc.” (Buytendijk, 1987, p. 120). Studying indi-
viduals’ experiences of significant organizational events is likely to highlight
the role of emotions. This is why IPA studies are not usually concerned
with the “smallest units” of experience that present themselves in the every-
day flow of life but, instead, with the “comprehensive units” of experience
that are meaningful or significant to individuals (Dilthey, 1976; Smith
A Phenomenology of Feeling 35

et al., 2009). For example, an IPA study could investigate a significant

organizational event from the perspective of the individual, like the experi-
ence of maternity leave and returning to work. A researcher utilizing IPA
would not be concerned with the nature of maternity leave itself but,
instead, with how particular individuals experience or make sense of these
events. An IPA study could pose the question “how does a women’s experi-
ence of identity change during maternity leave?”
The search for meaning that particular experiences hold necessitates
open research questions without hypotheses. This requires that a
researcher seeks to understand an experience without predetermining or
presuming the role of particular emotions. This does not mean that a
researcher cannot explore emotions in their data collection, or re-focus
their research, but that they should begin their research by generating a
research question that focuses on the experience of a particular organiza-
tional event or process. An example of this is Eatough and Smith’s (2006a)
study of how significant life events relate to experiences of anger. Initially,
this was a “study looking at how women experience and resolve conflict in
their lives” but began to examine anger explicitly as it emerged through
the narrative of one of their participant’s life story (Eatough & Smith,
2006a, p. 486). Through this research process, the authors provided a vivid
account of the experience of anger but did not presuppose the role of this
emotion at the outset.

Dasein: An Idiographic Approach to Sampling

Dasein is a term used by Heidegger to refer to individuals’ experience of

being the translation from German is “being-there,” where “there” is the
world. To Heidegger, individuals are thrown into, or always engaged with,
the world and cannot step outside of it. As Heidegger (1988, p. 297)
explained, the “self and world belong together in the single entity, Dasein.”
This rejection of the Cartesian divide between subject and object
(Gueldenberg & Helting, 2007) emphasizes the relational and contextual
nature of all experience. Many scholars, in their readings of Heidegger,
emphasize that emotion is an integral aspect of Dasein and how individuals
relate to the world (Morris, 2006; Ratcliffe, 2002; Solomon, 2006). Indeed,
Heidegger states that “Mood assails. It come neither from ‘without’ nor
from ‘within’, but rises from Being-in-the-world itself, as a mode of that
being” (Heidegger, 1996, p. 129). As Ratcliffe (2002, p. 289) explains,
“Moods, for Heidegger, give sense to Dasein’s world and to the manner in

which Dasein finds itself relating to the world.” Emotions therefore help to
make individuals’ experiences intelligible.
Phenomenological inquiry, which draws on Heidegger, emphasizes that
researchers must attend to participants’ cultural and historical context as
well as their emotions to understand their experiences. This necessitates an
idiographic approach with a desire to achieve an in-depth and detailed
understanding of each research participant or case. This matches IPA’s
focus on small samples, usually toward the lower end of a range from 1 to
30 participants (Brocki & Wearden, 2006; Cassell & Symon, 2004), as per
more traditional phenomenological studies (Starks & Brown Trinidad,
2007). The emphasis is on collecting rich data to understand the context
and perspective of an individual. This is in contrast to seeking sufficient
numbers of participants for achieving saturation, as is common in other
forms of qualitative research such as grounded theory (e.g., Strauss &
Corbin, 1990).
Given the small numbers of participants, IPA studies recruit participants
who can offer a meaningful perspective of the phenomenon of interest and
who therefore share a certain lived experience (Smith et al., 2009). For
instance, a shared experience could be an organizational event or process.
As such, the sampling of participants in IPA studies is purposive and
homogenous. This is similar to the ethnographic approach of social anthro-
pologists who research one particular community (Chapman & Smith,
2002). For instance, Murtagh et al. (2011) adopted an IPA study to under-
stand the experience and meaning of career changes and recruited eight
women, all of whom had voluntarily changed careers. This enabled the
researchers to focus on the rich accounts and context of each woman for
whom the research topic was meaningful. Their findings highlighted the
important role emotions played in driving career decisions and provided
empirical evidence for the other-than-rational perspective of decision-
making. For instance, when the authors asked their participants how they
made sense of their career decisions, “All of the participants described find-
ing that they were skilled and capable in the new field and drew on the
positive emotions related to the new career” (Murtagh et al., 2011, p. 255).
The participants in this study described how they felt “passionate” about
their work and described the central role of the positive emotions asso-
ciated with their decisions.
This study demonstrates IPA’s intention to seek out participants with a
shared experience so that the researcher can make claims about these peo-
ple but not all people (Langdridge, 2007). In this way, IPA draws explicitly
on Heidegger’s phenomenology to focus on a small number of participants
A Phenomenology of Feeling 37

to produce rich contextual accounts of individuals’ experiences. These

focused and rich accounts provide the opportunity to illuminate indivi-
duals’ emotions and role of these emotions in particular contexts.

The Life-World: Collecting Data that Reflects the

Participant’s Perspective

Husserl (1970, p. 49) describes “the only real world” as “the one that
is actually given through perception, that is experienced and experience-
able our everyday lifeworld.” Husserl used this concept of the life-world
to “describe the everyday world with which we are inevitably intertwined
and which we take for granted” (Sandberg & Dall’Alba, 2009, p. 1353).
Heidegger built on and developed this concept, to stress that being-in-the-
world is ontologically constitutive of the life-world (Sandberg &
Dall’Alba, 2009). In this way, the life-world describes the inseparable rela-
tionship between an individual and the world, which an individual always
experiences from their unique perspective (Sandberg, 2005). Furthermore,
for Heidegger, emotions and moods are a constituent of the life-world,
part of how we “attune” to our experiences of everyday life (Heidegger,
1996). As Todres, Galvin, and Dahlberg (2007, p. 57) explain: “Lived
experience is coloured by mood. It interpenetrates the other dimensions of
the lifeworld, shaping one’s spatial, temporal, intersubjective and embo-
died horizons. Anxiety reveals very different profiles of the world than joy
or tranquillity do.”
Phenomenological inquiry therefore seeks to examine individuals’ first-
person perspectives to understand how they experience particular phenom-
ena in particular contexts. This corresponds to IPA’s aim to describe the
“life world of the particular participants who have told their stories”
(Smith, 2004, p. 42). IPA studies share Heidegger’s stance that meaning
resides in the networks and relationships of individuals. Though an indivi-
dual’s account can never provide the full picture of an experience, it can
offer their perspective of their world as they subjectively experience it and
therefore some insights into a particular experience and associated
As Larkin et al. (2006) note, an IPA researcher must accept a third-
person perspective of a first-person account and, consequently, should
strive to see the world as the participant sees it by being sensitive to their
account. These ambitions explain the prevalence of semi-structured inter-
views as a method of data collection within IPA studies (Reid et al., 2005).

Semi-structured interviews offer the opportunity to focus on a particular

experience, but to capture much of the rich detail from individuals’ own
perspectives by enabling participants to steer an interview, shape the find-
ings and share their own story (for a detailed approach to planning and
conducting interviews see Smith & Osborn, 2003). For example, Wise and
Millward’s (2005, p. 403) IPA study investigated the career decisions of
“30-somethings.” They “employed semi-structured interviews to provide a
context of choice in which participants could provide a coherent and holis-
tic account of their career changes.” This approach allowed the authors to
demonstrate the important influence of context. For instance, how events
such as work-place bullying or getting divorced “invariably resulted in psy-
chological states such as depression and low self-worth that were found to
hamper aspects of the ensuing change process” (Wise & Millward, 2005,
p. 414). These contextualized accounts helped the authors to relay how par-
ticipants’ personal “feelings of dissatisfaction” and “feelings of regret”
crept into their career decisions. IPA’s procedures guide the researcher to
get as close as possible to each participant’s perspective during the collec-
tion of data to more accurately understand experiences and associated
emotions in their organizational life-world.

Hermeneutic Cycle: Iterative Data Analysis

Heidegger drew heavily on the ideas of hermeneutics to emphasize the role

of interpretation in any phenomenological effort. Indeed, he notes that the
“Phenomenology of Dasein is hermeneutics in the original signification of
that word, which designates the work of interpretation” (Heidegger, 1996,
p. 33). A key hermeneutic concept is the hermeneutic circle or cycle. This
refers to the idea that acts of interpretation are cyclical processes.
Philosophers such as Schleiermacher and Heidegger realized that “we can-
not understand a part without some understanding of the whole, yet we
cannot understand the whole without understanding its parts” (Inwood,
2000, p. 89). Smith et al. (2009) suggest one example of this process is that
the meaning of words only become clear when understood in the context of
the sentences that embeds them.
In this way, phenomenological inquiry requires an iterative analysis of
data that moves back and forth through participants’ accounts to under-
stand their experiences. IPA reflects this requirement by moving between
“emic” and “etic” positions, where the researcher moves between a phe-
nomenological or insider position to an interpretive or outsider position
A Phenomenology of Feeling 39

(Reid et al., 2005). This movement begins by initially considering each indi-
vidual’s account. IPA guides researchers to immerse themselves in the col-
lected data, typically interview transcripts, and searching for points of
significance. The comparison of one transcript with another grounds the
researcher in the data and leads back to further analysis in an iterative, cir-
cular fashion. As with other phenomenological approaches (e.g., Van
Manen, 1990), IPA studies eventually conduct a thematic analysis to unra-
vel the themes of shared experiences across the participants. Researchers
typically cluster these themes together into subordinate and, ultimately,
superordinate themes (for an example of this process in terms of the emo-
tion of anger, see Eatough & Smith, 2006b).
By interpreting shared experiences and developing associated themes,
IPA studies can illuminate how individuals understand or make sense of
particular experiences and the meaning or importance of particular features
of this experience. Researchers can use this information to answer particu-
lar research questions and inform organizational knowledge. For example,
Millward’s (2006) IPA study sought to understand how women experience
maternity leave. Her study of ten women’s experiences began with a
detailed analysis of each individual case and then looked for similar or
divergent accounts in other cases. This process allowed an array of sub-
themes to emerge, which drew on the participants’ accounts of feeling
guilty, happy, and insecure at various times during their experience. These
sub-themes in turn produced two master themes: changes in identity and
changed psychological contracts. These themes emerged across each partici-
pants account and clearly highlighted the need for a broader conceptualiza-
tion of maternity leave as more than a legal process for organizations. In
this way, Millward (2006) was able to demonstrate how particular emotions
or feelings informed and shaped participants’ experiences of maternity
leave, as one part of a whole experience. Such an iterative analysis practi-
cally reflects the concept of the hermeneutic cycle and reaffirms IPA’s theo-
retical links with hermeneutic phenomenology.

Double Hermeneutic: Reflexively Evaluating the Research

In recognizing that emotions help individuals to understand their engage-

ment with the world, Heidegger’s (1996) ideas also remind phenomenologi-
cal researchers that their own biases and emotions are likely to shape their
understating of participants’ experiences. In this way, hermeneutic phenom-
enology’s attention to context also extends beyond the participant under

study to consider the interaction of the researcher’s own context. This is

what Gadamer referred to as an encounter between two sets of prejudices
(Warnke, 1987). As the researcher is interpreting the experience of the par-
ticipant, who is in turn interpreting their experiences, a double hermeneutic
process operates (Smith & Osborn, 2003).
Phenomenological researchers should therefore appreciate that they can-
not conduct explorations of experience completely or directly and must
consider their own context and how this may shape any findings. The
guidelines of IPA emphasize reflexivity to examine the interpretations of
the researcher. IPA studies often cite Yardley’s (2000) principles for asses-
sing the quality of qualitative investigations (see Smith et al., 2009).
These are similar to many of the other criteria or procedures of quality
popularized in organizational and social science research (Creswell, 1998;
Lincoln & Guba, 1985) and include sensitivity to context and the need for
findings to be insightful and impactful. By challenging researchers to
explore their own context and situation as well as of those of their research
participants, IPA’s sensitivity to context is apparent. Moreover, as
Heidegger suggested, being methodologically self-aware and dwelling on
the conditions and procedures of interpretation will produce insightful
interpretations (Hoy, 1993).
Following Heidegger, it is impossible to “bracket” or remove our pre-
conceptions, as it is our assumptions and beliefs that allow us to interpret
and understand experiences. Nonetheless, researchers can attempt to con-
sider and reflect on their assumptions to consider how they may have
shaped their interpretations through several procedures. These are demon-
strated across several different IPA studies. For instance, Oakland,
MacDonald, and Flowers’s (2012) IPA study explored the impact of seven
opera choristers losing their employment. They utilized a reflexive diary to
“monitor any preconceptions and influences on the data collection and sub-
sequent analysis” (2012, p. 138). In a further example, Wise and Millward
(2005) requested other experienced researchers to challenge their findings
and asked several of the participants to provide feedback on the suitability
of their themes. Collectively, such procedures facilitate plausible and trans-
parent analyses thereby meeting much of the quality criteria of IPA
(Yardley, 2000) and those suggested by a variety of organizational scholars
(e.g., Creswell, 1998). These procedures therefore support researchers in
reflecting on the role of their own emotions in shaping their interpretations
of participants’ experiences.
By drawing on a range of phenomenological and hermeneutic concepts,
this section has elaborated the established procedures of IPA to provide
A Phenomenology of Feeling 41

guidelines for the examination of emotions. These guidelines consider a

variety of stages of research, from the selection of participants, through to
the methods of data collection and analysis. Whilst this guidance may
appear superficially similar to other approaches, there are substantial dif-
ferences that stem from the underlying assumptions of IPA.



IPA shares many similarities with other qualitative approaches that are
more widely recognized and adopted in organizational research, such as
grounded theory and discourse analysis. However, it is distinct from these
approaches and diverges from other forms of phenomenological inquiry. It
is important to explicate these differences to ensure researchers understand
the applicability and appropriateness of IPA to address particular research
Grounded theory (Charmaz, 2006; O’Reilly, Paper, & Marx, 2012) may
appear similar to IPA and is often discussed as a similar method (Smith &
Osborn, 2003). Both require researchers to produce findings grounded in
data and to conduct thematic analyses. However, the ambitions of the two
approaches differ. Whilst grounded theory intends to produce theories that
explain social processes, IPA seeks to understand the experience of specific
individuals or groups and does not necessarily seek to generalize. At a
more practical level, IPA’s commitment to understanding individuals’
experiences without establishing causal relationships requires a small sam-
ple without the theoretical saturation necessary for grounded theory
(Strauss & Corbin, 1990). Suddaby (2006) also clarifies these distinctions,
through a comparison of grounded theory and phenomenology more
broadly, and notes that the former tends to draw from multiple data
sources, whilst the latter typically utilizes interviews. Though there are
exceptions within the IPA corpus (Smith, 1999), it does appear reasonable
to suggest that IPA studies primarily use interviews for data collection.
IPA, alongside discourse analysis (Gee, 2010; Phillips, Sewell, & Jaynes,
2008), holds a view that language is crucially important in qualitative ana-
lysis. However, discourse analysis is typically skeptical of mapping indivi-
duals’ verbal accounts on to underlying subjective experiences, preferring
to analyze verbal or written accounts in their own right (Crossley, 2000). In
contrast, IPA assumes a “chain of connection” (Chapman & Smith, 2002)

between what a participants says and their cognitions. It is important to re-

iterate, however, that IPA contends that an individual’s language connects
to their lived experience of reality but not to an objective reality.
Just as grounded theory and discourse analysis contain variations in
form and application, so too does research within the family of phenomen-
ological approaches. The development of differing phenomenological
approaches has occurred in pedagogy (Van Manen, 1990), psychology
(Giorgi, 1985), and sociology (Psathas, 1973). Whilst Smith et al. (2009)
view these different approaches as complementary, they do possess distinc-
tive features. Although it is beyond the scope of this chapter to compare
and contrast these different approaches (see Gill, 2014), it is important to
note that they possess different aims and assumptions and therefore should
not be viewed as interchangeable.
Whilst IPA shares a number of similarities with other forms of estab-
lished qualitative and phenomenological inquiry, these should not obscure
fundamental differences. An idiographic focus that does not necessarily
seek to establish causal relationships, an assumption of a chain of connec-
tion between what individuals say and what they experience, and the
adoption of a Heideggerian emphasis on interpretation renders IPA a
unique qualitative approach. This chapter does not suggest that IPA
should supplant existing qualitative approaches or replace other forms of
phenomenological inquiry. However, IPA is a distinct approach that is
appropriate to address particular research questions consistent with its
epistemological position. That is, when researchers are trying to discover
how individuals make sense of their personal and social world (Smith &
Osborn, 2003).



IPA provides four advantages for organizational researchers seeking to

examine emotions. First, it can provide contextualized and rich accounts
that illuminate the nuances and shades of emotions that shape individuals’
experiences. Second, it can also help to explain social phenomena from the
“bottom up” and is particularly valuable in understanding the role of emo-
tions in processes. Third, it can complement existing research methods and
enrich extant theory by privileging the, often emotive, voice of participants
that other approaches often obscure. Fourth, it is applicable and of value
A Phenomenology of Feeling 43

to a diverse range of organizational subject areas, from entrepreneurship to

strategy. The following paragraphs consider each of these advantages.
The first advantage of employing an IPA approach is its ability to illumi-
nate the role of emotions in relation to individuals’ complex and nuanced
experiences. Phenomenological inquiry recognizes that feelings, like colors,
possess different shades. For example, some manifestations of an emotion
may be more intense than others (Gunther, 2004). IPA therefore seeks to
illuminate the role of emotion in individuals’ experiences within a particu-
lar context. This is of significant utility for developing organizational
insights. For example, it can enable researchers to explain how and why
certain individuals make particular decisions by establishing what is mean-
ingful to them in certain circumstances. Many IPA studies have revealed
that emotions intertwine the decision-making process. For instance,
Fitzgerald and Howe-Walsh (2008) applied the principles of IPA to estab-
lish and understand the factors that influenced female professional expatri-
ates’ decisions to pursue international work experience. By drawing on the
accounts of 10 women, the authors were able to provide rich accounts that
could highlight how their participants would “feel scared” and how this
would inform their experience of working abroad. In this way, IPA’s idio-
graphic focus enables researchers to explicate many of the personal experi-
ences and emotions that drive employees and entrepreneurs’ actions.
Organizational theorists equipped with these insights can better understand
the experiences of individuals and, in turn, the organizations that embed
them. Further IPA studies could address a whole host of valuable questions
such as what influences particular employees to leave or resign from their
work or to dedicate themselves to a chosen profession. In doing so, these
questions are likely to highlight the powerful role of emotions.
The second advantage of IPA is that it can also help to explain the role
of emotions in social phenomena and processes from the “bottom up.” As
Smith and Osborn (2003) note, IPA is useful when examining complexity
and processes, as it can unravel and explain these from the perspective of
individuals. For instance, Küpers, Mantere, and Statler (2012) used herme-
neutic and phenomenological methods, though not an explicit IPA
approach, to understand individuals’ lived experiences of strategy in an I.T.
corporation. The authors argued that these methods could enable research-
ers to come into “closer touch with individual and social realities” (2012,
p. 2). Their approach explicated the lived and emergent, as opposed to
deliberate, narrative processes of strategy development that occur within an
organization’s strategy workshop. In doing so, their results challenged
much of the cognitive bias that prevails in strategy research by emphasizing

the role of embodied, emotional and meaningful experiences in the emer-

gence of strategy. For example, how particular stories and phrases evoked
strong emotional reactions in employees, which guided their strategic rea-
soning. IPA, as a systematized form of inquiry, can equip researchers to
conduct similar investigations and to understand the potentially emotional
processes that facilitate the appearance and development of organizational
phenomena, such as strategy.
The third advantage of IPA studies is that they are well suited to com-
plementing some existing organizational theories, by privileging the experi-
ence of participants. Given the absence of individuals’ accounts from many
organizational studies, through methodological emphasizes on aggregation
and commonalities without being ground in each individual’s account, IPA
has the potential to revisit, develop and enrich much existing theory. For
instance, Murtagh et al.’s (2011) IPA study highlighted the need to incor-
porate emotional and non-rational perspectives into the existing theory of
career decision-making. In the same vein, Wise and Millward’s (2005)
research revealed that whilst 30-somethings sought career progression they
did not desire it in a traditional sense, instead seeking to adopt a protean
career. By examining individuals’ personal accounts of the meanings they
ascribe to their careers, these findings revise elements of the traditional
career theory model of structured vocational development in set stages and
highlight the neglected role of emotion. IPA could support further investi-
gation into other established constructs by considering how particular indi-
viduals experience and understand them.
Fourth, IPA is applicable and of value to a diverse range of organiza-
tional subject areas. For instance, both Berglund (2007) and Cope (2005)
posit that phenomenology is a useful tool for studying entrepreneurship and
make a convincing case for its potential to complement other approaches.
For example, by exploring how entrepreneurs experience particular con-
cepts, such as risk-taking and business planning. Cope (2011) conducted an
IPA study to explore how entrepreneurs learn from failure. He found that
“the emotional cost of failure was emphasized by all the participants”
(Cope, 2011, p. 610) and illuminates how this can have a significant impact
on individuals well-being. In doing so, Cope highlights how much of the
extant literature fails to recognize the emotional and social costs that entre-
preneurs’ endure and how these two sets of costs are closely related. By
focusing on the individual in their social context, IPA can provide richer and
broader meaning to a range of established concepts in organization studies.
As demonstrated in the examples cited in this section, the utility of an
IPA approach extends across many issues and fields. Alongside career
A Phenomenology of Feeling 45

theory, entrepreneurship and strategy, there is the potential for IPA to sup-
port research into a variety of other fields. For example, various streams
within the prominent body of institutional theory in organizational scholar-
ship call for the incorporation of individuals’ experiences (Suddaby, 2010)
and particularly emotional experiences (Clemente & Roulet, 2015;
Voronov, 2014; Voronov & Vince, 2012). IPA studies could shed light on
how emotions intertwine individuals’ experience of particular events, pro-
cesses, services, or products. Given the limited application of IPA studies
to areas of organizational research thus far, many established topics and
fields provide fertile ground for rich and original insights into the role of
emotions to emerge.


IPA is a young approach to qualitative analysis and continues to develop

(Larkin et al., 2006). It is therefore important to consider the potential lim-
itations of IPA to support its progress and establish its applicability to
organizational studies. It is beyond the scope of this chapter to consider all
the potential limitations of IPA, or of qualitative research more generally.
As such, this section reflects on IPA’s focus on interpreting of the subjec-
tive experience of individuals. Whilst this is a unique and valuable focus, it
also produces two important limitations. First, IPA is more appropriate for
small numbers of cases or individuals and does not lend itself to generaliza-
tion. Second, IPA’s use of interpretation is craft-like in nature, offering
guidelines more than rigid rules, which leads to some variability in its
IPA’s focus on individuals’ experiences circumscribes its utility. An IPA
study can help researchers to explore individuals’ understandings and
meanings, but cannot provide definitive answers as to what something “is”
(Larkin et al., 2006) or produce universal truths. IPA is not necessarily
opposed to making general claims for larger populations, but it is primarily
committed to an idiographic approach and therefore to the analysis of
small numbers of cases (Smith & Osborn, 2003). This focus can also lead to
practical difficulties. For example, given the associated need to recruit
homogenous samples, participants can be more difficult to recruit than ran-
dom samples, particularly over sustained periods (Smith et al., 2009).
Nonetheless, by focusing on a small number of individuals, an IPA
approach allows researchers to explore experiences and to relay emotive
and rich accounts that could be lost through the process of generalization.

In IPA studies, the researcher is an interpreter. This means that there is

a “double hermeneutic,” as the researcher attempts to make sense of an
individual’s account whilst that individual is making sense of an experience
(Smith & Osborn, 2003). The interpretation of data and the development
of themes are much simpler to describe than to achieve and require creativ-
ity and consistent effort on the part of the researcher. Given the subjective
nature of this process, IPA offers recommendations rather than rules. This
results in some variability in how researchers interpreted their data, as
Brocki and Wearden (2006) revealed in their critical evaluation of the use
of IPA by analyzing 52 published studies. As such, it would be more appro-
priate to describe IPA as a qualitative craft (Cunliffe, 2011), and to re-
iterate that an IPA is complex process that demands sustained sensitivity to
the accounts of participants. Nonetheless, Brocki and Wearden (2006, p. 1)
found much promise in the approach and stated that IPA is “applicable
and useful in a wide variety of research topics.”


A key assumption of hermeneutic phenomenological inquiry is that emo-

tions intertwine individuals’ experiences. By offering one way to interpret
how individuals make sense of particular experiences, IPA offers the possi-
bility of illuminating many of the emotions that are bound up in indivi-
duals’ experiences of organizational events or processes. Furthermore,
given its relative youth as a qualitative research approach, there is consider-
able potential for organizational researchers to employ IPA to reveal new
and valuable insights. As such, this chapter has clarified and elaborated the
guidelines of IPA, as one particular form of phenomenological inquiry, to
support scholars in eliciting the role of emotions in organizations and orga-
nizational theory. It has outlined IPA’s distinctive philosophical heritage in
phenomenology and hermeneutics, which emphasizes emotions as integral
to individuals’ experiences. It then outlined how a philosophically
grounded IPA research project could be structured to explore the role of
emotions. This chapter articulated four valuable contributions of the
approach. First, it can provide rich and contextual accounts that shed light
on the emotional meanings individuals’ ascribe to particular experiences.
Second, individuals’ emotive and rich accounts can explain the emergence
of particular behaviors and social phenomena. Third, it can complement
some existing research methods to enrich extant theory by privileging the
A Phenomenology of Feeling 47

voice of participants. Fourth, it is applicable to a diverse range of organiza-

tional subject areas. As Solomon (2006, p. 291) noted, the “realm of emo-
tions is rich and promising territory for phenomenologists.”

1. The spelling of “interpretative” in IPA is distinct from the “interpretive”
adopted by other approaches. This chapter adopts the former only when spelling
out the IPA acronym.
2. This chapter follows both Ratcliffe (2002) and Solomon (2006), who note that
whilst Heidegger refers primarily to “moods” his ideas can encompass emotions
more generally.


Ashforth, B. E., & Humphrey, R. H. (1995). Emotion in the workplace: A reappraisal. Human
Relations, 48(2), 97 125.
Berglund, H. (2007). Researching entrepreneurship as lived experience. In H. Neergaard &
J. P. Ulhøi (Eds.), Handbook of qualitative research methods in entrepreneurship
(pp. 75 93). Cheltenham: Edward Elgar Publishing, Inc.
Brocki, J. M., & Wearden, A. J. (2006). A critical evaluation of the use of interpretative phe-
nomenological analysis (IPA) in health psychology. Psychology & Health, 21(1),
87 108.
Buytendijk, F. J. J. (1987). The phenomenological approach to the problem of feelings and
emotions. In J. Kockelmans (Ed.), Phenomenological psychology (Vol. 103,
pp. 119 132). Netherlands: Springer.
Cassell, C., & Symon, G. (2004). Essential guide to qualitative methods in organizational
research. London: Sage.
Chapman, E., & Smith, J. A. (2002). Interpretative phenomenological analysis and the new
genetics. Journal of Health Psychology, 7(2), 125 130.
Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analy-
sis. London: Sage.
Clemente, M., & Roulet, T. (2015). Public opinion as a source of deinstitutionalization: A
“spiral of silence” approach. Academy of Management Review, 40(1), 96 114.
Cope, J. (2005). Researching entrepreneurship through phenomenological inquiry philosophi-
cal and methodological issues. International Small Business Journal, 23(2), 163 189.
Cope, J. (2011). Entrepreneurial learning from failure: An interpretative phenomenological
analysis. Journal of Business Venturing, 26(6), 604 623.
Creswell, J. W. (1998). Qualitative inquiry and research design: Choosing among five traditions.
Thousand Oaks, CA: Sage.
Crossley, M. L. (2000). Narrative psychology, trauma and the study of self/identity. Theory &
Psychology, 10(4), 527 546.

Cunliffe, A. L. (2011). Crafting qualitative research: Morgan and Smircich 30 years on.
Organizational Research Methods, 14(4), 647 673.
Dilthey, W. (1976). The development of hermeneutics (H. Rickman, Trans.). In H. Rickman
(Ed.), W. Dilthey: Selected writings (pp. 246 263). Cambridge: CUP.
Drummond, J. J. (2006). Respect as a moral emotion: A phenomenological approach. Husserl
Studies, 22(1), 1 27.
Eatough, V., & Smith, J. (2006a). I was like a wild wild person: Understanding feelings of
anger using interpretative phenomenological analysis. British Journal of Psychology,
97(4), 483 498.
Eatough, V., & Smith, J. A. (2006b). I feel like a scrambled egg in my head: An idiographic
case study of meaning making and anger using interpretative phenomenological analy-
sis. Psychology and Psychotherapy: Theory, Research and Practice, 79(1), 115 135.
Fineman, S., & Sturdy, A. (1999). The emotions of control: A qualitative exploration of envir-
onmental regulation. Human Relations, 52(5), 631 663.
Fitzgerald, C., & Howe-Walsh, L. (2008). Self-initiated expatriates: An interpretative phenom-
enological analysis of professional female expatriates. International Journal of Business
and Management, 3(10), 156 175.
Gee, J. P. (2010). An introduction to discourse analysis: Theory and method. Abingdon:
Gill, M. J. (2014). The possibilities of phenomenology for organizational research.
Organizational Research Methods, 17(2), 118 137.
Gill, M. J. (2015). Elite identity and status anxiety: An interpretative phenomenological
analysis of management consultants. Organization, 22(3), 306 325.
Giorgi, A. (1985). Phenomenology and psychological research. Pittsburgh, PA: Duquesne
University Press.
Gueldenberg, S., & Helting, H. (2007). Bridging “the great divide”: Nonaka’s synthesis of
“western” and “eastern” knowledge concepts reassessed. Organization, 14(1), 101 122.
Gunther, Y. H. (2004). The phenomenology and intentionality of emotion. Philosophical
Studies: An International Journal for Philosophy in the Analytic Tradition, 117(1/2),
43 55.
Hatzimoysis, A. (2010). Emotions in Heidegger and Sartre. In P. Goldie (Ed.), The oxford
handbook of emotion (pp. 215 236). Oxford: Oxford University Press.
Heidegger, M. (1988). The basic problems of phenomenology (A. Hofstadter, Trans).
Bloomington, IN: Indiana University Press.
Heidegger, M. (1996). Being and time (J. Stambaugh, Trans.). Albany: SUNY Press.
Hoy, D. C. (1993). Heidegger and the hermeneutic turn. In C. Guignon (Ed.), The Cambridge
companion to Heidegger (pp. 170 194). Cambridge: CUP.
Husserl, E. (1970). The crisis of European sciences and transcendental phenomenology: An intro-
duction to phenomenological philosophy. Evanston, IL: Northwestern University Press.
Inwood, M. (2000). Heidegger: A very short introduction. Oxford: Oxford University Press.
Küpers, W., Mantere, S., & Statler, M. (2012). Strategy as storytelling: A phenomenological
collaboration. Journal of Management Inquiry, 21, 298 317.
Langdridge, D. (2007). Phenomenological psychology: Theory, research, and method. Essex:
Prentice Hall.
Larkin, M., Watts, S., & Clifton, E. (2006). Giving voice and making sense in interpretative
phenomenological analysis. Qualitative Research in Psychology, 3(2), 102 120.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage.
A Phenomenology of Feeling 49

Millward, L. J. (2006). The transition to motherhood in an organizational context: An inter-

pretative phenomenological analysis. Journal of Occupational and Organizational
Psychology, 79(3), 315 333.
Moran, D. (2000). Introduction to phenomenology. London: Routledge.
Morris, W. F. (2006). Emotion and anxiety: A philosophic inquiry. Bloomington, IN: Xlibris
Moustakas, C. (1994). Phenomenological research methods. London: Sage.
Murtagh, N., Lopes, P. N., & Lyons, E. (2011). Decision making in voluntary career change:
An other-than-rational perspective. The Career Development Quarterly, 59(3), 249 263.
Oakland, J., MacDonald, R. A., & Flowers, P. (2012). Re-defining “me”: Exploring career
transition and the experience of loss in the context of redundancy for professional opera
choristers. Musicae Scientiae, 16(2), 135 147.
O’Reilly, K., Paper, D., & Marx, S. (2012). Demystifying grounded theory for business
research. Organizational Research Methods, 15(2), 247 262.
Phillips, N., Sewell, G., & Jaynes, S. (2008). Applying critical discourse analysis in strategic
management research. Organizational Research Methods, 11(4), 770 789.
Psathas, G. (1973). Introduction. In G. Psathas (Ed.), Phenomenological sociology (pp. 1 24).
New York, NY: Wiley.
Ratcliffe, M. (2002). Heidegger’s attunement and the neuropsychology of emotion.
Phenomenology and the Cognitive Sciences, 1(3), 287 312.
Rehman, S., & Roomi, M. A. (2012). Gender and work-life balance: A phenomenological
study of women entrepreneurs in Pakistan. Journal of Small Business and Enterprise
Development, 19(2), 209 228.
Reid, K., Flowers, P., & Larkin, M. (2005). Interpretative phenomenological analysis: An
overview and methodological review. The Psychologist, 18, 20 23.
Sandberg, J. (2005). How do we justify knowledge produced within interpretive approaches?
Organizational Research Methods, 8(1), 41 68.
Sandberg, J., & Dall’Alba, G. (2009). Returning to practice anew: A life-world perspective.
Organization Studies, 30(12), 1349 1368.
Smith, J. A. (1996). Beyond the divide between cognition and discourse: Using interpretative
phenomenological analysis in health psychology. Psychology & Health, 11(2), 261 271.
Smith, J. A. (1999). Identity development during the transition to motherhood: An interpreta-
tive phenomenological analysis. Journal of Reproductive and Infant Psychology, 17(3),
281 299.
Smith, J. A. (2004). Reflecting on the development of interpretative phenomenological analysis
and its contribution to qualitative research in psychology. Qualitative Research in
Psychology, 1(1), 39 54.
Smith, J. A. (2011). Evaluating the contribution of interpretative phenomenological analysis.
Health Psychology Review, 5(1), 9 27.
Smith, J. A., Flowers, P., & Larkin, M. (2009). Interpretative phenomenological analysis:
Theory, method and research. London: Sage.
Smith, J. A., & Osborn, M. (2003). Interpretative phenomenological analysis. In J. A. Smith
(Ed.), Qualitative psychology: A practical guide to research methods (pp. 51 80).
London: Sage.
Solomon, R. C. (2006). Emotions in phenomenology and existentialism. In H. L. Dreyfus &
M. A. Wrathall (Eds.), A companion to phenomenology and existentialism (pp.
291 309). Oxford: Wiley-Blackwell.

Stapleton, T. J. (1983). Husserl and Heidegger: The question of a phenomenological beginning.

Albany, NY: SUNY.
Starks, H., & Brown Trinidad, S. (2007). Choose your method: A comparison of phenomenol-
ogy, discourse analysis, and grounded theory. Qualitative Health Research, 17(10),
1372 1380.
Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory, procedures
and techniques. Newbury Park, CA: Sage.
Suddaby, R. (2006). From the editors: What grounded theory is not, Academy of Management
Journal, 49(4), 633 642.
Suddaby, R. (2010). Challenges for institutional theory. Journal of Management Inquiry, 19(1),
14 20.
Todres, L., Galvin, K., & Dahlberg, K. (2007). Lifeworld-led healthcare: Revisiting a humanis-
ing philosophy that integrates emerging trends. Medicine, Health Care and Philosophy,
10(1), 53 63.
Tomkins, L., & Eatough, V. (2014). Stop ‘helping’ me! identity, recognition and agency in the
nexus of work and care. Organization, 21(1), 3 21.
Van Manen, M. (1990). Researching lived experience: Human science for an action sensitive
pedagogy. USA: SUNY Press.
Voronov, M. (2014). Toward a toolkit for emotionalizing institutional theory. In N. M.
Ashkanasy, W. J. Zerbe, & C. E. J. Härtel (Eds.), Emotions and the organizational fabric
(Vol. 10, pp. 167 196). Research on Emotion in Organizations. Bingley, UK: Emerald
Group Publishing Limited.
Voronov, M., & Vince, R. (2012). Integrating emotions into the analysis of institutional work.
Academy of Management Review, 37(1), 58 81.
Warnke, G. (1987). Gadamer: Hermeneutics, tradition, and reason. Stanford, CA: Stanford
University Press.
Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. (2005). Organizing and the process of sensemak-
ing. Organization Science, 16(4), 409 421.
Wise, A. J., & Millward, L. J. (2005). The experiences of voluntary career change in 30-
somethings and implications for guidance. Career Development International, 10(5),
400 417.
Wrathall, M. A. (2007). Existential phenomenology. In H. L. Dreyfus & M. A. Wrathall
(Eds.), A companion to phenomenology and existentialism (pp. 31 47). Malden, MA:
Yardley, L. (2000). Dilemmas in qualitative health research. Psychology & Health, 15(2),
215 228.


Marilena Antoniadou, Peter John Sandiford,

Gillian Wright and Linda Patricia Alker


This chapter explores how Cypriot lecturers perceive and experience fear
while being at work. Drawing on the lens of interpretive inquiry, data
were collected through interviews with 19 lecturers. Analysis focused on
experiences of workplace fear offering rich insights into characteristics
of fear, eliciting events, and coping ways. Findings help to unveil the spe-
cific events that lead to fear in the Cypriot universities, and the ways lec-
turers manage their fearful experiences. The study contributes to the

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 51 79
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011004

study of discrete emotions, by empirically examining fear’s own storyline

through the workers’ own perspectives, within a specific context.
Keywords: Workplace fear; discrete emotions; higher education;
interpretive study; Cyprus


The view that organizations are emotional arenas in which different emo-
tions are elicited, displayed, suppressed, and managed did not develop
seriously in the literature until 1980s. Studies of workplace events
(e.g., Basch & Fisher, 2000; Boudens, 2005; Fineman, 1993; Weiss &
Cropanzano, 1996) reveal a wide range of emotional states generated in
organizations. However, scholars tend to treat most emotions in a relatively
undifferentiated manner, grouping them into positive or negative, primary
or secondary states, rather than as individually identifiable constructs (e.g.,
anger, envy), despite numerous calls for the study of discrete emotions
rather than emotions in general (Gooty, Gavin, & Ashkanasy, 2009;
Oatley, Dacher, & Jenkins, 1996; Tiedens & Linton, 2001). Such broad
conceptualisations seem to be inadequate indicators of workers’ emotional
experiences, in that they diminish the extent to which an emotion is consid-
ered distinctive in relation to its subjective experience, antecedents, and
consequences. However, despite this overemphasis on general conceptuali-
sations (e.g., Game, 2008; Watson, 2000), there have been some recent and
notable exceptions investigating discrete emotions, namely anger
(e.g., Domagalski & Steelman, 2005; Fitness, 2000; Geddes & Callister,
2007; Glomb, 2002), jealousy and envy (Vecchio, 2000), shame and guilt
(e.g., Barclay, Skarlicki, & Pugh, 2005). Certainly, it is necessary to focus
more on investigating the discrete nature of emotions if we are interested in
the different processes that drive them and the different outcomes that may
result from them (Gooty et al., 2009). Each emotion has its own storyline,
hence excluding discrete antecedents and effects results in fragmented
research that can confuse the field of study. One way to avoid this would
be to explore emotional events in organizations, based on workers’ narra-
tives about the emotions they experience at work. This study responds to
the call to research discrete emotions at work by adopting an interpretive
approach to explore workers’ narratives of fear in the workplace. Using
stories to explicate events and trace processes of emotion experience and
Understanding Lecturers’ Perceptions of Workplace Fear 53

expression help unravel some of the complex meanings that embrace emo-
tional lives at work (Lazarus & Lazarus, 1994).


Defining and Distinguishing Fear

From a psychological perspective, fear is an intense emotion experienced

in the presence of a perceived, immediate threat contrasting with anxiety
in that fear has an object (Ahmed, 2003), while anxiety is somewhat
more vague in focus, representing a longer-term state of uneasiness, often
without a clear and specific object. Thus, although, fear is described as
an emotional reaction to an identifiable threat, anxiety is described more
as a feeling of uneasy suspense (Rachman, 1974). Similarly, Heidegger
discusses anxiety, arguing that it does not have an easily identifiable
cause as it is characterized by the fact that “what threatens is nowhere”
(1962, p. 231). An anxious person may feel flushed, uncomfortable, and
an urge to flee, but if the person has no sense of a fearful object, or if
there is nothing obviously objectionable or fearful to the person, then
that feeling will not count as feeling afraid (Lazarus & Lazarus, 1994).
Therefore, an appraisal, belief, or judgment that there is a clear or
upcoming source of danger is essential to the experience of fear.
A number of early and contemporary researchers and theorists
(Aristotle, 1984; Darwin, 1872; Ekman, 1992; Fischer, Shaver, &
Carnochan, 1990; James, 1950; Plutchik, 1980; Shaver, Schwartz, Kirson, &
O’Connor, 1987) position fear within the basic human emotions. Social
constructivists also generally admit that fear is a primary emotion, as it
is hard-wired in the human neuroanatomy, although they would also
emphasize that the expression of fear is socially constructed (Kemper,
1987; Turner, 2000). Fear is broadly defined as an unpleasant emotion
triggered by perceived risk or danger, accompanied by a feeling of dislike
of particular conditions and objects (Gray, 1988). Current conceptualisa-
tions of fear have also characterized it as being a variable social emotion,
often contagious, that may affect group behavior and social policy; for
example, terrorist attacks, wars, and national threats have relied on the
emotion of fear, with different cultures, groups, and genders reacting dif-
ferently to it (Stearns, 2006). Indeed, emotions may be distributed differ-
ently across segments of a society, typically corresponding to each

segment’s social or economic status (e.g., individual shares of money,

power, and prestige) (Barbalet, 1998). Accordingly, fear is an emotion
that is differentially distributed across a society, resulting from a lack of
power and confidence that individuals may attribute to their own short-
comings (Barbalet, 1998). Moreover, its experience may be affected by
the context in which fear occurs, since a dangerous or threatening stimu-
lus, for instance a natural catastrophe, may pose little threat if it is seen
as part of watching a movie, rather than experienced in reality. It
appears, then, that evolution, social interactions, cultural prescriptions,
and context, penetrate every emotion, its occasions for appearing and its
duration (Frijda, 2007).

Fear’s Eliciting Events in the Workplace

Early research suggested that fear is elicited when workers face threatening
judgments about their role in the organization, especially when they are
faced with uncertainty about that role and when organizational changes
are anticipated (Kahn, Wolfe, Quinn, Snoek, & Rosenthal, 1964).
Accordingly, organizational fear may revolve around felt threats and loss
of control (Gibson, 2006). Although fear has long been seen as one of the
basic emotions, its presence in organizational theorizing is still limited
(Kish-Gephart, Detert, Treviño, & Edmondson, 2009). Workplace fear
appears not simply as a worker’s trait, rather as an emotion stemming
from several emotional interactions and conditions in the working environ-
ment. A broad view of fear is that it is a generalized experience of appre-
hension in the workplace (Rachman, 1974). Early research suggested that
fear is elicited when workers face threatening judgments about their role in
the organization, especially when they are faced with uncertainty about
that role and when organizational changes are anticipated (Kahn et al.,
1964). Accordingly, organizational fear may revolve around felt threats
and loss of control (Gibson, 2006).
Contemporary organizational research does draw out some of fear’s eli-
citing objects and events in the workplace. For example, conflict in hier-
archical relationships, as well as status differences and interactions with
disrespectful supervisors may generate fear in the worker (Tiedens, 2001);
when a high-performing employee is made redundant and/or humiliated by
supervisors, fear affects other employees (De Lara, 2006). As a result, fear-
driven workplaces with poor morale undermine employee commitment and
productivity (Namie, 2003). It was also found that new employment is a
Understanding Lecturers’ Perceptions of Workplace Fear 55

challenge for workers who may experience fear when entering a new envir-
onment and tend to seek job-related and emotional information from
supervisors and colleagues (Miller & Jablin, 1991; Morrison, 1993).
Research into fear at work often refers to job insecurity and worry
about the possible loss of employment (Burchell, Ladipo, & Wilkinson,
2002; De Witte, 2005; Dickerson & Green, 2012). Such research has shown
that job insecurity and fear of redundancy may adversely impact on psy-
chological well-being and job satisfaction, potentially leading to increased
psychosomatic complaints and physical strains (De Witte, 2005; Jordan,
Ashkanasy, Hartel, & Hooper, 2002; Wichert, 2005). Moreover, fear can
also be inspired by feelings of insecurity with respect to changes in the nat-
ure of the job or work role. Significant organizational change, such as mer-
gers or acquisitions, is a highly salient emotion-eliciting antecedent, during
which workers may experience a range of emotions and feelings such as
fear, anxiety, agony, sadness, powerlessness, and depression (Kirk, 1999;
Torkelson & Muhonen, 2003; Vince, 2006). Evidently, the majority of
research on fear is dominated by manifestations that fear is an emotion
that is negatively experienced. However, an increasing number of research-
ers have started to view emotions, like fear, as adaptively useful. For exam-
ple, fear may encourage avoidance of perceived threats and pessimistic
judgments, which protects the individual from risks and future unpleasant
outcomes (Izard, 1993; Lerner & Keltner, 2001; Maner & Gerend, 2007).
Fear can also influence a wide array of positive organizational phenomena;
for example, its expression may facilitate collective learning in organiza-
tions (Funlop & Rifin, 1997), influence team member/leader interaction,
communication, and improvement activity (Nembhard & Edmondson,
2006) and can influence decisions to reveal sexual orientation at work
(Ragins, Singh, & Cornwell, 2007).

Coping with Emotional Difficulties

Despite the positive outcomes that fear may be associated with, workers
tend to suppress expressions of fear more than emotions such as anger,
which are more likely to be overtly expressed and directed, at the person
who provoked the emotion (Gibson, 2006; Kish-Gephart et al., 2009). The
literature describes a number of behaviors that workers use to cope with
emotional difficulties and demands. Coping has been conceptualized as the
cognitive and behavioral effort and response of the person to events that
are perceived as negative (Weiten & Lloyd, 2003). Variations in the

individual’s coping responses to each emotion result from an appraisal pro-

cess that simultaneously takes into account personal factors along with
environmental demands, constraints, and opportunities (Lazarus &
Folkman, 1984).
Coping responses can be classified within three basic mechanisms:
appraisal-, problem-, and emotion-focused coping (Weiten & Lloyd,
2003). These three dimensions differ in the way coping behavior is tar-
geted: appraisal-focused coping aims to change one’s interpretations of
difficult experienced events (e.g., rationalization, re-interpretation);
problem-focused coping aims to alter the problem itself (e.g., seek social
support, improve time-management); and emotion-focused coping focuses
on managing the emotions of the person (e.g., meditating, exercising, dis-
traction). Emotion-focused coping has been particularly linked with emo-
tional labor as an attempt to regulate the expression of organizationally
desired emotions and fulfill the emotional display requirement of organi-
zations (Grandey, 2003). Workers are indeed selective in the degree to
which they authentically express and display their fear, which was con-
ceptualized as emotion work performance (Hochschild, 1983). Emotion
work leaves a discrepancy between felt and displayed emotions and is
seen as part of everyday social exchanges that ensure social stability and
the well-being of social actors (Hochschild, 1983). Emotion work can be
beneficial for the individual (e.g., Wouters, 1989), even though studies
have mainly emphasized the negative effects on the individual and organi-
zation (Grandey, 2003; Hochschild, 1983). In order to perform emotion
work and to regulate their emotions, workers, including university lec-
turers, use surface, and deep acting (Constanti & Gibbs, 2004). Surface
acting is performed when an individual hides spontaneous expression and
purposely puts on an appropriate emotion without actually feeling it,
while deep acting involves the reappraisal of a situation and an attempt
to modify the actual feeling by changing the determinants that gave rise
to it. Workers’ individual characteristics (personality) are significant
moderators of their engagement with different types of emotional man-
agement and expression (Fisher & Ashkanasy, 2000), apparently contri-
buting to the increasingly popular concept of emotional intelligence
(Goleman, 1995). This is based on the idea that personal skills and cogni-
tive abilities contribute to greater emotional sensitivity and ability to
cope with emotive demands at work, with more emotionally intelligent
individuals being better able to manage (and cope with) their fear, which
suggests that they might be more socially and emotionally capable than
others in other aspects of life.
Understanding Lecturers’ Perceptions of Workplace Fear 57

Fear in the Educational Context

In education, fear has long been identified in teachers’ emotional lives,

especially as being associated with losing control over their classes or losing
their job (Waller, 1932). Increased use of Information and
Communications Technology in lessons was also found to be another
determinant of teachers’ fear, often referred to as cyberphobia, which
includes the fear of using computers and lack of confidence with computer
jargon (Bradley & Russell, 1997). More modern changes in the educational
sector, such as the privatization of education, has also contributed to tea-
chers’ fear of change to their livelihood and professional accountability and
ultimately to fear of job loss (Conley & Glasman, 2008). Conversely, the
emotion of confidence can be generated from focused unit completion and
preparedness (O’Neill & Stephenson, 2012). Research suggests that profes-
sional development can help eliminate, or at least minimize, teachers’ fear
of teaching and helps them to experiment with new learning activities
(Buczynski & Hansen, 2010). Confidence building activities have also been
identified as a central element in the ability to learn about and master new
teaching practices and fulfill assessment responsibilities (Goos & Hughes,
In the context of Higher Education (HE), the emotional aspects of aca-
demic work remain under-researched (Pekrun, Goetz, Titz, & Perry, 2002),
although academics are also emotional and passionate beings (Hargreaves,
1998). HE institutions are often conceptualized as service providers with
customers, means of production, and service deliverers (Constanti &
Gibbs, 2004), although the nature of their customers and their relationships
with such service deliverers is rarely straightforward (Hoffman &
Kretovics, 2004). Academics interact with different stakeholders, such as
students, colleagues, administrators, and at times the media, which requires
producing an emotional state for other people. University work involves
emotional labor performance to achieve student satisfaction and attain
management goals (Bellas, 1999). It is not enough for academics to be
knowledgeable; they also need to communicate their knowledge in an enter-
taining way, as it is challenging to maintain the students’ interest and moti-
vation. Showing confidence in and enthusiasm for a subject is seen as a key
indicator of good teaching (Ramsden, 1992) and academics need to put on
a show and make use of paralanguage, nonverbal behaviors, humor, and
varied vocabulary (McKinney, 1988). Fear in HE, however, is seldom high-
lighted in the mainstream emotion literature. Relevant studies in the aca-
demic context have greatly dealt with organizational stress and emotional

labor as outcomes of certain events (e.g., Constanti & Gibbs, 2004;

Ogbonna & Harris, 2004), but not with the discrete emotions that these
stressors generate. Fear in academia was raised as a subsidiary outcome to
research projects rather than being central to them. For example, the aca-
demic context is characterized by fear of having their work stolen or plagi-
arized and fear of negative student evaluation (Zwagerman, 2008). Also,
decreased input and influence in policy generation and constant evaluation
and micromanagement of their work by managers and bureaucratic quality
assessment can create fear of loss of academic freedom in university tea-
chers (Conley & Glasman, 2008).


This study focuses on the emotion of fear, as a significant component of

organizational life which requires more attention (Ashkanasy & Nickolson,
2003), especially given its historical importance to psychology, sociology,
and psychiatry, together with its evolutionary nature (Plutchik, 1980).
From a psychological perspective, when attempting to explore people’s
behaviors, it is necessary to have a good understanding of the target per-
son’s perceptions of, and attitudes toward, the examined phenomenon.
Likewise, in therapeutic situations it is regarded as essential to obtain the
individual’s perception of the issue or problem before commencing therapy;
this understanding is likely to influence the individual’s responsiveness to
different therapeutic approaches (Cochran & Cochran, 2005; Egan, 2007).
Similarly, when exploring attitudes and beliefs toward a discrete emotion,
we can gain a better understanding of workers’ perceptions, and then
develop appropriate coping strategies with an increased probability of
We argue that having a deep understanding of the workers’ perspective
of fear is important for positive change to occur, and damaging outcomes
to be eliminated. Therefore, there is merit in understanding workers’ per-
spectives regarding fear in order to develop successful strategies to promote
organizational well-being. In the light of this, the aim of this research was
to examine academics’ understandings of, and attitudes toward, workplace
fear, with the intention of informing Universities about approaches that
might promote a greater awareness of fear and, therefore, mitigate any
potentially unpleasant or damaging consequences.
Understanding Lecturers’ Perceptions of Workplace Fear 59


The context of each study plays a significant role in the participants’ per-
ceptions of fear and the overall findings, since what people think, how they
feel, and what they do is shaped strongly by the social contexts in which
they live (Zembylas & Papanastasiou, 2006). Our research focused on HE
in Cyprus; exploring fearful experiences in Cypriot academia offers an
opportunity for a cultural perspective on discrete emotion research.
Uncovering culturally specific characteristics of emotional phenomena is an
important reason of conducting research in Cyprus, as these can be com-
pared to more universal findings on existing research in the field.
Insecurity and uncertainty have characterized the working environment
of academics in the United Kingdom and the United States for the last few
years, leading to the intensification of academic work accompanied by esca-
lating workloads, long hours, and increased surveillance and control
(Willmott, 1995). Notably, changes to the nature of academic work (stu-
dent teaching quality assessments, research assessment exercises, and teach-
ing quality reviews), together with increasing demands from other
stakeholders (students who demand greater levels of service, employers,
society) have provided tangible and comparable measures of lecturer per-
formance through which managers have tightened their control over the
academic labor process (Willmott, 1995).
Cyprus is experiencing what the United Kingdom and the United
States experienced during the past decades, with the introduction of fees
for students in HE. Although the HE sector in Cyprus is relatively
young, it has come a long way since independence. The increase in
demand for HE was growing (Ministry of Finance, 2013) and in 2005,
the House of Representatives approved the establishment and operation
of private universities (Ministry of Education and Culture, 2005). The
demand for HE exceeded public supply, so private markets took advan-
tage of this change, with three private colleges upgrading to private uni-
versity status. Cypriot universities are largely teaching-focused after the
accreditation of their programs with a few now investing in research
(Varnava-Marouchou, 2007). In this regard, little is known about how
these changes and the profession’s nature affect the emotions of aca-
demics. Moreover, cultural meanings of the profession have not been
raised in the organizational literature in any depth. There appears to be a
lack of research about the role of emotions in HE teaching, how lec-
turers’ emotional experiences relate to their teaching practices, how they

cope with their emotions, and how the socio-cultural context of teaching
interacts with their emotions (Bellas, 1999).


The research objective of the methodology used was to map the qualita-
tively different conceptualisations of fear held by lecturers’ working experi-
ences. Consistent with Heideggerian phenomenology, the authors assumed
that lecturers’ experiences would include expressions about what concerned
them most (Heidegger, 1962). Thus, the study took an interpretive stand,
aiming to capture the voice of the lecturers in order to make sense of the
nature of their fearful experiences. An interpretive approach adopts a posi-
tion that knowledge of reality is a social construction by human actors,
therefore, to understand a phenomenon requires studying it within its
natural environment from the participants’ meanings, without any attempts
to control or generalize (Ashforth & Humphrey, 1993). The interpretive
researcher assumes that interpreting the meanings of the lecturers’ fearful
experiences would uncover new understandings of how fear is generated
and managed within the workplace. This approach suggests that people in
organizations make retrospective sense of unexpected and disruptive events
through an ongoing process of action, selection, and interpretation (Weick,
1995). The concept of making sense of a phenomenon encourages the active
agents to construct sensible events and requires placing a stimuli into a
framework that enables its comprehension, construction of meaning, inter-
acting in pursuit of mutual understanding and patterning (Weick, 1995).
An interpretive approach can be more effective at capturing the richness asso-
ciated with emotional experiences, as it is primarily descriptive and focuses
on the perspective participants attach to everyday life (Sandberg, 2005).
The strength of applying an interpretive approach to the experience of fear
lies on the fact that value-free data cannot be obtained, since the interpretive
researchers use their preconceptions to guide the process of enquiry, while
closely working and interacting with the human subjects of the enquiry, chan-
ging the preconceptions of both parties (Ashforth & Humphrey, 1993;
Sandberg, 2005). This contrasts with the positivistic perspective, which
assumes that specific stimuli elicit the same specific emotions in a more or less
same biologically determined way, whereas the interpretive approaches
support that a given stimulus may elicit various emotional states and conse-
quences depending on how it is interpreted (Ashforth & Humphrey, 1993).
The choice of an interpretive approach is primarily geared around addressing
Understanding Lecturers’ Perceptions of Workplace Fear 61

the specific aims of the study, without explicitly entering into a discussion
about the relative merits of the positivist and constructivist approaches to the
study of emotion. In particular, the purpose of this approach is to elicit and
make sense of the data to show how fear and fearful events appear from the
perspective of the lecturers’ in their working environment. The authors
believe that within the context of workplace fear the interpretive approach is
appropriate, in order to capture the socially constructed contextual richness
and complexity of the emotion as experienced by lecturers.


A sample of 19 lecturers from three universities and from various disci-

plines were invited to participate in the study. All replied and were further
informed about the study and given the opportunity to ask questions about
participation. Following this second contact, all agreed to participate to the
research and signed consent forms. The sample consisted of 10 women and
9 men, lecturers and senior lecturers, having academic experience ranging
from 3 to 30 years, and ranged in age from early thirties to late fifties. The
sample size was determined following the notion of theoretical sampling
(Lincoln & Guba, 1985) in which the recruitment of participants stopped
when no new conceptions of the phenomenon under study emerged. In this
study, the variation in fear descriptions began to repeat itself after about 15
participants. All participants had been born and brought up in Cyprus and
were PhD holders or candidates. The interviews were conducted from 2011
until 2013. Participants were informed that participation was voluntary
and that they had the right to withdraw from the study. Anonymity was
promised to the interviewees. In the findings’ section, the participants were
referred to by number and length of service in years in parenthesis (e.g., P1,
3 refers to participant one, with 3 years of service), in order to help identify
potential differences in fear experiences and reactions between the younger
and senior academics.


Data were collected using semi-structured interviews conducted by the first

author and lasted between one to three and a half hours. The interviewees
were informed of the study’s objectives and approach, and were offered the
chance to see the interview transcript to check for correctness. All of the

interviews were face-to-face and were conducted in a private location at the

participants’ workplace. The aim of the interviews was to capture the possi-
ble variation in conceptions of fear in a rich and comprehensive way. The
interviews began with the following open-ended question: Can you bring to
mind a particular time when you felt fear at work and what that experience
meant to you? Tell it in as much detail as you can, as if you were reliving it
again. They were also asked to describe the emotion in order to identify
fear’s characteristics. Additional questions were asked to explore in more
detail the emotion, such as (a) the situational context of the experience,
(b) descriptions of the people with whom the experience was shared, (c) the
source of the event that stimulated fear, (d) the bodily symptoms experienced
and expressive reactions shown, and (e) how they attempted to deal with it.


Interviews were tape-recorded and transcribed word for word. Data were
analyzed using phenomenological analysis using Moustakas’ (1994)
approach to data analysis. The analysis consisted of the following phases:
first, familiarizing with the data with repeated readings of the transcripts as
it was important to understand the data and generate any descriptions
related to fearful aspects of the lecturers’ profession. Second, the descrip-
tions of the participants’ experiences of fear were coded and grouped into
coded categories by excerpting exemplars. Third, different categories were
sorted into potential themes, such as the characteristics of fear, its eliciting
events, and collated all the relevant coded data extracts within the identified
themes. The next step involved rereading exemplars from each category, and
identifying problematic themes or generating new ones (such as expressive
ways of fear) while later the themes were defined and labeled based on what
each theme was about. Then, interpretations of exemplars that disclosed the
meanings of participants’ fearful experiences were reflected in a concise,
coherent, logical account. Finally, to augment understandings of the col-
lected data, findings were further explored in relation to the literature.


From the interviews, three primary themes were identified. Each theme is
discussed in turn and presents a distinct but important understanding of
participants’ experiences of fear.
Understanding Lecturers’ Perceptions of Workplace Fear 63

Theme 1: A Social Emotion with Physical and Psychological Disruption

A dominant theme in participant accounts was that fear is an emotion of

physical and psychological disruption, which represented the anxiety and
worry expressed regarding certain social events and situations of unwanted
consequences. Lecturers described fear as an uncomfortable emotion,
unpleasantly experienced, and emphasized the impact of their actions on
others and how the display of their fear may have had harmful conse-
quences for their reputation. Their fear was experienced with physical
changes, such as shivering, palpitations, trembling voice, sweating, continu-
ous swallowing, and body temperature decrease. Psychological changes
were also mentioned such as sense of insecurity and self-doubt, agony, ner-
vousness, and feeling of uncertainty of the unknown. Fear was experienced
as feeling weak and needing relief, accompanied with feelings of vulnerabil-
ity, powerlessness, outpourings of crying, and agony of what will follow.
Participants remembered experiencing fear when lecturing, making them
think that they were not good enough to be lecturers, or that they were not
in the right job. Other participants described feeling frozen, exposed, hav-
ing butterflies in their stomach, and sweating when they are on stage.
For the lecturers, being fearful means having thoughts of self-doubt.
Academia entails many fearful moments and according to the interviews,
fear emphasizes weakness and a pressure to be unmistakable and perfect.
Participants referred to the social aspects of the emotion as, according to
them, experiencing fear creates thoughts about being professionally and
socially right. Descriptions about a loss of personal dignity and status, a
lack of confidence, a sense of failure to perform professionally and nervous-
ness were common. Central to the experience of fear is the insecurity and
agony that their reputation may be damaged. A male lecturer explained
what fear meant to him in relation to keeping his good reputation with his
students, because as he said students tend to see academics as people they
respect and not as people they make fun of:

I kept thinking that this is not where I want to be. The insecurity of the unknown is
fearful. I had to be there for two hours. I kept wondering, “What will I say for two
whole hours? How many slides is two hours? Will I remember everything? How will I
speak if I have 200 people looking at me? What will they think of me?” (P10, 4)

Fearing he would not complete an adequate and satisfying lecture, the

participant was filled with self-doubt, insecurity, and self-questioning on
whether he would respond to the professions’ duties. Opening an inner-
dialogue appeared as part of the experience of fear, with other lecturers

mentioning self-talking whilst being in the fearful situation of reputation

damage. Fear, then, appears to be an emotion that values the importance
of other people involved in the situation.

Theme 2: Determinants of Fear

The insecurity of losing their job as an outcome of the financial crisis, lack
of experience and confidence in lecturing, possible failure to deal with the
job’s expectations, and the authoritative behavior of managers were the
main determinants of the participants’ fear. They recognized the recent
financial recession as a key factor affecting lecturers’ psychological well-
being. Fear was linked with the potential loss of employment and with
uncertainty regarding career progression and promotional opportunities.
From the stories, it was clear that the HE industry had been subjected to
many difficulties over the last years, including the economic recession and
the privatization of a number of institutions. This generated descriptions of
fear about job security. A 35-year old female lecturer who recently started
her career in academia, having only two years working experience in her
current institution described how her fear was caused when her manager
informed her about a salary reduction:
It’s this damned recession. They cut and cut and cut salaries. But when she informed
me about a reduction of 300Euros I got scared. I couldn’t say anything … I was a
wreck. I mean, I do realise it wasn’t just me, but I got scared that this is just the begin-
ning and the next step would be to lose my job. (P13, 2)

The recession contributed to feelings of insecurity concerning the nature

and future existence of the academics’ jobs. For the participants, job inse-
curity was perceived as a significantly damaging experience, which
impacted their physical health, mental well-being, and generated job-
induced stress symptoms. Recurrent job insecurity can result in widespread
fear in organizational settings (Astrachan, 1995), with detrimental conse-
quences for the individual’s well-being. In contrast to actual job loss, job
insecurity refers to the anticipation of this stressful for the academics situa-
tion in such a way that the nature and continued existence of one’s job are
perceived to be at risk (Sverke & Hellgren, 2002). This implies that the feel-
ing of and reaction to job insecurity can be different among workers even if
they are exposed to the same objective situation.
Comments of failing to respond to others’ expectations and conse-
quently losing their job status and reputation were mentioned. When
Understanding Lecturers’ Perceptions of Workplace Fear 65

participants were asked to recall events from fearful experiences, a num-

ber of them remembered their first lecturing experience. Fear was specifi-
cally generated during the first lecturing sessions when they feared failing
to respond to their job and losing their credibility with the students.
Both young and senior academics suffered agonies of fear before every
big lecture and reflected on how “terrifying” and “frightening” it was to
deal with students and their questions. Participants remembered feeling
uncertain of the unknown, feeling unprepared to face the students’ ques-
tions during their lectures. Some participants attributed their fear not
only to their limited working experience, but also to their limited skills to
deal with students whose age was similar to theirs. A young and rela-
tively inexperienced female participant commented how she sees fear as a
problem in her job:
Self-confidence has been a major problem in my job. I have many problems with fear
of not being able to do all the things I need to do to perform the job. I found myself
struggling with classroom management, I have problems being organised and have
bouts of fear and depression at work. Whenever these hit me, I do not have the con-
fidence I need to control my classroom. I am now three years into this job, which
means eight months for each academic year, and I feel like I am not doing any
better. (P13, 2)

For some participants the small age difference with their students reg-
ularly generates fear for failure, which was attributed to their limited
experience. This limited work experience was discussed as a factor that
may generate fear, especially to the young participants who recognized
their fear as a result of their lack of experience to deal with large audi-
ences. The participants’ fear results from a threat of being confronted in
the “students’ eyes” as being unable to manage difficult situations, and
which in turn generates feelings of depression and anxiety. The initial
stages in an educator’s life do involve fear because of the unpredictability
and unknown nature of the job (Erb, 2002). New teachers may experi-
ence anxiety because of the complexity of learning to teach and the
uncertainty of achieving goals.
Another reason for the workers’ fear was related to the unwanted
social consequences of saying something that might disappoint other
people or that would not meet their demands, such as, complaining or
disagreeing to imposition of rules. Fear was described as being elicited by
a fear of speaking up and of being unprepared to deal with student and
managers’ queries, which brought fearful feelings of how other people
may react. This fear was, therefore, an adaptive reaction to hierarchy
issues (manager lecturer, student lecturer) which made them feel

powerless (Kemper, 1978; Plutchik, 1980). Participants described events

during which they felt afraid to express to their managers their thoughts
and concerns, consciously recognizing the negative consequences on their
image and professionalism. This is something that comes in alignment
with other studies, which reported fear as an important driver of silence
in organizations (Kish-Gephart et al., 2009; Milliken, Morrison, &
Hewlin, 2003). Moreover, managers’ insulting behavior generated stories
of fear of speaking up and defending the self. A 33-year old male senior
lecturer explained that fear of upsetting managers and of speaking up
was partly caused by his managers’ behavior and explained how he felt
when his manager allocated him for a demanding administrative position
against his will:

She (manager) treats you as if you are nobody. She believes that because she is the
Head you need to say “Yes” to everything, otherwise you have a problem. This dicta-
torial attitude makes me want to smash her into pieces and then step on them to make
sure she is vanished. But I cannot do any of these, so I keep these thoughts to myself.
(P10, 4)

Some lecturers clearly showed their desire to defend themselves from

their managers, but after considering the estimated costs of doing so they
often decided not to argue with them. Similarly, other participants told
stories of how they concealed any fear generated by their managers,
because their fear extended to repercussions resulting from the expression
of real emotions combined with the fear of seeming unprofessional. Fear of
speaking up to one’s boss is described primarily as a function of the boss’s
behavior or style (Ashford, Rothbard, & Dutton, 1998; Milliken et al.,
2003), which is confirmed by the above descriptions of the above manager’s
dictatorial attitude.
Fear of speaking up to one’s superior seems to have much broader and
more distal roots. Participants described a centralized educational system,
which they saw as reducing professional autonomy and promotion pro-
spects. Events were mostly centered on the managers’ leadership style, the
support, or lack of support, provided by a centralized, bureaucratic educa-
tional system, and the imposition of rules with minimal input from the aca-
demics. This bureaucratic liability stands in direct conflict with the
academics’ professional accountability as participants were disappointed
with their senior management and experienced feelings of insecurity of not
receiving acknowledgment and not being included in decision-making
Understanding Lecturers’ Perceptions of Workplace Fear 67

Theme 3: Coping with Fear using Appraisal-, Problem- and

Emotion-Focused Mechanisms

Participants discussed coping tactics, based on all the three basic mechan-
isms discussed by Weiten and Lloyd (2003), namely appraisal-, problem-,
and emotion-focused coping. Appraisal-focused coping occurred when the
lecturers attempted to modify the way they feel, by avoidance, withdrawal,
and distancing themselves from the fearful event. Specifically, participants
mentioned withdrawal as a common mechanism when they experience fear.
Withdrawal took the form of either psychologically withdrawing from a
fearful event or physically leaving the fearful event. In terms of physical
withdrawal, academics found it useful to immediately leave the University
to avert attention from the fear-related event. Participants also reported
staying away from the workplace for a period of time or working from
home in order to cope with fear and tension and to engage in amusing
activities outside work. Their justification for this was that they find it
impossible to relax in a “poisonous” and “intense” atmosphere, like their
Attempts to emotionally distance themselves from the event included reflec-
tion and engaging in non-work activities such as listening to music, reading,
fishing, and cycling. A number of participants reported seeking moments of
solitude, slowing down work pace, and taking time to reflect and reappraise
the fearful situations that occurred at work. This reflection time, which can be
spent while driving home, seemed helpful to emotionally distance them from
fearful events, to recognize the pros as well as cons of events and to get the
work feelings out of their system and calm down. This eventually helped
lecturers to relax and to create a positive and self-supportive mind set.
Problem-focused coping attempted to deal with the cause of the fear
with lecturers trying to resolve the problem by adopting new methods
that would help manage their fear. This category of coping appeared in
the form of improving time-management and of prioritizing. Most of the
participants indicated that they act proactively as a coping strategy to
deal with fear in teaching as they think it is an important component of
job performance (Crant, 2000). Planning and preparing, both practically
and emotionally, to prevent falling into pitfalls contributed to developing
a sense of control over managing the demands of the job, leading to
reduced fear, especially during lecturing. Length of professional experi-
ence appeared to impact on the academics’ emotional maturity, which in
turn affected the way they coped with fear. Proactive coping, such as

good time-management and prioritizing, has been shown to prevent or

lessen the impact of stress in university lecturers (Devonport, Biscomb, &
Lane, 2008) and also to facilitate the achievement of personal goals and
personal growth in the general occupational context (Greenglass, 2002).
Emotion-focused coping involved the participants’ attempt to display
more socially appropriate emotions or changing one’s own emotional reac-
tion. Emotion work was associated with fear, when events involved con-
trolling the experienced emotion and displaying a more socially
acceptable one, such as confidence. The participants revealed efforts to
fake, suppress, and change their authentic emotions to display a more pro-
fessional image to others. With this emotion regulation, they showed a con-
scious awareness that commoditizing their emotions is part of their job.
Controlling their emotions was stressful and emotionally demanding
which derived from emotional encounters with students and managers.
Participants recognized that their job role is very different from their pri-
vate self and that their emotions and emotional skills are used to keep other
people satisfied as part of their paid work. They discussed a perceived need
to leave aside their private self while at work and agreed that codes of emo-
tional display are essential. In many cases, participants felt the need to
show appropriate self-confidence and enthusiasm during the delivery of
their lectures in order to meet the students’ expectations. There were many
examples of participants describing how they felt when they needed to dis-
play certain emotions in reaction to their students’ questions, with all of
them pointing out the damaging effects that faking and the inability to
answer had on them.
Suppression of fear and showing appropriate understanding and accep-
tance of an event were recognized as necessary. Surface acting was evident
with the participants feeling the need to portray the image of a professional
and confident person in an attempt to control their fear during teaching.
Based on the participants’ stories, surface acting served an important pur-
pose in being professional at work. P11 (3) focused on the idea of profes-
sionalism by rationalizing his behavior:

I had to be confident. I had to feel it and I had to show it. If the students see the slight-
est fear, they will take advantage of it. We have to show courage, we have to display
the appropriate professionalism and expertise to remain a respectful person in class.

The use of surface acting was a way to change, instead of “hiding” fear.
The participants reported using personal self-talking and deep breathing as
ways to alter their fear in order to fit the culturally appropriate emotion.
For example, internal dialogue works as a tactic to change the state of fear
Understanding Lecturers’ Perceptions of Workplace Fear 69

when students become demanding and continuously ask challenging ques-

tions. As P16 (13) commented, “I re-assure myself that everything will be
okay and must show confidence rather than express fear because it is part
of the job.” In the participants’ descriptions, there was a feeling that sur-
face and deep acting techniques were important when dealing with the chal-
lenges of the profession. This was evidenced by stories, in which they
described their strategy in moments of fear during lecturing:
I started talking to myself, saying “Relax, it’s ok to be scared but let it go.” I was talk-
ing to myself and it’s something I still do. When I am nervous or scared, before going
to teaching, I think, “You are fine. Take some deep breaths, it’s normal to feel this
way, but you can handle this.” And for some magic reason, it works! (P5, 6)


This research investigated workplace experiences of fear within Cypriot

HE, the situations that give rise to fear and the ways in which lecturers
cope with their fear. The study contributes to the literature on organiza-
tional emotions, responding to calls for more research into the nature of
discrete emotions (e.g., Gooty et al., 2009) by offering insights into fear’s
own storyline. The study also offers a new way of studying emotions by
shifting the existing focus on Western spaces and conducting research in
other European workplaces, the Cypriot academia. Learning from the
Cypriot workplace offered opportunities for understanding fearful experi-
ences across national and organizational boundaries.
The findings provide evidence of what fear means to Cypriot lecturers
and how they experience it in the workplace. Fear can manifest itself in
relation to potential loss of employment, personal failure, loss of status and
reputation, and discourages speaking up when dealing with powerful stake-
holders. Fear was confirmed by the participants as an existential emotion
(Lazarus & Lazarus, 1994) because the threats it is based on are concerned
with meanings about who they were and the quality of their existence in
their workplace. It is an emotion that involves the subject’s reputation,
together with the recognition placed on others’ judgments. Fear was
described as a psychological and physiological state of disturbance, which
has numerous personal impacts, such as lack of sleep, depression, feelings
of vulnerability, self-doubt, insecurity, and worry. Several pressures and
challenges of the nature of the job, which were the personal meaning that
aroused their fear, undermined the participants’ security, reputation, and

autonomy. Lecturers constructed these meanings out of their experience

and the values of the culture in which they live in and are committed to
preserving them (Gendlin, 1962). In the stories, key areas of perceived
threat were the academics’ job security, academic status and respected
reputation, and autonomy in speaking up, which led to perceptions of fail-
ure and serious doubts about their own adequacy. This personal meaning
made the participants experience fear as a silent inner struggle, which even
though their fear was sometimes suppressed it still made them feel threa-
tened (Lazarus & Lazarus, 1994).
The situations most likely to cause fear at work involved feelings of
exposure during lectures, job loss, fear of failure and of reputation loss,
and of speaking up. Interestingly, these situations were linked only to
teaching and to the general nature of the profession and not related to any
research outcomes, something that confirms the Cypriot tertiary institu-
tions as largely teaching-oriented (Varnava-Marouchou, 2007). Participants
viewed their fear a primary emotion of their job. A common form of fear
was worry about loss of employment with the institution, which was influ-
enced by the financial crisis in 2008, confirming the large impact of employ-
ment insecurity on well-being (e.g., Dickerson & Green, 2012; Wichert,
2005). The emotion of fear in teaching was previously mentioned as related
with job insecurity (Waller, 1932).
Fear was mainly generated during their first lecturing sessions and at the
beginning of an academic year when the audience is still unknown. Both
young and senior academics reflected on how fearful their lecturing experi-
ences can be and the majority of them shared how frightening it was to
face the students and to deal with their questions. They attributed their
fear not only to their limited experience as lecturers, but also to their lim-
ited communication skills to deal with students whose age was similar to
theirs. They admitted that the small age difference they have with some stu-
dents regularly generated fear for failure as they found themselves unable
to respond to their challenging questions. The study lends support to
research that the initial stages in an educator’s life involve fear and anxiety
because of the unpredictability and unknown nature of the job; hence, the
emotional world of beginning teachers was regarded as a “whirlpool,”
because it is never still (Erb, 2002, p. 1). What is also interesting, is that
participants mentioned that training was often not provided for lecturers,
since PhDs provide research training, but teaching qualifications were
rarely compulsory and even when offered were taken during rather than
before starting to teach.
Understanding Lecturers’ Perceptions of Workplace Fear 71

Senior management affected the participants’ psychological well-being.

It was evident from the stories that the participants were disappointed with
the way in which management treated them; this disappointment gave rise
to feelings of feeling unappreciated, which in turn resulted to fear of speak-
ing up. Empirical studies found that workers frequently remain silent in
situations that call for voice, like employee treatment and managerial beha-
vior (Bowen & Blackmon, 2003; Milliken et al., 2003). Workers often hesi-
tate to speak up or raise an issue of concern to their managers because of
fear of repercussion and of experiencing unwanted social consequences for
saying something that might disappoint others (Milliken et al., 2003; Ryan
& Oestreich, 1991). Likewise, the potential benefits of employee participa-
tion in various organizational processes support that organizational partici-
pation plays a major role in reducing fear generated by organizational
The interviews indicated that lecturers controlled their fear with all three
broad types of coping strategies, namely appraisal-, problem- and emotion-
focused (Weiten & Lloyd, 2003). An important factor offsetting fear at
work was the degree of preparation that allowed lecturers to respond to the
challenges of teaching. Participants referred to proactive strategies to emo-
tionally and practically prepare themselves for the demands of the job and
to act in a professional, non-fearful manner. As a powerful, evolutionary-
based emotion, fear also encouraged distancing, withdrawal, and avoidance
behavior, which entailed a cognitive focus on perceived threats, and pessi-
mistic judgments about risks and future outcomes (e.g., Izard, 1993).
Participants highlighted their need to seek different spheres to “escape”
from their fears at work and to get involved in more relaxing activities.
Participants also showed a conscious awareness that commoditizing their
emotions was part of their job, and claimed to suppress emotional displays
that would be considered inappropriate; this, in turn, affected their emo-
tional experiences of fear. Evidently, the lecturers’ priority was to perform
emotion work to ensure that their authentic emotion was suppressed or
presented in a socially acceptable way (Hochschild, 1983). They all dis-
cussed the impact of having to regulate their fear and exhibit false emotions
to other people in their attempt to manage the emotion and its conse-
quences and to meet social norms. However, controlling their fear to show
a professional image to others was stressful and emotionally demanding.
Most lecturers made use of surface and deep acting (Hochschild, 1983) as a
means of regulating their emotional experiences. A number of participants
felt that there are unwritten guidelines directing their emotional expressions

and experiences, and therefore they made extraordinary efforts to portray

socially, occupationally, and organizationally expected images, thus illus-
trating their emotional laboring (Ogbonna & Harris, 2004).
Participants seemed to rely on their personal emotional skills to carry out
their job well. Confidence was an emotion that academics deliberately choose
to show in order to gain control during teaching or to diminish their fear and
shame. Emotion regulation, as the hiding of unacceptable emotional impulses
from public view, was evident in the stories, as participants demonstrated
their own social beliefs about their role in the University and the expectations
the students have from them in that role (Averill, 1980). Also, surface acting,
meaning the compliance with accepted display rules by displaying emotions
that are not actually felt (Hochschild, 1983), was described by most of the
lecturers in an effort to control their fear and display their confidence during
teaching or responding to difficult questions. In addition, in an emotionally
intelligent way, participants referred to fearful events they experienced with
students and tried to remember successful ways that helped them in the past.
This led them to think of and adopt scripts that they used in similar situations
in their attempt to have a positive impact on their students, something that
was related to high emotional intelligence (Lindebaum & Jordan, 2012).


Evidence from the study showed that fear at work is an increasing concern
in today’s Cypriot HE sector. Closer attention is needed to the climate of
employee fear and its impact on workers’ well-being, as the participants
clearly perceive this to be deteriorating significantly. Fear of employment
loss, fear of failure, job status and reputation damage, and fear of speaking
up were all implicated as etiological factors for workplace fear, leading to
mostly damaging psychological and physiological consequences. However,
when participants adopted mechanisms that gave them a sense of confi-
dence and autonomy at work, their fear was less prevalent. This suggests
that developing strategies that enhance employee security, confidence, and
participation, especially at times of economic recession, could be the key
for beneficial outcomes and satisfaction at work. The management’s practi-
cal contribution is important, because refined understanding of the sources,
nature, and expressions of fear of speaking up is essential to guide those in
power who seek to improve their organizations via routine upward input
from employees at all levels. In addition, the role of management in helping
Understanding Lecturers’ Perceptions of Workplace Fear 73

lecturers limit the experience of fear and develop the skills needed, by pro-
viding opportunities for teaching courses, or by providing ongoing mentor-
ing seems essential considering the fearful nature that this aspect of the job
Implications for research suggest that utilizing an interpretive approach,
as a new way of studying the emotion of fear, gave the opportunity
to identify important areas of the emotion’s nature in the Cypriot acade-
mia. It is argued that a more in-depth insight of workplace specific
emotions would help to create a “portrait” of other emotions in more


Some limitations of the research must also be recognized. An obvious lim-

itation is that the study’s findings are context-related, since the study was
conducted amongst one professional group in one country, so one should
be cautious when generalizing to other contexts. The small sample size also
limits the ability to generalize the findings from this study, as there is an
evident risk that the findings and participants’ personal stories are not
necessarily transferable to other occupations and contexts. However, the
purpose of this study was not to obtain knowledge that can be generalized.
Indeed, phenomenologists would argue that such an aim would be proble-
matic given the subjective nature of emotion. Thus, this chapter works
toward developing understanding about a single emotion in a specific con-
text; it provides a voice to a group of people with limited attention in the
published emotion literature. The interpretive approach encouraged the
participants to reflect on and construct personal stories influenced by their
workplace interactions; however, this type of inquiry is retrospective and
dependent on the participants’ recollection. Therefore, we recognize that
this reflection and meaning-making approach produced findings that would
perhaps be different from a longitudinal approach, for example, and which
may provide an even deeper understanding of fear. It is nevertheless note-
worthy to recognize that in any form of research, some choices leave out
the possibility of other meanings or angles that could have been uncovered.
Although objectivity and generalizability are neither attainable, nor parti-
cularly desirable in interpretive investigations, it is important for the
reader(s) to be able to ground this work in the context of the researcher’s
influences and views of the world.

The fact that the interviews were conducted in Greek and not English
should also be recognized, as translation from one language to another can
distort the meaning of the participant’s original expression and this could
be culturally related and difficult to translate. The authors recognize that
the way the data were presented reflect their own understanding of the texts
and their understanding of the context in which events took place in the
world of academics. It is important, therefore, to recognize that the three
main themes identified in the study represent only one interpretation of the
data. In interpretive research, a single text is indeed open to many different
interpretations because the meaning of a text is usually the outcome of the
fusion of the horizons of the text and interpreter (Moustakas, 1994). Space
did not allow for a discussion of all the data, and although a large amount
of the interview quotes were presented, it was not possible to use all the
text that was produced. However, the quotes were not extracted from their
context, in the hope that the reader would get a flavor of the original text
and have the opportunity to judge the authors’ interpretation.


Ahmed, S. (2003). The politics of fear in the making of worlds. International Journal of
Qualitative Studies in Education, 16(3), 377 398.
Aristotle. (1984). In J. Barnes (Trans. & Ed.), The complete works of Aristotle: The revised
Oxford translation (Vol. 2). Bollingen Series. Oxford: Princeton University Press.
Ashford, S. J., Rothbard, N. P., & Dutton, J. E. (1998). Out on a limb: The role of context
and impression management in selling gender-equity issues Sandy Kristin Piderit.
Administrative Science Quarterly, 43, 23 57.
Ashforth, B. E., & Humphrey, R. H. (1993). Emotional labor in service roles: The influence of
identity. Academy of Management Review, 18(1), 80 115.
Ashkanasy, N. M., & Nickolson, G. J. (2003). Climate of fear in organisational settings:
Construct definition, measurement and a test of theory. Australian Journal of
Psychology, 55(1), 24 29.
Astrachan, J. H. (1995). Organizational departures: The impact of separation anxiety as stu-
died in a mergers and acquisitions simulation. The Journal of Applied Behavioral
Science, 31(1), 31 50.
Averill, J. R. (1980). A constructivist view of emotion. In R. Plutchik & H. Kellerman (Eds.),
Theories of emotion (Vol. 1, pp. 305 339). Orlando, FL: Academic Press.
Barbalet, J. M. (1998). Emotion, social theory, and social structure: A macrosociological
approach. Cambridge: Cambridge University Press.
Barclay, L. J., Skarlicki, D. P., & Pugh, S. D. (2005). Exploring the role of emotions in injus-
tice perceptions and retaliation. The Journal of Applied Psychology, 90, 629 643.
Basch, J., & Fisher, C. D. (2000). Affective events-emotions matrix: A classification of work
events and associated emotions. In N. M. Ashkanasy, C. E. J. Hartel, & W. J. Zerbe
Understanding Lecturers’ Perceptions of Workplace Fear 75

(Eds.), Emotions in the workplace: Research, theory, and practice. (pp. 36 48).
Westport: Greenwood Publishing Group.
Bellas, M. L. (1999). Emotional labor in academia: The case of professors. The ANNALS of
the American Academy of Political and Social Science, 561(1), 96 110.
Boudens, C. J. (2005). The story of work: A narrative analysis of workplace emotion.
Organization Studies, 26(9), 1285 1306.
Bowen, F., & Blackmon, K. (2003). Spirals of silence: The dynamic effects of diversity on
organizational voice. Journal of Management Studies, 40(6), 1393 1417.
Bradley, G., & Russell, G. (1997). Computer experience, school support and computer anxi-
eties. Educational Psychology, 17(3), 267 284.
Buczynski, S., & Hansen, C. B. (2010). Impact of professional development on teacher prac-
tice: Uncovering connections. Teaching and Teacher Education, 26(3), 599 607.
Burchell, B., Ladipo, D., & Wilkinson, F. (2002). Job insecurity and work intensification.
Routledge studies in employment relations. London: Routledge.
Cochran, J. L., & Cochran, N. H. (2005). The heart of counselling: A guide to developing thera-
peutic relationships. Belmont, CA: Wadsworth.
Conley, S., & Glasman, N. S. (2008). Fear, the school organization, and teacher evaluation.
Educational Policy, 22(1), 63 85.
Constanti, P., & Gibbs, P. (2004). Higher education teachers and emotional labour.
International Journal of Educational Management, 18(4), 243 249.
Crant, J. M. (2000). Proactive behaviour in organizations. Journal of Management, 26(3),
435 462.
Darwin, C. (1872). The expression of the emotions in man and animals. London: HarperCollins.
De Lara, P. Z. M. (2006). Fear in organizations: Does intimidation by formal punishment
mediate the relationship between interactional justice and workplace internet deviance?
Journal of Managerial Psychology, 21(6), 580 592.
De Witte, H. (2005). Job insecurity: Review of the international literature on definitions, pre-
valence, antecedents and consequences. SA Journal of Industrial Psychology, 31(4),
1 6.
Devonport, T., Biscomb, K., & Lane, A. M. (2008). Sources of stress and use of anticipatory,
preventative and proactive coping strategies by higher education lecturers. The Journal
of Hospitality, Leisure, Sport and Tourism, 7(1), 70 81.
Dickerson, A., & Green, F. (2012). Fears and realisations of employment insecurity. Labour
Economics, 19(2), 198 210.
Domagalski, T. A., & Steelman, L. A. (2005). The impact of work events and disposition on the
experience and expression of employee anger. Organizational Analysis, 13(1), 31 52.
Egan, G. (2007). The skilled helper: A problem-management and opportunity-development
approach to helping (8th ed.). Belmont, CA: Thomson/Brooks/Cole.
Ekman, P. (1992). An argument for basic emotions. Cognition & Emotion, 6(3 4), 169 200.
Erb, C. S. (2002). The emotional whirlpool of beginning teachers’ work. In paper presented at
the annual meeting of the Canadian Society of Studies in Education, Toronto, Canada.
Fineman, S. (1993). Organizations as emotional arenas. In S. Fineman (Ed.), Emotion in orga-
nizations (pp. 9 35). London: Sage.
Fischer, K. W., Shaver, P. R., & Carnochan, P. (1990). How emotions develop and how they
organise development. Cognition & Emotion, 4(2), 81 127.
Fisher, C. D., & Ashkanasy, N. M. (2000). The emerging role of emotions in work life: An
introduction. Journal of Organizational Behavior, 21(2), 123 129.

Fitness, J. (2000). Anger in the workplace: An emotion script approach to anger episodes
between workers and their superiors, co-workers and subordinates. Journal of
Organizational Behavior, 21(2), 147 162.
Frijda, N. H. (2007). What might emotions be? Comments on the comments. Social Science
Information, 46(3), 433 443.
Funlop, L., & Rifin, W. D. (1997). Representing fear in learning in organizations.
Management Learning, 28(1), 45 63.
Game, A. (2008). Negative emotions in supervisory relationships: The role of relational
models. Human Relations, 61(3), 355 393.
Geddes, D., & Callister, R. R. (2007). Crossing the line(S): A dual threshold model of anger in
organizations. Academy of Management Review, 32(3), 721 746.
Gendlin, E. T. (1962). Experiencing and the creation of meaning: A philosophical and psycholo-
gical approach to the subjective. Illinois, IL: Northwestern University Press.
Gibson, D. E. (2006). Emotional episodes at work: An experiential exercise in feeling and
expressing emotions. Journal of Management Education, 30(3), 477 500.
Glomb, T. M. (2002). Workplace anger and aggression: Informing conceptual models
with data from specific encounters. Journal of Occupational Health Psychology, 7,
20 36.
Goleman, D. (1995). Emotional intelligence: Why it can matter more than IQ. New York, NY:
Bantam Books.
Goos, M., & Hughes, C. (2010). An investigation of the confidence levels of course/subject
coordinators in undertaking aspects of their assessment responsibilities. Assessment &
Evaluation in Higher Education, 35(3), 315 324.
Gooty, J., Gavin, M., & Ashkanasy, N. M. (2009). Emotions research in OB: The challenges
that lie ahead. Journal of Organizational Behavior, 30, 833 838.
Grandey, A. A. (2003). When “the show must go on”: Surface acting and deep acting as deter-
minants of emotional exhaustion and peer-rated service delivery. Academy of
Management Journal, 46(1), 86 96.
Gray, J. A. (1988). The psychology of fear and stress (2nd ed.). Cambridge: Cambridge
University Press.
Greenglass, E. R. (2002). Proactive coping and quality of life management. In E. Frydenberg
(Ed.), Beyond coping: Meeting goals, visions, and challenges (pp. 37 62). London:
Oxford University Press.
Hargreaves, A. (1998). The emotional practice of teaching. Teaching and Teacher Education,
14(8), 835 854.
Heidegger, M. (1962). Being and Time. New York, NY: Harper & Row.
Hochschild, A. (1983). The managed heart: Commercialization of human feeling. Berkley, CA:
University of California Press.
Hoffman, K. D., & Kretovics, M. A. (2004). Students as partial employees: A metaphor for
the student-institution interaction. Innovative Higher Education, 29(2), 103 120.
Izard, C. E. (1993). Four systems for emotion activation: Cognitive and noncognitive
processes. Psychological Review, 100, 68 90.
James, W. (1950). The principles of psychology (Vol. 2). New York, NY: Dover.
Jordan, P. J., Ashkanasy, N. M., Hartel, C. E. J., & Hooper, G. S. (2002). Workgroup emo-
tional intelligence. Scale development and relationship to team process effectiveness
and goal focus. Human Resource Management Review, 12(2), 195 214.
Understanding Lecturers’ Perceptions of Workplace Fear 77

Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D., & Rosenthal, R. A. (1964).
Organisational stress: Studies in role conflict and ambiguity. New York, NY: Wiley-
Kemper, T. D. (1978). A social interactional theory of emotions. New York, NY: Wiley-
Kemper, T. D. (1987). How many emotions are there?: Wedding the social and autonomic
components. American Journal of Sociology, 93, 263 289.
Kirk, P. (1999). Understanding emotions the key to preventing feelings of alienation. British
Journal of Healthcare Management, 5(2), 76 77.
Kish-Gephart, J. J., Detert, J. R., Treviño, L. K., & Edmondson, A. C. (2009). Silenced by
fear: The nature, sources, and consequences of fear at work. Research in Organizational
Behavior, 29, 163 193.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY: Springer
Publishing Company.
Lazarus, R. S., & Lazarus, B. N. (1994). Passion and reason: Making sense of our emotions.
New York, NY: Oxford University Press.
Lerner, J. S., & Keltner, D. (2001). Fear, anger, and risk. Journal of Personality and Social
Psychology, 81, 146 159.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Newbury Park, CA: Sage.
Lindebaum, D., & Jordan, P. J. (2012). Positive emotions, negative emotions, or utility of dis-
crete emotions? Journal of Organizational Behavior, 33(7), 1027 1030.
Maner, J. K., & Gerend, M. A. (2007). Motivationally selective risk judgments: Do fear and
curiosity boost the boons or the banes? Organizational Behavior and Human Decision
Processes, 103, 256 267.
McKinney, K. (1988). Faces: Five components of quality teaching. Teaching Sociology, 16(3),
298 301.
Miller, V. D., & Jablin, F. M. (1991). Information seeking during organizational entry:
Influences, tactics, and a model of the process. Academy of Management Review, 16(1),
92 120.
Milliken, F. J., Morrison, E. W., & Hewlin, P. F. (2003). An exploratory study of employee
silence: Issues that employees don’t communicate upward and why. Journal of
Management Studies, 40(6), 1453 1476.
Ministry of Education and Culture. (2005). The private universities (establishment, operation
and control) law 109(1) of 2005. Retrieved from http://www.ecpu.ac.cy/en/legislation.
html. Accessed on November 17, 2014.
Ministry of Finance. (2013). Number of students in tertiary education in Cyprus and abroad
1985/86 2011/12. Retrieved from http://www.mof.gov.cy/mof/cystat/statistics.nsf/
sel=2. Accessed on November 17, 2014.
Morrison, E. W. (1993). Longitudinal study of the effects of information seeking on newcomer
socialization. Journal of Applied Psychology, 78(2), 173 183.
Moustakas, C. (1994). Phenomenological research methods. London: Sage.
Namie, G. (2003). Workplace bullying: Escalated incivility. Ivey Business Journal, 68(2), 1 6.
Nembhard, I. M., & Edmondson, A. C. (2006). Making it safe: The effects of leader inclusive-
ness and professional status on psychological safety and improvement efforts in health
care teams. Journal of Organizational Behavior, 27(7), 941 966.

Oatley, K., Dacher, K., & Jenkins, J. (1996). Understanding emotions. Cambridge: Blackwell
Ogbonna, E., & Harris, L. C. (2004). Work intensification and emotional labour among UK
university lecturers: An exploratory study. Organization Studies, 25(7), 1185 1203.
O’Neill, S., & Stephenson, J. (2012). Does classroom management coursework influence pre-
service teachers’ perceived preparedness or confidence? Teaching and Teacher
Education, 28(8), 1131 1143.
Pekrun, R., Goetz, T., Titz, W., & Perry, R. P. (2002). Academic emotions in students’ self-
regulated learning and achievement: A program of qualitative and quantitative,
research. Educational Psychology Review, 37(2), 91 106.
Plutchik, R. (1980). A general psychoevolutionary theory of emotion. Emotion: Theory,
Research, and Experience, 1(3), 3 33.
Rachman, S. (1974). The meanings of fear. Harmondsworth: Penguin.
Ragins, B. R., Singh, R., & Cornwell, J. M. (2007). Making the invisible visible: Fear and dis-
closure of sexual orientation at work. Journal of Applied Psychology, 92(4), 1103 1118.
Ramsden, P. (1992). Learning to teach in higher education. London: Routledge.
Ryan, K. D., & Oestreich, D. K. (1991). Driving fear out of the workplace: How to overcome
the invisible barriers to quality, productivity and innovation. San Francisco, CA: Jossey-
Bass Publishers.
Sandberg, J. (2005). How do we justify knowledge produced within interpretive approaches?
Organizational Research Methods, 8(1), 41 68.
Shaver, P., Schwartz, J., Kirson, D., & O’Connor, C. (1987). Emotion knowledge: Further
exploration of a prototype approach. Journal of Personality and Social Psychology,
52(6), 1061 1086.
Stearns, P. N. (2006). American fear: The causes and consequences of high anxiety. New York,
NY: Routledge.
Sverke, M., & Hellgren, J. (2002). The nature of job insecurity: Understanding employment
uncertainty on the brink of a new millennium. Applied Psychology, 51(1), 23 42.
Tiedens, L. Z. (2001). Anger and advancement versus sadness and subjugation: The effect of
negative emotion expressions on social status conferral. Journal of Personality and
Social Psychology, 80(1), 86 94.
Tiedens, L. Z., & Linton, S. (2001). Judgment under emotional certainty and uncertainty: The
effects of specific emotions on information processing. Journal of Personality and Social
Psychology, 81(6), 973 988.
Torkelson, E., & Muhonen, T. (2003). Coping strategies and health symptoms among women
and men in downsizing organisation. Psychological Reports, 92(3), 899 907.
Turner, J. H. (2000). On the origins of human emotions: A sociological inquiry into the evolution
of human affect. Stanford, CA: Stanford University Press.
Varnava-Marouchou, D. (2007). Teaching and learning in an undergraduate business context:
An inquiry into lecturers’ conceptions of teaching and their students’ conceptions of learn-
ing. Nottingham: University of Nottingham.
Vecchio, R. P. (2000). Negative emotion in the workplace: Employee jealousy and envy.
International Journal of Stress Management, 7(3), 161 179.
Vince, R. (2006). Being taken over: Managers’ emotions and rationalizations during a com-
pany takeover. Journal of Management Studies, 43(2), 343 365.
Waller, W. (1932). The sociology of teaching. New York, NY: Wiley.
Watson, D. (2000). Mood and temperament. New York, NY: Guilford Press.
Understanding Lecturers’ Perceptions of Workplace Fear 79

Weick, K. E. (1995). Sensemaking in organizations. California, CA: Sage.

Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion of the
structure, causes and consequences of affective experiences at work. In B. M. Staw &
L. L. Cummings (Eds.), Research in organizational behaviour (Vol. 19, pp. 1 74).
Greenwich, CT: JAI Press.
Weiten, W., & Lloyd, M. A. (2003). Psychology applied to modern life: Adjustment in the 21st
century (7th ed.). London: Wadsworth.
Wichert, I. (2005). Job insecurity and work intensification: The effects on health and well-
being. In B. Burchell, D. Ladipo, & F. Wilkinson (Eds.), Job insecurity and work intensi-
fication (pp. 92 111). London: Routledge.
Willmott, H. (1995). Managing the academics: Commodification and control in the develop-
ment of university education in the U.K. Human Relations, 48(9), 993 1027.
Wouters, C. (1989). The sociology of emotions and flight attendants: Hochschild’s managed
heart. Theory, Culture and Society, 6(1), 95 123.
Zembylas, M., & Papanastasiou, E. (2006). Sources of teacher job satisfaction and dissatisfac-
tion in Cyprus. Compare: A Journal of Comparative and International Education, 36(2),
229 247.
Zwagerman, S. (2008). The scarlet P: Plagiarism, panopticism, and the rhetoric of academic
integrity. College Composition and Communication, 59(4), 676 710.
This page intentionally left blank



Markus Plate


Shame is a common, yet seldom acknowledged emotion. Shame signals a

threatened social bond in which the claim of as what one wants to be
seen (i.e., the claim for a certain relational identity and relative status
positioning) is neglected by the other party. Using a case study
approach, this chapter provides insights into how shame shapes the rela-
tionship and leadership structure in organizations. The case used is based
on a documentary TV show; hence this chapter also provides insight in
the use of visual/TV material to gain insight in relational leadership
Keywords: Shame; face; leadership; relationship; reality TV

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 81 107
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011005


Shame and it’s weaker cousin embarrassment are quintessential human

emotions that regulate everyday social interaction (Barrett, 1995), both on
the micro level of individual human interaction as well as on the macro
level of society (Goffman, 1967; Scheff, 2000a, 2000b). In its essence,
shame is both a self-conscious and social emotion. On the psychological
level, it signals that your self (who you are, your personal identity) is
faulty, worthless or wrong. On the social level, it signals that your self-in-
relationship (who you claim to be in relation to the other person, your
social identity) is faulty, as it is not accepted and acknowledged by the
other party. Despite its importance, shame seems to be vanishing from
open discourse in modern Western societies (Scheff, 2014) and subse-
quently finds rare treatment in organizational, management or leadership
studies. The lack of research is surprising, though, as one would expect
more scholarly interest in a factor that is at the center of regulating human
interaction and thus influences management and leader effectiveness and
business performance.
In a recent review, though, Creed, Hudson, Okhuysen, and Smith-
Crowe (2014) acknowledged the influence of shame in organizational pro-
cesses by providing a framework that highlights the interplay of individu-
ally felt shame, the disciplinary power of shaming and the influence of
shared understandings of shameful behavior. Creed et al. (2014) emphasize
the importance of shame for a recent discourse in institutional theory which
aims at uncovering the micro foundations of institutions such as interac-
tions and sensemaking (Powell & Colyvas, 2008). A similar discourse can
be found in leadership theory, in which the interactional and relational
foundations of leadership are under investigation (Fairhurst & Uhl-Bien,
2012; Graen & Uhl-Bien, 1995; Uhl-Bien, 2006). In this tradition, Sauer
and Ropo (2006) investigate the role of shame in leadership in the context
of theater and report that although shame is experienced as aversive, it can
encourage actors to work outside the normal way of acting and thus pro-
vide beneficial effects. However, shame seems to be a rather toxic emotion
that often goes along with conflict. For example, Crowley (2014) empha-
sizes the negative effects of shame in relation to coercive measures of orga-
nizational control in manual labor and service sector firms. According to
Crowley (2014, p. 430), these measures “promote[s] abuse that intensifies
shame, generates hostility toward management, and contributes to conflict
among coworkers” and as such have a detrimental effect on leadership and
Shame and the Undoing of Leadership 83

This chapter investigates the interplay of shame and leadership by look-

ing at the level of interaction and communication. The investigation is based
on a case study of a lace-dying factory in England and aims to answer the
question: Which role does shame play in establishing a leader follower
The case is based on an uncommon data source, namely visual material
based on a reality TV show. As a general trend, the use of visual materials
in organization and management research gained popularity in recent years
(Bell & Davison, 2013). Especially movies and TV provide the opportunity
for unique insights into otherwise neglected topics: “While business school
curricula, traditional case studies and textbooks tend to emphasize the
rationality and order associated with organization, film draws attention to
the embodied, personal and emotional nature of organizational life” (Bell,
2008, p. 1). The case in question thus provides a unique opportunity to gen-
erate interesting insight into the neglected topics of shame in organizations,
especially leadership.
The aim of this chapter is thus twofold: It provides an introduction
about the use of TV/movie material for research on emotional issues in
organizations, as well as an insight into the function of shame and its rela-
tionship to leadership.
The reminder of this chapter is structured as follows. In the next section,
an overview on movies as data source for emotional and organizational
phenomena is provided. “Analysis of Emotions in Visual Material” shows
examples for research on emotions that makes use of movies and TV mate-
rial. “A Primer on Shame, Identity and Relationship” establishes the theo-
retical framework used in the case analysis. The analytical strategy for the
case is explained in “Method of Analysis of the Case Study”, followed by
an in-depth case reconstruction in “How Shame Undermines Leadership
Insights from a Case Study.” “Discussion and Conclusion” briefly sum-
marizes and discusses the findings.



Scholars have used a great variety of movies as inspiration for generating

interesting insight into organizational and institutional phenomena. On one
side of the spectrum is the purely fictitious like the TV crime series “The
Wire” (Holt & Zundel, 2014; Zundel, Holt, & Cornelissen, 2013), feature

films like “12 Angry Men” and “13 Days” (Buchanan & Huczynski, 2004),
the movie drama “Thelma and Louise” (Ashford, 2013), the British sitcom
“Only Fools and Horses” (Infante, 2011) or Italian, Swedish and South
African soap operas (Czarniawska, Eriksson-Zetterquist, & Renemark,
2013). The analysis of these movies and shows provides insights into insti-
tutional work and organizational patterns and dynamics (Zundel et al.,
2013), reveals “taken-for-granted assumptions about women and work that
are firmly embedded in the public discourse” (Czarniawska et al., 2013,
p. 267), exhibits the connection of work-family discourse and gender roles
(Infante, 2011), or provides inspiration for scholarly writing (Ashford,
2013). In general, Hassard and Buchanan (2011, p. 620) claim that feature
films can be “analyzed as research data in the form of process narratives
‘proxy documentaries’, and case studies to inform our understanding of
organizational behavior and to develop theory.”
On the other side of the spectrum are observational documentaries,
visual ethnographies or cinema ve´rite´ (Hassard, 2011). One remarkable
example is the documentary “Corporation: After Mr. Sam” which depicts a
series of board meetings in which the succession of the CEO of Steinberg’s
Supermarket (at one point in time the largest supermarket chain in the
Canadian province of Quebec) is discussed. The documentary has been
intensely studied from many perspectives amongst others, from an emo-
tional perspective about emotional display rules in organizations and the
function of emotions in “rational” business negotiations (Fitch & Foley,
2007; Putnam, 2007; Tracy, 2007).
These extremes cinema ve´rite´ on the one hand and pure fiction on the
other hand open up a continuum. Movies in the tradition of cinema ve´r-
ite´ don’t interact or interfere with the happenings that are filmed (except
the influential fact of the filming per se). In general, the director tries to dis-
play the happenings as close to reality as possible. On the other side, the
happenings of pure fiction are under complete control of the writer and
director. In between these extremes are genres like TV game shows and rea-
lity TV. Both genres show non-scripted action and interaction of “real”
people in unique situations. Both genres can be considered as suitable
sources for academic research on emotions and/or organizational phenom-
ena. Examples for research on game shows are Scheff and Retzinger
(2001b), who analyzed episodes of “Candid Camera” and two US TV game
shows for occurrences of face-loss and shame; or Culpeper (2005), who
provide a micro-interactional account on face-loss and impoliteness on
“The Weakest Link.” As this chapter focuses on reality TV as a source for
academic interrogation, this genre is elaborated upon further below.
Shame and the Undoing of Leadership 85

Reality TV

In general, reality TV is a genre where “the unscripted behavior of ‘ordin-

ary people’ is the focus of interest” (Bignell, 2013, p. 310). In reality TV,
the real behavior of ordinary people is not only sought to be described,
though, but the aim is also to “explain […], especially through the form
of narrative” (Bignell, 2005, p. 62). One subgenre of reality TV is the
“lifestyle makeover show,” in which the participants receive advice and
help by experts in transforming a bad situation into a better one be it
about homes, fitness, parenting, nasty pets (Lewis, 2008) or failing busi-
nesses. For this chapter, I focus on the British makeover series “I’ll show
them who is boss” (hereafter ISTWIB) in which an experienced business
man (the “consultant”) visits struggling businesses in order to provide
advice and save the business. In each episode of ISTWIB, the business
man Gerry Robinson visits a failing enterprise, talks with manage-
ment, owners and staff, comments on the unfolding situation, offers diag-
nosis on the problem as well as advice to solve it. Usually, these kind of
shows combine formulaic and spontaneous elements. The formulaic ele-
ments arises due to a general scripted plot that can best be summarized
as “Gerry Robinson tells the business people how to fix the firm.” Several
types of scenes occur: The consultant comments on previously filmed
scenes of the business, visits the business, and comments on it, intervenes
and comments on the intervention. These kind of scenes are part of the
overall plot that allows the “fixing” narrative to occur. Nevertheless, each
episode also contains scenes that depict unique characters (managers,
staff, family owners, etc.) and their spontaneous behavior. These scenes
show interactions between business members and between business mem-
bers and the consultant, as well as personal comments about the situation.
From a research perspective, these scenes provide insights into patterns of
interaction and communication, as well as the relationship structure of
the participants.


Shame is a self-conscious emotion (Lewis, 2003). Unlike primary emotions,

it does not clearly link to codable facial expressions (Tangney & Dearing,
2004). Similar problems arise with self-reports, as interviewees might con-
fuse shame with other self-conscious emotions like guilt (Tangney &

Dearing, 2004). What’s more, if the data source in question is a TV episode

or movie, the usual interview methods and self-reports can’t be employed
either. Hence, shame has to be analyzed indirectly. It is beyond the scope
of this chapter to provide an in-depth review of research methods on emo-
tions and visual material in general and shame in TV/movies in particular.
Instead, I provide examples of studies to provide an overview of possible
approaches. In general, these research approaches focus on the individual
level of verbal and nonverbal behavior, or the social level of micro interac-
tion and strategic interaction, and narratives or system rules.

Individual Level

To a certain degree, analyzing emotions like shame based on TV shows or

movies is not different than analyzing a videotaped interview session.
Hence, one can analyze the verbal (actual words employed), paraverbal
(loudness, pitch, speed, pauses, etc.) and nonverbal level (gestures, stance,
facial expressions, etc.) of the interviewee. These cues focus on the shame
experience of a specific individual.
An example of this approach is the study of Scheff and Retzinger
(2001a), who analyzed the TV show Candid Camera. Normal people are
filmed being in a “rigged” or unusual situation in which the host plays a
prank on the involuntary participants. In one moment, the host reveals
that the participant have been the victim of a prank (“moment of truth”)
and the participants usually experience intense emotions of embarrass-
ment and shame, relief and surprise, and anger. Scheff and Retzinger
(2001a) analyzed the nonverbal, paraverbal and verbal behavior of the
participants during the moment of truth. Their nonverbal proxy of shame
is related to hiding behavior. Hiding behavior includes actions such as:
the hand covering the face, hands touching face, hair, neck, and chest, a
turned head and/or torso, hiding behind an object or even running out of
the immediate situation, bending over at the waist level or bending knees,
and averting or avoiding gaze/eye contact. The paraverbal proxy of
shame is related to loudness and pitch (speaking very soft, almost inaud-
ible) and long pauses and fragmented sentences. Their analysis of the ver-
bal level is based on the Gottschalk Gleser content analysis scale for the
analysis of psychological states (Gottschalk, Wingert, & Gleser, 1969)
which allows for the coding of words and identifying the state of shame.
See Retzinger (1995) for an overview of verbal, paraverbal and beha-
vioral cues to shame.
Shame and the Undoing of Leadership 87

Social Level

Studying emotions on the social level shifts the focus from individual
expression of emotion to social micro interaction in which emotions arise,
to the function of emotional expression for communication and interaction,
and to matters of emotions and narratives.
One example for the first research approach is the study of the mutual
construction of “face” in interaction. Face the “public self-image that
every [person] wants to claim” for him- or herself (Brown & Levinson,
1987, p. 61) is often studied within the tradition of pragmatics in linguis-
tic research on politeness in conversation. If the maxims of polite conversa-
tions are violated and face is lost, shame is thought to be triggered (Scheff,
1994, 2000b; Scheff & Retzinger, 2001a). Hence, shame is inherently
embedded in social interaction. This level of analysis also includes a verbal
and paraverbal level, and here is used to analyze how face is mutually con-
structed in conversation. Naturally, this involves a change from behavior
and words to sentences, context, and communication strategies, as polite-
ness and impoliteness are “not inherent in particular linguistic and nonlin-
guistic signals” (Culpeper, 2005, p. 41). What’s more, face threats might be
hidden or implicated and thus can’t be coded in a straightforward way.
Instead, analyzing face-loss and shame relies on unfolding the implied
meaning established in conversation. One example of this approach can be
found in Culpeper (2005), who studied impoliteness on the TV show The
Weakest Link. The Weakest Link is a game show where a team of players
has to answer general knowledge questions; each round a contestant is
voted off by the other contestants until only two remain. Culpeper ana-
lyzed the interaction of the host with her guests, and the way she is system-
atically impolite and putting the contestants “one down.” The host’s
behavior partially serves “entertainment” purposes and partially aims to
identify weaknesses and thus the weakest player. Culpeper pays attention
to rhetorical devices such as one-liners, rhetorical questions, and (pseudo)
aphorisms, and paraverbal cues such as intonation (i.e., the rise and fall of
A second tradition analyzes the function of emotions in organizational
communication and discourse. The basis for these kinds of analysis is the
actual text as transcript, in combination with paraverbal and nonverbal
cues. These cues partially relate to the individual (like face color or the
coordination of hand gestures and speech), but also include a social dimen-
sion, that is, address the relationship of the interactants (gazes, trunk
movement, etc.). For example, Tracy (2007) analyzed the role of

“irrational” emotions in supposedly “rational” endeavors like providing

arguments and evidence in decision-making groups and highlights the
importance of emotional display rules. Putnam (2007, p. 95) analyzed how
metatalk about feelings “functions to express and suppress feelings, open
and close spaces for action, and resist and comply with power relation-
ships.” Fitch and Foley (2007) analyzed the persuasive and cultural func-
tion of emotions in communication. By providing a fine-grained analysis of
the transcript, they studied the cultural embeddedness of emotional display
rules in organizational discourse.
A third tradition addresses the function of narratives. Movies and TV
shows tell a certain narrative which constitutes the thesis of the film,
what it is about (Hassard & Buchanan, 2011). While stories and narra-
tives might be discredited as fictitious (and thus unacademic) on the one
hand, they provide important insight if employed fruitfully. For instance,
stories are important sensemaking devices and are already employed in
organizational practice (Weick, 1995). As sensemaking devices, they serve
the same function as theories in science, or might even be understood as
theory, which is “about the connections among phenomena, a story
about why acts, events, structure, and thoughts occur” (Sutton & Staw,
1995, p. 378). Dyer and Wilkins (1991, p. 617) also argue that it is the
story which gives scientific constructs its functionality as “the very clarity
of the constructs stems from the story that supports and demonstrates
them.” Langley (1999) and Pentland (1999) take this argument further
and shows how narratives can be treated as data to construct process
theories of organizations. As movies have narrative(s), Hassard and
Buchanan (2011, p. 625) argue that these “must be plausible if they are
to work, and through the rich description of context and complexity,
films can construct accounts of characters, settings, and events that are
more realistic than the contents of sanitized textbooks and the data gen-
erated by conventional research methods.” Although movies tell a narra-
tive, and thus can be analyzed according to the role and function of
emotions, narratives are also important for the characters involved. In
this regard, narrative and stories play an important role for the construal
of identity (Ibarra & Barbulescu, 2010). An example of the analysis of
narrative and emotions is the work of Gordon (2011). She studies the use
of narrative in the self-construal of parental identity in the American
Reality TV show “Honey We’re Killing the Kids” and show the connec-
tion between the use of visual imagery and visual narrative and face-
saving/embarrassment preventing strategies and self-construals.
Shame and the Undoing of Leadership 89



This chapter takes a view on shame that emphasizes the social side of
shame and embarrassment and its connection with matters of identity and
relationship. Both embarrassment and shame occur in social situations,
hence in interaction with other people, when the core of “who we are” is
questioned. In the following, two approaches to conceptualize the self-in-
relationship and its connection to shame are provided. The first tradition
highlights more the notion of “face” and status, while the second tradition
focuses more on the relationship aspect.
The first tradition is associated with Goffman’s understanding of face
and embarrassment. In every interaction, the participants make a claim
about how they see the other participants and themselves (Goffman, 2003,
p. 7, calls this a “line”). The line provides an evaluation of the relative
“social value” of the participants. The effective positive social value
claimed by a person is the “face” of that person in interaction (Goffman,
2003). The face claimed in interaction is usually accepted (“granted”) by
the other participants. If the line is not accepted and the claimed face not
granted by the other participants, though, the person loses face, feels a
sense of embarrassment (Goffman, 1956), and is “shamefaced” (Goffman,
2003, p. 7).
The second tradition is associated with the work of Scheff and
Retzinger. For instance, Retzinger (1991, p. 38; emphasis in original)
argues that humans “live in the mind of others: we feel pride when we are
noticed favorably, and shame when our bonds are threatened.” According
to Scheff and Retzinger (2001a), this argument goes back to Helen Lewis
who argued that “that shame and threats to the social bond are interdepen-
dent aspects of the same reality. Shame is the emotional aspect of disconnec-
tion between persons” (Scheff & Retzinger, 2001a, p. 27; emphasis in
Both perspectives are complementary to each other if we consider the
connectedness of identity and relationship. Who we are our identity
exhibits different levels of reference. For example, we have a personal
identity, which refers to the unique attributes that characterize ourselves,
and a social identity, which characterizes who we are in relationship to
specific others or in relationship to other groups (Brewer & Gardner,
1996). The social self also encompasses the role identity, hence the identity
we have as a leader, craftsman, teacher, etc. Our social identity is

negotiated in communication and interaction. Each communication

encompasses a relationship aspect (Watzlawick, Bavelas, & Jackson,
1967), that is, a comment on how the “sender” defines him- or herself, the
other, and the relationship itself. Watzlawick et al.’s (1967) understanding
of the relationship dimension already implies that the interactant’s identity
is intertwined with the actual relationship. On the level of communication,
identity is expressed and constituted by narratives (Ibarra & Barbulescu,
2010) and negotiated with the other interactants (Spencer-Oatey, 2007). At
the same time, the relationship itself is a source of identity, as people not
only define themselves by their individual attributes (personal identity) or
group memberships and professional roles (social identity), but also by
their relationships with specific individuals (Sluss & Ashforth, 2007).
Hence, the social identity of a person exists in relation to other people,
and defines who we are for them, in relation to them, and what they are
for us, in relation to us (Andersen & Chen, 2002; DeRue & Ashford,
The face/social value claimed is connected to the notion of “status” in
relationships. Status refers to how “interactors reciprocally define their
positions relative to one another” (Courtright, Millar, & Rogers-Millar,
1979, p. 180) and is determined by the respect and deference received from
others (Carson Marr & Thau, 2014). In terms of social identity, we make a
certain claim about who we are in a specific relationship, that is, how we
see our status in relation to the other person (i.e., whether one is in the
superior, inferior, or equal position). The status and identity claim is
contingent on the other person, as both status and social identity become
“reality” if it is granted by the other party. Hence, on the level of micro
interaction, the “sender” implicitly or explicitly provides a definition how
(s)he sees him- or herself, the other, and the relationship. The “receiver”
then might accept or reject this definition (Haley, 1963). Accepting the defi-
nition means going along with the definition of the other person. Rejecting
could be either implicit or explicit and often implies a counter offer. If the
matter is not settled, a continuous struggle about the relationship and who
you are in relation to the other arises.
If we don’t get the respect we claim in a certain relationship, that is, our
claim about who we are and who the other is in this relationship is denied
by the other person, both the relationship and identity is threatened. The
discounting of one’s identity (“Who you are”; Tangney & Dearing, 2004),
not being accepted as the person that you claim to be, is part of the shame
experience. At the same time, the discounting of one’s claimed identity
threatens the social bond, as there is no agreement on how both
Shame and the Undoing of Leadership 91

participants relate to each other. From this point of view, face and social
identity are both negotiated in communication and relationship, and shame
occurs if face and social identity are not granted or threatened. Both face
and social identity share similar aspects as both refer to positive attributes
that a person claims for him- or herself (Spencer-Oatey, 2007). The social
identity, though, might encompass also neutral and negative attributes.


Shame is defined as the emotional experience in reaction to a threatened
social bond, claimed identity, and social status (face), and thus both the
relationship (status) and identity are negotiated and granted through com-
munication. The way the interactants discursively negotiate and define rela-
tionship and identity can be analyzed. In cases of threatened face, messages
that put the other person one down and when identity claims are denied,
shame is likely to occur.
In order to analyze the material, the episode was recorded from TV,
transcribed and scrutinized multiple times, both on the textual level of dis-
course and the visual-auditive level of the program itself. The analysis was
done in a sequence analytical way to reconstruct how the social reality of
the case was constructed in discourse (Froschauer & Lueger, 2003;
Przyborski & Wohrab-Sahr, 2010). The analysis focused on how the inter-
actants defined the situation they are in, their identity, the identity of their
counterpart, and their relationship. Each of these definitions were also cate-
gorized in terms of leader and follower position, relative status, and claim-
ing, granting and denying a certain face and identity. As noted above, both
face and identity share attributes, and are thus analytically hard to discern.
For the purpose of this analysis, both terms are mentioned in general.
Analytically, face is rather stressed when (relative) social value is claimed
or denied, while identity is rather stressed when the (social or role) identity
is claimed or denied.
As relationships include specific actors and groups, special attention was
paid to how the main actors define and negotiate their relationships. The
main interactants and their relationship are:

• Richard (owner-manager) and Geoff (manager)

• Workers (Staff)
• Relationship between management and staff

The consultant framed the case initially as a leadership problem, hence

the claimed and granted face and identities and relationships between
Richard and the workers are the main focus. Additionally, further informa-
tion was included to highlight specific contexts and further information:
• Spouse and child of Richard
• Relationship between Richard and his wife and daughter
• Gerry Robinson (consultant)
• Relationship between Gerry and Richard
During the analysis, three main situations or patterns emerged that con-
nect shame and face to the negotiation of identity and relationship in the
business context. The first is connected to the feedback of the consultant in
regard to Richard’s aggressive leadership style and the subsequent trigger-
ing of identity work. The second is related to the identity of Richard and
how a belief makes shame-prone relationships due to status put-downs
more likely. The third is related to Richards authoritarian leadership style,
its effect on the relationship to the staff, and a shame-based explanation of
the leadership problem initially diagnosed by consultant and management.


In the following sections, the case analysis is presented. First, the case set
up is displayed, that is, background information and the initial framing of
management and of the consultant about the problem they encounter.
Next the case is reconstructed, based on a framework that highlights
leadership as a certain relationship between leader and follower, based
on mutual respect (or the lack of it), and how shame plays its part in the
relation to leader identity and the relationship to the workforce.

Leadership under Pressure The Case of a Newly Acquired

Lace-Dying Factory

The main characters of this case are Richard, Henry, Geoff, and the staff
at a lace-dying factory which was recently acquired by Richard and his
father Henry. Richard is the managing director and is running the floor
together with manager Geoff. Geoff is also new to this company, but has
Shame and the Undoing of Leadership 93

been working with Richard and Henry for nearly two decades at other
businesses. In what is to follow, I provide a reconstruction of the situation
and contrast the different explanations of management, the consultant and
Richard and Henry describe the problematic situation they’re facing as
such: Richard and his father have been working in the textile industry for
at least two decades. Eighteen month prior to the filming, Richard (at that
point in time 39 years old) and his father bought an old lace-making busi-
ness, founded in 1899. Richard describes the relationship with the staff as
problematic: “We have to push all the time to get things done, push, push,
push to get things done.” Workers are having pauses, working slow, doing
shoddy work, or not following instructions:

Richard: I am telling them what I want and it is not being done. I will write it down,
but yet it is not being done, the simplest of tasks are not being done. They it seems to
be traditional that they just throw it in. It just seems amazing, it is such a valuable fab-
ric and they just throw it through the door. Now what do you do with people that are
just basically either not listening to you or feel that what you are telling them is wrong.

Consequently, productivity and quality of work are low, which means

high costs. In combination with the high debt due to the acquisition, the
financial situation puts Richard and Henry under a lot of pressure and
As a result, Richard and Geoff are following a rather authoritarian lea-
dership style which emphasizes high demands and control: “… it is just a
big job to put it right, you have got to be on the factory floor all of the
time.” When asked about the causes, Richard answers: “It is quality issues,
it’s people.” Manager Geoff has similar, but even more telling answer: “It’s
tending to be a little bit in the dark ages here and it is just a case of educat-
ing people as to what the current trade requires.”
From the point of view of management, the situation is framed quite
clearly. The problem is “people” people who work lazily, don’t have an
acceptable quality standard, and need to be educated to move them from
the “dark ages” to modern times. The authoritarian leadership style of
management is a reaction to this situation. From this point of view, it fol-
lows logically that you have to exercise a high amount of control over your
lazy workers from the “dark ages.” The followers “are” the problem. To
treat the problem, they got to be controlled and “educated.”
From the point of view of the consultant, though, it’s not the followers
who are the problem it’s the leader. Robinson diagnoses: “it’s a sort of
incredible aggression” that the management emanates, as well as “a

panicky feel about him [Richard], he’s frightened, that anger all comes out
of insecurity,” and relates this to the “huge financial commitment.” The
leader “needs help[, that] is very alarmingly clear.” The consultant also
highlights the other side of the coin the situation of the workers, and
connects it to the leadership style and personality of Richard:
Gerry Robinson: You have got a work force that is pretty disillusioned, they see you
buzzing around. You are frantic about making things happen more quickly than they
do. There is a kind of aggression there which maybe isn’t real at all, but stylistically you
are actually quite frightening.
Gerry Robinson: You do frighten them in my view, let alone they are wary of you, I
think you frighten them.

So far, the portrayed discourse is rather straightforward. Management

and consultant frame their perspective in “personal” terms (people, leader).
Emotions are also acknowledged in an implicit and explicit way, mainly
fear (panicky, frightened) and anger (aggression), and some kind sadness
Shame seems to be absent from the overt framing by management and
consultant, but it plays a major role in this case and it’s mostly hidden.
The first contribution of shame can be seen when Richard is confronted
with his leadership style. Richard loses his projected face, which triggers
shame and a process of identity work, to find a new face and identity to be
claimed. The second contribution lies in a personal belief of Richard that is
prone to create potentially shaming situations like a self-fulfilling pro-
phecy. The third and major contribution lies in the vicious cycle of “one
down” messages from management, which results in “one down” behavior
of the staff, and subsequently prevents leadership to occur.

Shame as a Threat to Leadership Case Reconstruction

In what is to follow, three situations are reconstructed from the perspective

of shame and leadership, based on the relational identities of the interac-
tants. When needed, further theoretical elaboration is provided.

Shame and Identity Work

The first instance of shame can be seen in the reaction of Richard, after
being confronted with the idea that he is aggressive and frightening the
staff. Richard seems flabbergasted: “I don’t know how to take that …”
Robinson frames this as being defensive and provides comforting words:
Shame and the Undoing of Leadership 95

“It is important Richard not to be defensive here because I think you are a
decent bloke.” Richard continues: “It’s not defensive it is just a little bit of
well defending yourself a little bit, I don’t think I am right aggressive.”
The face and identity Richards claims not being an aggressive guy
is threatened by the feedback of the consultant. Robinson tries to save face
by attributing being a “decent bloke” to Richard. Richards tries to stick to
the claimed face by denying that the granted identity suits him. Later,
though, the feedback triggers a process of identity work. Identity work can
be understood as the “forming, repairing, maintaining, strengthening or
revising the constructions that are productive of a precarious sense of
coherence and distinctiveness” (Alvesson & Willmott, 2002, p. 626). The
feedback of the consultant created a sense of self-doubt and skepticism
about Richards perception of who he thinks he is in interaction (not an
aggressive guy), and who he “really” is (a frightening, aggressive guy). The
realization of this difference and the doubt that he might be somebody else
as he thinks he is “a shock, the aggression side is a shock” for Richard. At
home, he discusses the identity granted by the consultant with his wife and
daughter: “I am still quite amazed that he [Gerry Robinson] has come up
with that I’m aggressive and saying that to me.” His wife responds: “It is
not that you are aggressive, it is just that you look … just you look aggres-
sive, your hair is short and …” The wife also tries to save face and the old
identity by making the issue about looks (the outer appearance), not the
inner identity. Richard also checks the granted identity with his daughter:
Richard: Do you see me as aggressive Ann?
Daughter: Not at home, but at work I think you are, when people aren’t doing the job
that you want them to do you get angry.

During the next days, Richard continues to work on the feedback and
thus on the identity and face that he feels being rightful to claim in
Richard: It is obviously going around now, it is always in the back of my mind as to
what he has said about my own personal approach to people and that’s err something I
am going to work on over a period of …, well as long as it takes.

He checks the old identity with a staff member:

Richard: When he [Robinson] came he actually made a few comments that I come
across as being very abrasive and very aggressive.
Worker: Yeah, that’s right.

Richard: How do you feel about that?


Worker: I think most people found that, you were like untouchable.

The worker confirms the consultant’s comment on Richard’s effect. His

answer furthermore indicates a comment not only on Richard (aggressive)
but on the relationship between management and staff which indicates a
very distanced quality (untouchable). This comment already foreshadows
the insecure bond between management and staff that was created during
the first months after Richard took over (see the analysis on the shame
spiral for an in-depth look into this issue).
To summarize: Richard was used to claim a certain face and identity in
interaction, which is discursively established as “not being an aggressive
guy.” The consultant doesn’t grant this face and identity; which “shocks”
Richard. Within the shame framework, I interpret this as a loss of face,
and the experience of shame (“shock,” being “surprised” that the issue of
aggression was “openly said” to him, etc.). We see face-saving work by the
consultant (“decent bloke”) and wife (“maybe you just look aggressive”)
and a process of identity work where Richard questions his old identity
and works on a new understanding of himself as leader.

Preemptive One-Upmanship
The second contribution of shame in understanding the dynamics of the
case is also connected to Richard’s identity, especially a core belief of his
own. His wife characterizes this belief as such that “a sign of kindness is a
sign of weakness and if you are too nice then people will just take advan-
tage and walk all over you.” The notion of status and relative positioning
is quite clear from this characterization. Explicitly labeled is the weak side,
and the weak side is associated with kindness. If you are weak, people will
walk all over you and take advantage of you. If we assume that people in
general don’t like to be “walked upon,” the image painted here is putting
the actor “one down,” on the lower side of the relative status positioning.
By anticipating this reaction, Richard positions himself as tough and
aggressive, a bad target for being walked upon and taken advantage of.
But at the same time, he sends the relationship message that he is “one up”
compared to the other interactant, no matter what, which means that his
communication partner is “one down.” Potentially, he invites other interac-
tants to enter a status struggle and shame-inducing experience as not being
accepted by him as highly valued interactant, but rather as a person in
lower status.
Richard’s belief and its potential influence underline the importance of
shame in interaction. Scheff (1994, 2000b) stresses the importance of not
Shame and the Undoing of Leadership 97

only the actual shame experience but the avoidance of anticipated shame.
Relying on Cooley (1902) and Goffman (1967), he argues that humans con-
tinually monitor the relationship status and how they are seen by the other
party. Thus, humans continually anticipate shame and embarrassment and
aim at avoiding it. Richard’s belief, though, indicates a shame avoidance
not by mutual face-saving (a symmetric relationship, based on equality),
but anticipates being but down and thus reverses the complementary rela-
tionship by putting himself one up.
To summarize: Part of Richards identity is a belief that is thought to
prevent being “put down” by others, hence lose face and experience shame.
This belief lets Richard to anticipatorily prevent shame by being “tough”
and position himself on the upper level of the relative status hierarchy. This
preventive move potentially put’s other people one down and bears the
potential for shame-inducing encounters.

Collective Spiral of Unacknowledged Shame

Both episodes above refer to the experience of shame as the identity
claimed in interaction is not granted. The question remains how the issue
of leadership relates to issues of shame, relationship, and identity. In recent
time, leadership has been conceptualized as a special kind of relationship
(Fairhurst & Uhl-Bien, 2012; Uhl-Bien, 2006) that encompasses the roles of
leader and follower (Uhl-Bien, Riggio, Lowe, & Carsten, 2014). What’s
more, role incumbents develop identities based on their professional roles
(Ashforth, Kreiner, & Fugate, 2000) and the relationships connected to
that role (Sluss & Ashforth, 2007), hence, who you are as a leader/follower,
and in relation to your follower/leader. Leadership thus is created by a pro-
cess of reciprocal claiming and granting of leader and follower identities
(DeRue & Ashford, 2010). Watzlawick et al. (1967) argue that relationship
is either based on equality and sameness or inequality and complementar-
ity. In the context of leadership, the leader-follower relationship can be
understood as a complementary relationship as the role of leader and fol-
lower are unequal in terms of responsibilities, status, and authority (Yukl,
2013). Importantly, the granting of a leader identity by the follower is a
process that relies on “a complicity or process of negotiation through which
certain individuals, implicitly or explicitly, surrender their power to define
the nature of their experience to others. Indeed, leadership depends on the
existence of individuals willing, as a result of inclination or pressure, to sur-
render, at least in part, the powers to shape and define their own reality”
(Smircich & Morgan, 1982, p. 258). In the leader-follower relationship,
shame could occur in two scenarios: First, the leader is not accepted by the

followers, the face/identity claimed by the leader not granted, and thus a
shame experience is triggered. Second, the follower is not accepted by the
leader, the face/identity claimed by the follower not granted, and thus a
shame experience is triggered.
Usually, a shame reaction is connected to blushing, making yourself
smaller, even fleeing the scene (Tangney & Dearing, 2004), starting face
work and trying to repair or regain the claimed face (Domenici &
Littlejohn, 2006), or entering a relationship conflict (Scheff & Retzinger,
2001a; Tangney & Dearing, 2004). According to Scheff and Retzinger
(2001a), whether face work that is, trying to save face and repairing
the relationship or conflict occurs depends on whether shame is
mutually acknowledged or occurs covertly. If shame is mutually acknowl-
edged, both interactants are aware about the loss of face and can engage
in repairing the relationship. If shame is not acknowledged, though, it
can trigger anger and rage instead of withdrawal (Retzinger, 1991), and
set in motion a shame spiral of mutual disrespect: Let’s assume that two
people, A and B, interact, and A disrespects B, which denies the social
identity claimed by B (you are not who you claim to be), equals to a loss
of face (you are not as socially valuable and deem respect as you claim
to be) and triggers the experience of shame in B. If the shame episode is
acknowledged by B (i.e., shame is experienced authentically) and this is
acknowledged by A, the interactants can take action to repair the social
bond. If the shame episode is not acknowledged, though, anger occurs in
B, and a shame spiral of mutual disrespect occurs. B feels disrespected by
A and subsequently disrespects A. A experiences shame, which will, as it
also goes unacknowledged, cause anger and result in subsequent disre-
spect of B. A spiral of unacknowledged disrespect, shame, and anger
occurs which erodes the social bond in each cycle. Scheff and Retzinger
(2001a) demonstrate the workings of shame spirals in different contexts
like marital quarrels and family or societal conflicts. But how does a
shame spiral might look like in the business context, especially in the
context of leadership? This question leads us to the third contribution of
shame to understand the dynamics of this case, which starts where the
other two episodes end.
Richards aggressive style, combined with a situation of high financial
pressure (and thus a huge need to turn around and reorganize the business
rather sooner than later), led him to send the relationship message that the
staff is to blame for the insolvency of the business before, and thus has to
change quick. Whether or not Richard intended to send this message this
is the way it was received, as one worker tells him:
Shame and the Undoing of Leadership 99

Worker 1: Do you think we are performing as a workforce now, because we don’t get
any feedback to say how well we are doing. It’s just when you took over you just seem
like you were trying to have a bit of a go; it was all “Vernon Road employees that sank
Vernon Road,” I feel that that was the sort of impression that you gave us, that it was
our fault.

Two issues are of importance here: A previous put-down and a subse-

quent lack of appreciation. The worker acknowledges that Richard leader-
ship talk and behavior when he and his father took over the business was
interpreted as a huge put-down. Within the context of Richards and Geoff
statements that the issue is “people,” who work like in the “dark ages” and
need to be “educated,” and the question of the worker about the quality of
their work (how they perform as a workforce), we can assume that the staff
understood the message as them being “bad” workers (a loss of face in
regard to quality of their work). If we assume that the workers wouldn’t
claim to be responsible for the failure of the business, Richards relationship
message puts them lower down than a worker with pride would accept.
The message attacks the “quality face” (Spencer-Oatey, 2002), which is
connected with the claim for valuable attributes (e.g., being a competent
worker of good quality). Instead, Richard grants the workers a faulty iden-
tity and collectively shames them. What’s more, the worker indicates that
the staff didn’t receive any collective acknowledgment of the quality of
their work after the initial put-down. Within the shame framework estab-
lished above, the lack of public reappraisal can be understood as an indica-
tor for unacknowledged shame.
The problems that Richard, Henry, and Geoff encounter can now be
interpreted as a shame-based spiral of passive resistance. In essence, by tak-
ing breaks, working slowly and working shoddily, the workers refuse follo-
wership. Leadership and followership both are labels for a certain kind of
relationship, and the workers are refusing to enter a follower role due to the
lack of respect of their leaders. They answer the put-down of Richard with a
put-down by themselves by neglecting the leadership claim Richard and
Geoff make. Manager Geoff takes an even harder stance. His comment that
the workers needs to be “educated” and that they work like in the dark ages
puts him one up as being in charge as well as the expert who can educate the
“darkagelings.” The identity and face he offers for the workers is highly
unattractive, and the workers react negatively to these relationship messages.
What’s more, Geoff not only offers a negative face, but also disconfirms
their identity as valuable human beings worth his attention if they refuse fol-
lowership. As one worker puts it: “If he comes up to you and says ‘Can you
work overtime’ and you say no, he doesn’t talk to you for two weeks, he will

totally blank you.” This reaction of “blanking” is a special kind of rejection

of the relationship definition called “disconfirmation” (Watzlawick et al.,
1967). If the workers make a relationship claim to being “quality workers,”
the rejection of this claim by management still acknowledges the existence of
the person as such and confirms their “humanity.” “Blanking” or discon-
firming, however, is a rejection on a deeper relationship level. Now, the
other person is neglected as a person and human being, that is, “You don’t
count” or even “You don’t exist.” The messages of disconfirmation, the
authoritarian understanding of relationships, and acts of unilateral control
to ensure “quality work” provide a set of one down messages on the rela-
tionship level. The workers lose face, experience shame and are confronted
with a rather negative face and identity granted by management.
After experiencing the shame associated with the put-down, the staff has
the option to enter face work (in case of mutually acknowledged shame) or
a state of aggression and relationship conflict (in case of unacknowledged
shame). The TV episode only provides evidence for the latter. The workers
refuse the offered face and identity, and reclaim part of their status by not
accepting the leadership claim of Richard and Geoff. The complaints of
Richard about “people being the problem” who need constant control mir-
ror the not granted leader identity. The workers take breaks, work slow,
and deliver bad quality which are examples of behavior that neglects the
leadership claim of management and shows the autonomy and self-
determination of the staff. By retaining the right to refuse the leadership
claim of Richard and Geoff, they put themselves “one up,” as it is them
who decide who is the leader, and who is not.
Richard realizes that his leadership position is not being accepted
(hence, he is automatically not in the strong, one-up position). He reacts to
this put-down accordingly: He doesn’t accept the relationship and identity
definition of the workers and tries to enforce followership by being control-
ling and authoritarian. Combining both the managements and staffs reac-
tions of put-downs, the spiral of unacknowledged shame is complete. Both
parties answer disrespect with disrespect. Although each party supposedly
struggles for respect, their mode of conduct of unilaterally redefining the
granted or claimed face and identity induces shame in the other party.
Fig. 1 provides an overview of the shame spiral and the contributing con-
text factors.
The feedback of the consultant about Richard’s aggressive appearance
triggered a process that led to an open talk of Richard with a group of
workers. Within this meeting, the occurrence of shame was acknowledged
(although not in these words) and a process of relationship repair was
Shame and the Undoing of Leadership 101

Fig. 1. Shame Spiral and Contributing Context Factors.

initiated. After being confronted with Richards initial leadership speech

included a put-down (“… when you took over it seemed like you were try-
ing to have a bit of a go, that it was all Vernon Road employees that sank
Vernon Road and I feel that that was the sort of impression that you gave
us, that it was our fault”), Richard acknowledges his contribution in this
shaming experience and momentarily lowers his status claim by admitting
mistakes and providing insight into the stressful situation he was facing:
Richard: That was wrong. The approach I came in with was wrong, okay, and I am tell-
ing you that now, it was wrong, but the time I came in here it was literally ‘hit the
ground at a gallop and sort everything out’…, I mean we had got no gas, they were
going to cut that off, they were cutting the tele …, everything, every day we …

This admission in turn is answered by further moves to repair the rela-

tionship by acknowledging the need for communication and admitting that
both parties have mutually suffered in this process:
Worker 2: Why didn’t you talk to people about that at the time?
Richard: Literally, I haven’t had time until now.
Worker 2: It was rough for all of us.

Richard: It was rough for me as well.


Both parties are levelling, being equal in having suffered all are “vic-
tims” in a sense. The staff furthermore addresses the issue of mutual respect
by stressing their importance for the success of the company and that both
parties are in mutual reliance of each other (emphasis added):
Worker 1: You [Richard] can stand there and I am not criticizing you, you can say that
“I’ve been working hard, I have phone calls until 11.00 at night,” you have got the firm
on an even keel, you want it to progress, you have done it, Richard, but you have done
it with us, because without us you are nothing and without you we are nothing, so it is a

This dialogue acknowledges the occurrence of shame and face-loss that

had occurred in the past. Instead of neglecting the other party, the dis-
course establishes “good will” and mutual acknowledgment, and empha-
sizes mutual dependency and the need for cooperation. In the program, the
dialogue between workers and management provided the groundwork for
an initial healing of the relationship. Unfortunately, the further develop-
ment didn’t follow this promising path as Richard turned back to his old
style and manager Geoff who was first fired due to his antagonistic lea-
dership style and then rehired to force a higher production later contin-
ued being a tough leader. Several rounds of layoffs had a further
detrimental effect on the motivation of the staff. The Vernon Road
Bleaching and Dyeing Company was dissolved in 2009.
To summarize: Richard sends a relationship message that defines the
workers as “faulty,” being responsible for the failure of the previous busi-
ness. By neglecting a claim for a “quality face,” being a worker of
respectable competence, the workers were collectively shamed. They in turn
reject the relationship definition by management by rejecting the claim of
management to be in a leader position. Both parties refused to grant the
other party the claimed faces and identities which in turn leads to further
acts of control by management, which in turn shames the staff as not being
able to work on their own. A shame spiral occurs that erodes the relation-
ship between management and workers. Leadership as such can’t be exer-
cised as management denies the workers the identity as valuable workers
and the staff denies the claim for leadership.


This chapter provides insight into (1) the function of shame (understood as
a threatened social bond due to non-granting a claimed face and identity
Shame and the Undoing of Leadership 103

and a devaluation of the self) within the context of business leadership, and
(2) the opportunity to do research on emotions in organizations, based on
readily available TV/movie materials.
The case makes three points about shame, identity/relationship, and lea-
dership. First, we see the connection of shame and identity, as the expert
a person of high status grants an “aggressive” identity to the leader. This
feedback is highly disturbing, as the leader first denies this identity, and
later enters identity work in order to accommodate to this situation. This
finding is in line with the conception of shame as a threatened (i.e., claimed,
but not granted) identity. Second, we see how beliefs about relationship (as
part of the identity) ready a person to enter a shame-prone relationship. In
order to avoid being put down, the leader preemptively takes a powerful
stance and thus puts the other person one down. While this might ensure
one-upmanship, it might trigger a shame reaction in the other person as
s(he) feels put down. The social bond is thus under continuous threat.
Although a leader is in a slightly higher status position by definition, this
kind of relationship is granted by followers who give up their power to
shape their own reality temporarily (Smircich & Morgan, 1982). In case of
a status put down by a leader, handing over the power to the leader
becomes less likely. Third, we see a shame spiral in an organizational con-
text. While these kinds of shame spirals are known in the realm of marriage
and family (Scheff & Retzinger, 2001a), this case provides insight into
shame spirals in the business context. The experience of unacknowledged
shame undoes the creation of leadership. As argued above, leadership is
a relational phenomenon that relies on the followers constantly handing
over power to leaders and by this creating a certain relationship. The put
down from the leader side leads to a refusal to grant a leader identity. By
denying a leader identity, the workers also deny followership, hence leader-
ship as such is not established. Instead, the workers work reluctantly and
follow the directions of management mostly in situations of direct control.
In the absence of direct control, though, there is no voluntary following the
instructions. Instead, the workers do as they prefer, take breaks, etc.
Ironically, the denying of a leader identity makes management claim their
superior position even harder, which results in further put downs. A spiral
of unacknowledged shame is complete, and leadership as such is undone.
Instead, compliance of workers is ensured by control and punishment
A limitation of the method is that the case reconstruction is based on
edited film material, and thus no claim is made to have reconstructed the
case as a whole, as it “really” occurred. There might be other factors that

contributed to the problematic leadership situation that weren’t highlighted

in the TV reel (e.g. factors of social identity due to unionized workers and
the micropolitical implications thereof, a real need for training and new
equipment, etc.). Nonetheless, these factors complement the explanation
provided in this chapter by further contextualizing and they also compli-
cate the situation (but they don’t invalidate the analysis). Future research
could analyze these factors under a different frame (e.g., power and poli-
tics) to provide an alternative explanation. Even so, the boundary condi-
tion would remain of being a reconstruction of the case from a different
point of view.
The strengths of the method of using publicly available visual material
as a data source for investigating emotional and organizational phenomena
is that it is easily accessed at little to no cost. Second, it provides an oppor-
tunity to code verbal and nonverbal emotional information over time.
Third, it provides rich observational data.
With respect to our understanding of shame, the application of the con-
cept of shame spirals using organizational theories, herein leadership, pro-
vides useful insight into how shame shapes behavior in organizations.
Future research is needed to investigate shame spirals within the context of
other organizational theories. Further, the literature on shame spirals
shares some similar assumptions as research on the crossover of emotional
and psychological states and the process of discrete emotional crossover.
Application of this literature to the study of shame in organizations would
provide another fruitful avenue for future research (Härtel & Page, 2009).


This chapter was enabled by the support of CeFEO and the Jan
Wallanders och Tom Hedelius Stiftelse.


Alvesson, M., & Willmott, H. (2002). Identity regulation as organizational control: Producing
the appropriate individual. Journal of Management Studies, 39(5), 619 644.
Andersen, S. M., & Chen, S. (2002). The relational self: An interpersonal social-cognitive the-
ory. Psychological Review, 109(4), 619 645.
Ashford, S. J. (2013). Having scholarly impact: The art of hitting academic home runs.
Academy of Management Learning & Education, 12(4), 623 633.
Shame and the Undoing of Leadership 105

Ashforth, B. E., Kreiner, G. E., & Fugate, M. (2000). All in a day’s work: Boundaries and
micro role transitions. Academy of Management Review, 25(3), 472 491.
Barrett, K. C. (1995). A functionalist approach to shame and guilt. In J. P. Tangney & K. W.
Fischer (Eds.), Self-conscious emotions (pp. 25 63). New York, NY: The Guilford
Bell, E. (2008). Reading management and organization in film. Basingstoke: Palgrave
Bell, E., & Davison, J. (2013). Visual management studies: Empirical and theoretical
approaches. International Journal of Management Reviews, 15(2), 167 184.
Bignell, J. (2005). Big brother: Reality TV in the twenty-first century. New York, NY: Palgrave
Bignell, J. (2013). An introduction to television studies (3rd ed.). London: Routledge.
Brewer, M. B., & Gardner, W. (1996). Who is this “we”? Levels of collective identity and self
representations. Journal of Personality and Social Psychology, 71(1), 83 93.
Brown, P., & Levinson, S. (1987). Politeness: Some universals in language usage. Cambridge:
Cambridge University Press.
Buchanan, D., & Huczynski, A. (2004). Images of influence: 12 angry men and thirteen days.
Journal of Management Inquiry, 13(4), 312 323.
Carson Marr, J., & Thau, S. (2014). Falling from great (and not-so-great) heights: How initial
status positions influences performance after status loss. Academy of Management
Journal, 57(1), 223 248.
Cooley, C. H. (1902). Human nature and the social order. New York, NY: Charles Scribner’s
Courtright, J. A., Millar, F. E., & Rogers-Millar, L. E. (1979). Domineeringness and domi-
nance: Replication and expansion. Communication Monographs, 46(3), 179 192.
Creed, D. W. E., Hudson, B. A., Okhuysen, G. A., & Smith-Crowe, K. (2014). Swimming in a
sea of shame: Incorporating emotion into explanations of institutional reproduction
and change. Academy of Management Review, 39(3), 275 301.
Crowley, M. (2014). Class, control, and relational indignity: Labor process foundations for
workplace humiliation, conflict, and shame. American Behavioral Scientist, 58(3),
416 434.
Culpeper, J. (2005). Impoliteness and entertainment in the television quiz show: The weakest
link. Journal of Politeness Research: Language, Behavior, Culture, 1(1), 35 72.
Czarniawska, B., Eriksson-Zetterquist, U., & Renemark, D. (2013). Women and work in
family soap operas. Gender, Work & Organization, 20(3), 267 282.
DeRue, D. S., & Ashford, S. J. (2010). Who will lead and who will follow? A social process of
leadership identity construction in organizations. Academy of Management Review,
35(4), 627 647.
Domenici, K., & Littlejohn, S. W. (2006). Facework: bridging theory and practice. Thousand
Oaks, CA: Sage.
Dyer, W. G., Jr., & Wilkins, A. L. (1991). Better stories, not better constructs, to generate bet-
ter theory: A rejoinder to Eisenhardt. Academy of Management Review, 16(3), 613 619.
Fairhurst, G. T., & Uhl-Bien, M. (2012). Organizational discourse analysis (ODA): Examining
leadership as a relational process. The Leadership Quarterly, 23(6), 1043 1062.
Fitch, K. L., & Foley, M. (2007). The persuasive nature of emotion and the cultural nature of
feelings in organizations. In F. Cooren (Ed.), Interacting and organizing (pp. 113 126).
Mahwah, NJ: Lawrence Erlbaum Associates.

Froschauer, U., & Lueger, M. (2003). Das qualitative Interview. Wien: Facultas.
Goffman, E. (1956). Embarrassment and social organization. American Journal of Sociology,
62(3), 264 271.
Goffman, E. (1967). Interaction ritual. New York, NY: Anchor Books.
Goffman, E. (2003). On face-work: An analysis of ritual elements in social interaction.
Reflections, 4(3), 7 13.
Gordon, C. (2011). Impression management on reality TV: Emotion in parental accounts.
Journal of Pragmatics, 43(14), 3551 3564.
Gottschalk, L., Wingert, C., & Gleser, G. (1969). Manual of instruction for using the
Gottschalk-Gleser content analysis scales. Berkeley, CA: University of California Press.
Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to leadership:
Development of leader-member exchange (LMX) theory of leadership over 25 years:
Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6(2),
219 247.
Haley, J. (1963). Strategies of psychotherapy. New York, NY: Grune & Stratton.
Härtel, C. E. J., & Page, K. M. (2009). Discrete emotional crossover in the workplace: The
role of affect intensity. Journal of Managerial Psychology, 24(3), 237 253.
Hassard, J. S. (2011). Researching work and institutions through ethnographic documentaries.
In D. A. Buchanan & A. Bryman (Eds.), The Sage handbook of organizational research
methods (pp. 270 282). London: Sage.
Hassard, J. S., & Buchanan, D. A. (2011). From modern times to Syriana: Feature films as
research data. In D. A. Buchanan & A. Bryman (Eds.), The Sage handbook of organiza-
tional research methods (pp. 620 632). London: Sage.
Holt, R., & Zundel, M. (2014). Understanding management, trade, and society through fiction:
Lessons from The Wire. Academy of Management Review, 39(4), 576 585.
Ibarra, H., & Barbulescu, R. (2010). Identity as narrative: Prevalence, effectiveness, and conse-
quences of narrative identity work in macro work role transitions. Academy of
Management Review, 35(1), 135 154.
Infante, E. R. (2011). Researching work-family discourses: Step by step audiovisual analysis of
the British sitcom Only Fools and Horses (1981 2003). Newcastle upon Tyne:
Cambridge Scholars Publishing.
Langley, A. (1999). Strategies for theorizing from process data. Academy of Management
Review, 24(4), 691 710.
Lewis, M. (2003). The role of the self in shame. Social Research, 70(4), 1181 1204.
Lewis, T. (2008). Changing rooms, biggest losers and backyard blitzes: A history of makeover
television in the United Kingdom, United States and Australia. Continuum, 22(4),
447 458.
Pentland, B. T. (1999). Building process theory with narrative: From description to explana-
tion. Academy of Management Review, 24(4), 711 724.
Powell, W. W., & Colyvas, J. A. (2008). Microfoundations of institutional theory. In
R. Greenwood, C. Oliver, R. Suddaby, & K. Sahlin (Eds.), The SAGE handbook of
organizational institutionalism (pp. 276 299). London: Sage.
Przyborski, A., & Wohrab-Sahr, M. (2010). Qualitative Sozialforschung Ein Arbeitsbuch.
München: Oldenbourg.
Putnam, L. L. (2007). Contradictions in the metatalk about feelings in corporation: After
Mr. Sam. In F. Cooren (Ed.), Interacting and organizing (pp. 95 111). Mahwah, NJ:
Lawrence Erlbaum Associates.
Shame and the Undoing of Leadership 107

Retzinger, S. (1991). Shame, anger, and conflict: Case study of emotional violence. Journal of
Family Violence, 6(1), 37 59.
Retzinger, S. M. (1995). Identifying shame and anger in discourse. The American Behavioral
Scientist, 38(8), 1104.
Sauer, E., & Ropo, A. (2006). Leadership and the driving force of shame: A social construc-
tionist analysis of narrative. In W. J. Zerbe, N. M. Ashkanasy, & C. E. J. Härtel (Eds.),
Individual and organizational perspectives on emotion management and display (Vol. 2,
pp. 57 80). Bingley, UK: Emerald Group Publishing Limited.
Scheff, T. (2014). The ubiquity of hidden shame in modernity. Cultural Sociology, 8(2),
129 141.
Scheff, T. J. (1994). Microsociology. Chicago, IL: The University of Chicago Press.
Scheff, T. J. (2000a). Bloody revenge. Lincoln: iUniverse Inc.
Scheff, T. J. (2000b). Shame and the social bond: A sociological theory. Sociological Theory,
18(1), 84 99.
Scheff, T. J., & Retzinger, S. M. (2001a). Emotions and violence Shame and rage in destruc-
tive conflicts. Lincoln: iUniverse Inc.
Scheff, T. J., & Retzinger, S. M. (2001b). Two videotape studies of social action. In T. J.
Scheff & S. M. Retzinger (Eds.), Emotions and violence (pp. 41 60). Lincoln: iUniverse
Sluss, D. M., & Ashforth, B. E. (2007). Relational identity and identification: Defining our-
selves through workplace relationships. Academy of Management Review, 32(1), 9 32.
Smircich, L., & Morgan, G. (1982). Leadership The management of meaning. Journal of
Applied Behavioral Science, 18(3), 257 273.
Spencer-Oatey, H. (2002). Managing rapport in talk: Using rapport sensitive incidents to
explore the motivational concerns underlying the management of relations. Journal of
Pragmatics, 34(5), 529 545.
Spencer-Oatey, H. (2007). Theories of identity and the analysis of face. Journal of Pragmatics,
39(4), 639 656.
Sutton, R. I., & Staw, B. M. (1995). What theory is not. Administrative Science Quarterly,
40(3), 371 384.
Tangney, J. P., & Dearing, R. L. (2004). Shame and guilt. New York, NY: The Guilford Press.
Tracy, K. (2007). Feeling-limned talk: Conduct ideals in the Steinberg succession meeting. In
F. Cooren (Ed.), Interacting and organizing (pp. 77 94). Mahwah, NJ: Lawrence
Erlbaum Associates.
Uhl-Bien, M. (2006). Relational leadership theory: Exploring the social processes of leadership
and organizing. The Leadership Quarterly, 17(6), 654 676.
Uhl-Bien, M., Riggio, R. E., Lowe, K. B., & Carsten, M. K. (2014). Followership theory: A
review and research agenda. The Leadership Quarterly, 25(1), 83 104.
Watzlawick, P., Bavelas, J. B., & Jackson, D. D. (1967). Pragmatics of human communication.
New York, NY: W. W. Norton & Company.
Weick, K. E. (1995). Sensemaking in organisations. Thousand Oaks, CA: Sage.
Yukl, G. (2013). Leadership in organizations (8th ed.). Boston, MA: Pearson Education.
Zundel, M., Holt, R., & Cornelissen, J. (2013). Institutional work in The Wire: An ethological
investigation of flexibility in organizational adaptation. Journal of Management Inquiry,
22(1), 102 120.
This page intentionally left blank


Dirk Lindebaum


The processes that underlie ability emotional intelligence (EI) are barely
understood, despite decades of management research. Furthermore, the
outcomes of these processes have been narrowly and prescriptively
defined. To address this deficiency, I conducted a phenomenological
study (n = 26). Findings from a public sector sample suggest that the
underlying emotional processes of meaningful life events are at least
for now better defined through the construct of emotion regulation.
While it is part of the ability EI model, the emotional processing that
occurs prior to emotion regulation being initiated is likely to be less con-
sistent with current EI theory. Likewise, these processes lead to out-
comes considerably more nuanced than currently appreciated in the EI
literature. Consequently, what started as a gap-filling approach to

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 109 137
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011006

research eventually turned into a problematization of what scholars seem

to know about EI. I outline the theoretical and practical implications of
this study for management, and offer suggestions for future research.
Keywords: Emotional intelligence; emotion regulation; phenomenol-
ogy; processes

The purpose of this study is to elicit the processes that underlie ability emo-
tional intelligence (EI), and to examine their associated perceived individual
outcomes. EI has been defined as “ability to perceive accurately, appraise,
and express emotion; the ability to access and/or generate feelings when
they facilitate thought; the ability to understand emotion and emotional
knowledge; and the ability to regulate emotions” (Mayer & Salovey, 1997,
p. 10). Despite decades of (largely) quantitative management research, evi-
dence that EI is a valid construct is mixed. Among the key issues are ques-
tions about its incremental validity over and above general mental ability
and personality (Antonakis, Ashkanasy, & Dasborough, 2009; O’Boyle,
Humphrey, Pollack, Hawver, & Story, 2011), as well as problems concern-
ing the factor structure and reliabilities of the dominant EI measure
(i.e., the MSCEIT or Mayer-Salovey-Caruso-Emotional-Intelligence-Test,
see these studies for empirical divergence: Føllesdal & Hagtvet, 2013;
Mayer, Salovey, Caruso, & Sitarenios, 2003).
One contributing factor to these mixed findings may be an incomplete
understanding of the processes that underlie EI. For instance, Fiori (2009,
p. 24) argues that “the analysis of processes underlying EI may reveal that
individuals differ in how they engage in mechanisms responsible for emo-
tionally intelligent performance.” Mayer, Salovey, and Caruso (2004) also
recognize that “understanding the processes underlying EI” (p. 211) is not
well understood. Others bemoan the predominantly descriptive nature of
EI (Matthews, Roberts, & Zeidner, 2004). To date, available descriptors of
EI make reference to (i) the maximum capacity to process emotional data,
(ii) the configuration of mental processes and a tapping into intra-psychic
experiences, and (iii) the assimilation of emotional experiences into mental
life (e.g., Mayer, Caruso, & Salovey, 1999; Mayer, Roberts, & Barsade,
2008). However, these descriptors are somewhat vague, since they do not
account for generative processes that explain how and why they work. For
instance, when it is stated that EI pertains to tapping into intra-psychic
A Qualitative Study of Emotional Intelligence 111

experiences, does this imply a tapping at conscious or subconscious levels?

In addition, with regard to configuring mental processes, it is not clear
what factors might initiate them, how the process can be defined, and what
consequences might be experienced.
In consequence, the first theoretical contribution of this study is to elicit
the processes underlying EI by examining the lived experiences of indivi-
duals through a phenomenological study. Phenomenology seeks to render
explicit the often implicit meaning and structure of human experience
(Sanders, 1982). The choice of qualitative inquiry is appropriate in this
study, since it offers the possibility to explore emotional processes at work
in ways that quantitative studies cannot (Maitlis & Ozcelik, 2004). Also, it
is useful to help explain quantitative results requiring further analysis
(Creswell & Plano Clark, 2011), which the mixed quantitative findings on
EI discussed earlier suggest. A better theoretical understanding of these
processes is of fundamental importance for management studies, for this
enhanced understanding can help better delineate the theoretical para-
meters of EI theory, as well as its empirical testing.
The second theoretical contribution concerns a broader appreciation of
outcomes resulting from these processes. Beyond existing concerns about
the descriptive nature of EI, I suggest that EI is also prescriptive in nature
(cf. also Lindebaum, 2009; Lindebaum & Jordan, 2014). That is, abilities
associated with EI have been hitherto exclusively likened to positive and
linear outcomes for individuals, for instance, in terms of job performance
or health outcomes (Joseph & Newman, 2010; Martins, Ramalho, &
Morin, 2010). These findings converge insofar as they reflect linear and
positive correlations between EI and outcome variables. In this prescriptive
respect, it is noteworthy that the definition of EI as stated earlier is often
accompanied by the annex that it “promotes emotional and intellectual
growth” (see e.g., Mayer & Salovey, 1997). However, there is a nascent
theoretical debate that being emotionally intelligent may also lead to nega-
tive outcomes (such as deviance, see Winkel, Wyland, Shaffer, & Clason,
2011), or in the form of curvilinear effects (e.g., Jordan, Dasborough,
Daus, & Ashkanasy, 2010). This resonates with Lincoln’s view (2009),
when she points to the pitfalls of the “research that is being done here,
that is, experimental-heritage, conventional paper-and-pencil tests based
on a somewhat scientific (and often researcher-determined) definition of
EI” (p. 785).
Combining these two contributions, this phenomenological study seeks
to extend and elaborate theory in terms of how individual differ in pro-
cessing emotional information, and how this initiates varying emotional

processes over time, which, in turn, can lead to a multiplicity of out-

comes. Given the general lack of qualitative EI research, these findings
have considerable potential to shed light on the why EI tends to increase
over time as expounded by the lived experiences of individuals as
opposed to merely noting that it increases (Kafetsios, 2004). The practical
implications of this enhanced understanding for management can be sig-
nificant. Especially in terms of EI interventions (see, e.g., Clarke, 2006;
Thory, 2013), this better understanding of processes enhances the pro-
spects that interventions can succeed, which is of critical importance vis-
à-vis the financial resources that organizations invest in EI interventions
(Kunnanatt, 2004).
The article is structured in the following manner. First, I outline the EI
construct in detail. Second, I report findings from a phenomenological
study designed to address the two research questions posed in this study.
Finally, I elaborate upon the theoretical and practical ramifications of this
study for management. Importantly, my original intention was to fill an
important gap in the body knowledge on EI. However, and consistent with
inductive research sometimes yielding surprising findings (Eisenhardt,
1989), the ultimate framing of this study’s contribution is more akin to
what scholars refer to as the problematization of knowledge (i.e., how can
we think differently about we already know, see Alvesson & Sandberg,
2011). Findings suggest the need to probably think differently about EI in
terms of processes and its outcomes, while they point to emotion regulation
as an alternative construct that for the time being potentially better
captures the emotional processes that inform the lived experiences of indivi-
duals. Indeed, emotion regulation is part of the ability EI model. However,
the emotional processing that occurs prior to emotion regulation being
initiated is likely to be less consistent with the hierarchical and sequential
logic of the ability EI model.

There are two reasons for selecting ability EI as a theoretical framework.
First, there is evidence to suggest that emotional abilities increase with age
when they are assessed with performance tests (Kafetsios, 2004), which are
said to enable the objective determination of right or wrong answers
(Mayer et al., 2004). A positive correlation between EI and age is often
lacking when self-report measures of EI are employed (Hemmati, Mills, &
A Qualitative Study of Emotional Intelligence 113

Kroner, 2004). Second, self-report measures of EI are typically encapsu-

lated within personality theory (Petrides, Pita, & Kokkinaki, 2007) and,
therefore, represent an individual’s “preferred way of behavior” as opposed
to “mental performance on emotion-related abilities” (Mayer et al., 1999,
p. 270). Given this study’s interest in the underlying emotional processes
that lead to increased EI, it shall be understood that the ability version of
EI is a preferable theoretical foundation of this study as opposed to the so-
called mixed models or Trait EI models, which rely on self-report measures
(see Mayer, Salovey, & Caruso, 2000 for a comprehensive review of major
Advocates of the EI construct frequently point to studies highlighting
the intimate nexus between emotion and cognition to suggest that EI may
aid individuals in pursuing adaptive goals and behaviors (e.g., Mayer &
Salovey, 1995). For instance, Baron (2008) notes that emotions and cogni-
tion are entwined by means of processes “through which information is
entered into memory, processed, and retrieved for later use” (p. 328). As
Mayer and Salovey (1997) note, “using the emotions as one basis for think-
ing, and thinking with emotions themselves, may be related to important
social competencies and adaptive behavior” (p. 22).
The ability model of EI is hierarchically and sequentially organized.
That is, the complexity of emotional abilities increases from lower to higher
tiers, and comprises four abilities. Further to this, each branch has asso-
ciated with its stages or levels of ability, which are mastered by individuals
sequentially. As noted by Mayer et al. (1999), at the lowest level, indivi-
duals first perceive and express emotions (e.g., in faces). Second, they
assimilate basic emotional experiences into their mental life, evaluating
emotions against each other, and against various sensations and thoughts
and permitting emotions to direct attention. Third, individuals seek to
understand and reason with and about emotions and determine how these
emotions change over time, for each emotion follows a specific rule
(Caruso & Salovey, 2004). That is to say, anger arises in response to injus-
tice, fear often turns into relief and so forth (Solomon, 1993). Fourth, at
the highest level, the management of emotions concerns knowing how to
calm down after anger or assuaging the anxiety of another person. The
aforementioned four aspects are said to be interrelated. For instance,
awareness of emotions is indispensable for their regulation (e.g., empathy
may be conducive to managing emotion in others). In congruence with this,
studies suggest that all four aspects are positively interrelated (Gannon &
Ranzijn, 2005). The four-branch EI ability model has been operationalized
with the MSCEIT (see Mayer, Salovey, & Caruso, 2002). While advocates

report acceptable psychometrics properties of this measure, such as reliabil-

ities and factor structure (Mayer et al., 2003), other studies suggest findings
to the contrary, especially in terms of factor structure (Føllesdal &
Hagtvet, 2013; Palmer, Gignac, Manocha, & Stough, 2005).
Since how individuals reason about and with emotions is central to the
EI construct, it is relevant to refer to the work of Feldman Barrett,
Mesquita, Ochsner, and Gross (2007). They argue that a direct examination
of an individual’s mental representation is best achieved by investigating
the verbal responses concerning their own behavior. Extending this view,
they also argue that “an adequate description of what people feel is
required so that scientists know what to explain in the first place” (p. 374).
This is consistent with Lincoln (2009), who calls for a removal of
researcher-defined representations of what constitute emotionally intelli-
gent behavior. The arguments concerning the processes that underlie EI
and what outcomes they may yield inform the two research questions
(RQs) of this study. Specifically, through the study of lived experiences:
RQ1. What processes can be identified that underlie EI and help differ-
entiate how individuals engage in mechanisms responsible for a variety
of outcomes?

RQ2. What do these outcomes look like?

I am concerned here with the phenomenological accounts provided by par-
ticipants. This allows to embrace “the complexity of the human individual
and their actions” (Ardley, 2011, p. 637), as well as prioritizing the indivi-
dual’s own interpretation of a situation. Phenomenological approaches
focus on the participants’ own interpretation of their lived experience
(Sanders, 1982). Here, it is assumed that the emotional processes under
investigation are individually constructed through active sense-making.
This sense-making process mainly unfolds in the interaction with others
(Johnson & Cassell, 2001). The role of the researcher is of fundamentally
interpretive nature (Henwood & Pidgeon, 1992). Encouraging individuals
to share their emotional experiences was key, especially in terms of emo-
tionally challenging life situations. Balancing job performance and one’s
well-being served as a starting point to elicit these challenging life situa-
tions, for individuals often struggle to maintain a healthy balance between
A Qualitative Study of Emotional Intelligence 115

these factors (Gooty, Gavin, Ashkanasy, & Thomas, 2014).

Correspondingly, I asked participants whether success at work or their
well-being is more important to them, followed by promptings as to
whether they can recall concrete situations at work. Another key question
sought to elicit more generally whether participants could recall a life event
that they found emotionally very challenging, to the extent that their beha-
vior changed as a result (i.e., exploring maximum challenges in processing
emotional information consistent with EI theory). Note that while some
participants were grateful, others were adamant that they were not grateful
at all for the emotional challenges they faced.
From the above follows that this study (and qualitative research more
generally) typically excludes a concern for validity, reliability, objectivity,
generalizability, and verification (Amis & Silk, 2008). Instead, and consis-
tent with writings of other qualitative researchers (Lincoln & Guba, 1985),
I focused upon (i) credibility (instead of internal validity), (ii) transferabil-
ity (instead of external validity), (iii) dependability (instead of reliability),
and (iv) confirmability (instead of objectivity) of findings. Thus, I concur
with Cassell, Bishop, Symon, Johnson, and Buehring (2009), who “do not
subscribe to the view that there is a fixed point from which qualitative
research can be considered ‘good’ or ‘bad’ in an externally-legitimated
way. Rather, individuals will construct their own sense of what is good,
which will privilege their own understanding and sensemaking processes”
(p. 517). However, being rigorous in the collection and analysis of data
has been a central concern not in the sense of the creation of distance
and non-involvement (Etherington, 2004), but rather as a systematic read-
ing and re-reading of the interview scripts, coupled with a persistent oscil-
lating between theory and data to ensure intimate communication between
the two.

Approach to Data Collection and Analysis

I conducted 26 interviews with participants from a local authority (n = 12)

and University administration (n = 14) in the United Kingdom between
April and July 2010. Underlying this choice is a consideration for purposive
sampling, as it is essentially strategic and implies an effort to establish a
good correspondence between research question and sampling (Bryman,
2004). Both organizations underwent considerable restructuring and reor-
ganization at that time, which can evoke anxiety and fear for individuals
(Huy, 1999). It shall be understood that the data emerging from the

interviews do not always correspond with my theoretically informed ques-

tions. That is, since the data creation process is governed by the accounts
that participants provide, the interviews contained considerably more depth
and width than could be mapped neatly onto the interview guide (which is
available upon request).
I negotiated access with key managers from the local authority and uni-
versity departments. Managers passed on an Informed Consent Sheet to
staff, with an emphasis that participation in this study is entirely volunta-
rily. In doing so, participants were assured of the strict confidentiality and
anonymity inherent in the interview process, and that numbers would be
used to communicate the identity of participants.
The sample consisted of 15 female and 11 male employees. Duration of
interviews ranged between 40 and 60 minutes. The sample size was influ-
enced by the fact that, after the 23rd interview, participants did not pro-
duce any significant new data, concepts, or themes (relative to the themes
outlined later). This corresponds to what Lincoln and Guba (1985) describe
as theory saturation. All participants agreed to their interview being audio-
recorded, and a professional service provider transcripted all interviews in
full. To disguise participants’ identities, I use pseudo names when reporting
the data.
During the data analysis, the interviews were subjected to four iterations
of scrutiny. First, I listened to interviews to obtain a sense of the whole and
to double-check transcription accuracy, while also noting emphasis in voice
or speech pauses as well as non-verbal cues, such as finger-tipping on the
table. The second iteration of data analysis aimed to separate the interview
material according to relevance to the research questions. I then partially
followed the widely used recommendations for phenomenological data ana-
lysis by Hycner (1985).1 This enables the creation of stronger findings since
a “clear methodological process is used” (Gephart, 2004, p. 458). Having
engaged in “bracketing” to note down conscious presuppositions (see also
Wright, Murray, & Geale, 2007),2 I perused the interviews again (i.e., the
3rd iteration) to delineate general units of meaning (e.g., words, phrases, or
non-verbal cues that reveal unique and coherent meaning). At this stage, I
also produced a hand-written summary (between 1 and 2 pages) of each
interview. I then read the scripts again in order to delineate more closely
units of meaning as they relate to the research questions. Hycner (1985)
argues that this enables the clustering of relevant units (i.e., do units of rele-
vant meanings cluster together naturally?). It was at this stage that I rea-
lized the original intention of a gap-filling approach was not
suitable anymore, and that framing the contribution in terms of
A Qualitative Study of Emotional Intelligence 117

problematization is necessary. In the 4th iteration, I could determine

themes from the specific units of meaning. Guided by the research ques-
tions, three themes emerged from the data analysis. These themes (which
reflect clusters of relevant units of meaning) are labeled (i) factors initiating
emotional processes, (ii) characteristics of processes (in terms format, inde-
pendence of emotional processes, mechanisms, and contents), and (iii)
work and health-related outcomes of these processes. Following this, I
involved a second experienced researcher in phenomenological data analy-
sis in order to assess the plausibility of the themes I identified. Importantly,
the joint deliberation of the themes’ contents did not focus upon calculating
the inter-rater reliabilities of both researchers’ scores (but rather upon cred-
ibility of findings as highlighted earlier). Overall, this consultation
prompted some minor changes in the description of themes. Lastly, in qua-
litative analysis data are often first dissected to create new meaning struc-
ture that correspond to the questions posed, while new connections
between the new structure of the data are allowed to emerge (Miles &
Huberman, 1994). As the analysis suggests, several new connections
became apparent.


Factors Initiating Emotional Processes

While filtering iteratively the data in quest of emotional processes, it

became evident that without an appreciation of factors initiating emotional
processes, any subsequent processural understanding is likely to remain
incomplete. Two factors stood out in this respect, namely, (i) courage and
(ii) honesty about individuals’ own emotions and thoughts, as outlined


In terms of a presence of courage, several participants shared similar

experiences. For instance, Sally spoke of her experience in dealing with a
colleague at work from a different department, with whom she had to pro-
duce a policy document. Following some misunderstandings about her
intentions of changes to the document, she noted:

I said: Well I think we should meet for a coffee and talk it through. And that was extre-
mely difficult because I’m absolutely adverse to confrontation. I really didn’t find that
comfortable at all. But it was really … interesting that we sat down. It was very difficult
at first.

Confronting unpleasant situations emerged also in David’s account. He

shared an example with regard to the link between courage, the expression
of work-related frustration, and stress relief. The situation in question per-
tained to a workshop led by senior management, which he saw as a cause
of that work-related frustration on the restructuring of the organization.
He said (in front of 20 colleagues):
“You’ve been talking about this for a long time and you’re achieving nothing” …
“Why don’t you just get on with it, if you’re going to do it, do it” … I could see the
room fall silent. One of my colleagues … said, ‘You’re like Wolfe Smith out of Citizen
Smith’, an old television program, power to the people … I don’t normally get on my
soapbox and argue, but it’s a great stress buster”.

The above quote indicates that speaking truthfully (which requires cour-
age) in such a situation was seen by David as source of stress relief.
However, there were several accounts that suggested, just as some parti-
cipants displayed the courage to confront uncomfortable situations, others
struggled with it. One informative account stems from Martin:
When I was younger my father died which in itself is obviously life changing … As a
person I felt it was a time when … your emotions were most on public display and
I … reacted to that by completely … shutting off any … emotional response when I
was in public … Looking back on it all that happened was that the grief was just buried
until later and came out over a number of years. And these insights you only have with
age and experience, but it would be nice to talk to that young man and explain what he
was going through and how things would change and what was the best.

Despite him being now an adult, this reflection still did not come easily
to him, as underlined by the swallowing speech pause before the last word
in the above excerpt, which in itself was uttered in a quivering voice. It
appears that Martin decided to suppress the emotion-eliciting event for a
long time.


This theme emerged from several accounts, in which life-changing experi-

ences (e.g., divorces, grievance cases at work, and stress at work) initiated a
greater honesty about the participants’ own emotions.3 Like courage, it
should be borne in mind that this sense of honesty is intimately linked to
A Qualitative Study of Emotional Intelligence 119

the outcomes individuals experience, as highlighted later. For instance,

Andrew appreciated that following his divorce and a grievance case at
work that the important
thing is recognizing, has something, has anything changed … I think maybe what’s
changed from all those circumstances is the sense of … being more honest.

A similar storyline surfaced in Wayne’s account, who told of his hospita-

lization some time ago, which was for him a life-threatening experience. It
turned out, however, that it was a condition that could be cured with minor
surgery. Yet, his angst made him become “like a child … crying every time
somebody came to see” him at his hospital bed. While previously he
thought of himself as “strong,” this situation made him realize that he was
“not as strong as [he] thought [he] were,” and that he “stopped pretending”
that he is “very strong,” which brought the “honest person” out of him.
Intriguingly, like several other participants in this study, he felt grateful for
that life-changing experience, to the extent that he felt grateful for it. In all
cases, reflections on the pain-instilling experiences over time were essential
in leading to that sense of honesty with which they can now deal with, or
relate to, emotional experiences in more adaptive ways.

Processes in Terms of Format, Independence of Emotional Abilities,

Mechanism, and Content

Clearly distinguishable themes emerged from the data in terms of format

(i.e., curvilinear vs. linear effects), independence of emotional abilities
(i.e., do individual experiences reflect the sequential logic of the ability EI
model?), mechanisms (e.g., cognitive transformation of emotion elicitors),
and contents (i.e., conscious vs. subconscious). Note that these themes,
despite the need to distinguish them, are nevertheless often related, as high-
lighted below.


The lived experiences of several participants were often inconsistent with

the linear logic of EI, which almost exclusively assumes positive and linear
relationships between EI and important life outcomes. That is, it became
evident that sometimes processing emotional information was engaged in
too much, or participants could not stop thinking about particular events,

which was perceived as an annoyance. In terms of excessive thinking, for

instance, Terry said that he “probably worries too much about things over
which [he has] got no control.” In a similar vein, on being asked whether
she has a tendency to worry or ruminate, Marie stated:
I think I do that … too much. I think about things a lot … I’m probably a bit
paranoid … I over-analyze things all the time. I wake up in the night thinking about
things, write them down. It’s everything in my life, that’s not just work.

Sally made similar suggestions. When asked whether she has a propen-
sity to ruminate, she responded that she “tick(s) that box every time” and
that she “chew(s) over things a lot.” An inability to stop thinking thoughts
that are emotionally draining emerged from the account of Leanne.
Following an outline of numerous events that she currently has to organize
or attend to, I raised the question whether the planning worries or agitates
her, or whether it is a positive agitation. Her response was she has “a ner-
vous energy probably that [she] get(s) a bit hyper,” but also that she “can’t
switch off sometimes.”
By contrast, when asked whether he has a tendency to worry excessively,
Murray was quick to respond that “worry, no … it’s not in [his] genes,” a
statement that was accompanied by what might be described as hysterical
laughter. A very similar storyline was shared by David, who noted that he
can only worry about things when they arise, not when they are imagined
future situations, adding that “if you go over it in your mind all the time
you can plan and that is a good thing but to over-plan” was not regarded
as helpful.

Independence of Emotional Processes

The label of this theme reflects sequential issues (relative to the hierarchi-
cally and sequentially organized ability EI model outlined earlier) that ema-
nated from several participants’ accounts. For instance, Damian elaborated
on an important experience he had at university, when he failed an accoun-
tancy exam (i.e., he felt “completely at a loss,” as if it was “some sort of
grief”). Being forced to sit in the exam again, he noticed that this comple-
tely changed him “emotionally,” insofar as he understood better that he is
“going to fail things,” that it is bound to happen at some point, and that
“you should … prepare for failure” since it “is not the end of the world.”
The consequence of better understanding the causes and consequences of
emotions made him more resilient (note, this pertains to the third branch
of the ability EI model). However, elsewhere Damian was candid that he
“wouldn’t use [emotions] at all really” in fear of coming across “as being
A Qualitative Study of Emotional Intelligence 121

angry or upset” when influencing others (i.e., managing of emotions in

others, fourth branch). Interestingly, when asked whether he puts his heart
and mind into performing well at work, he thought that “it’s important to
engage both the heart and mind because … sometimes you can be
too … logical and factual about things” whereas “a certain amount of how
things feel” should be allowed “to influence how [he] act(s).” Thus, the case
of Damian underscores that he does not use emotions at the intra versus
interpersonal level to the same degree. It is not clear whether this reflects a
lack of ability per se, but his determined response in terms of not using
emotions to influence others suggests a discontinuity as far as the sequen-
tial logic of the ability EI model is concerned. The case of Sally offers
further support that individuals can perceive their emotional abilities as
varying in proficiency. When asked how she believes her health affects her
performance at work (and vice versa), she suggested that:

You could almost force yourself into that situation [i.e., feeling more positive and ener-
getic] … and think, ‘Oh gosh I’ve got a lot to do, but you know what I’m going to get
this this and this,’ and you can bring yourself to a positive mental attitude if you like.

What surfaces in her account is that she, with some effort, can induce a
positive mental attitude if she “feel(s) a bit sick,” that she has “to be more
driven,” and that she has “to force (her)self” on occasion to get the job done.
This corresponds with the second branch of the model (i.e., using emotions
to facilitate thinking). However, as the interview progressed, she admitted
that she “(doesn’t) think (she’s) mastered” putting herself emotionally in the
shoes of another person (i.e., first branch of model). She continued:

Sometimes I’m surprised at how people are feeling. I misjudge things. In a meeting I’ll
think that person’s really got it and after the meeting that person will be ‘Oh my good-
ness that was awful,’ and I’ve been quite surprised … I’ve got a lot more to learn.

More insights on whether participants excelled at emotional abilities

along the logic of the ability model resulted from Lindsay. Like other parti-
cipants, I asked her whether she is able to put herself emotionally in the
shoes of another person (i.e., first branch of EI model). She replied that she
does that “quite well,” since she experienced repeated struggle with the per-
formance management side of her job, when she asks herself “how would I
feel?” However, she noted that, while she mostly feels “alright” in terms of
her empathetic skills, she is “not one of these ones that reads it, processes
it, and modifies behavior straight away.” Like other accounts before, hers
suggests a discontinuity among the EI branches that have been hitherto
assumed to be hierarchical and sequential.


Participant accounts falling within this theme serve very closely to answer
RQ1 (i.e., what processes can be identified that underlie EI and help differ-
entiate how individuals engage in mechanisms responsible for a variety of
outcomes?) Recall that current descriptors of EI tend to be rather vague,
referring to tapping into intra-psychic experiences or configuration of men-
tal processes. As such, this theme is concerned with a more tangible
description of mechanisms that steer the processes underlying EI. Two sub-
themes emerged as I deepened the analysis: (i) cognitive transformation of
the emotion elicitor and (ii) changes in attention focus.
In terms of participants’ narratives indicating a cognitive transformation
of the emotion elicitors, the data analysis revealed a high frequency of its
occurrence. To begin with, Nina shared the view that “work isn’t every-
thing to [her] at all” when asked whether success at work or happiness is
more important to her, adding later that she has stopped “taking work as
serious” following the death of a very close friend (i.e., it changed her out-
look on life to appreciate family and friends more). In a similar vein,
Wayne alluded to a significant change in his outlook on life following a
previous hospital admission, during which he sensed a real death anxiety.
That experience transformed “problems” into “situations” for him in terms
of living with other peoples’ problems, since not “everybody is the same as
[him].” In addition, David engaged in cognitively transformed situations to
the effect that:
Life can’t always be great and neither should it always be bad … it does move around
and I think that people can have far too greater expectations or become very disap-
pointed or disillusioned and depressed if it’s bad.

With regard to changes in attention focus, one informative account was

provided by Chris. Increasingly disliking his job at the local authority, he
alluded that he always wanted to pursue a career as a musician.
Sadly, I always had the music that was … an escape and with that music is a dream
of … having success with what you do because you feel it’s good. So I think that was
always not like a backup but a buffer … or maybe it’s like a dream. That was your
dream [and] you’re thinking: ‘Okay, this will only last for so long because we are doing
this with the music.

His reference to “escape” underlines that he distracted his attention

focus to pursuits considered more life-enhancing than the constant fear of
losing his job as a result of multiple rounds of restructuring. He even
A Qualitative Study of Emotional Intelligence 123

admitted that he would “never reach that light” (i.e., being a professional
musician), yet he was adamant that he was “always trying to head towards
it.” In a related fashion, Bill talked about how he became less “emotional”
having taken on previously the emotional burdens of colleagues who
worked with him in an organization with a bullying culture. He indicated
that he has is a “nurtured a little bit the ability to detach [him]self from
both staff, both people and both the emotional journey that people go on
at work,” to the point he pretends that he is not interested in conversations
at work during which he would feel like the “sponge for everyone else’s
issues and problems.” Wayne, by contrast, did not attempt to distract him-
self, but rather consciously concentrated on the pleasure and life-enhancing
events in life (following his hospital admission). In his words:
I said: ‘You have life, enjoy it, otherwise you’ll lose it … the joy of being there, so work
with the greatest mood I always did it but this actually gave more boost, go for it
tomorrow you may, you know, you just walk and somebody may run over you by car
and you are finished.

Again, even though his lived experienced is contained within this theme,
it is worth repeating that the act of concentrating on life-enhancing events
has important consequences both for himself and his social environment.


In terms of contents of processes, only few participants made reference to

the distinction between conscious and sub-or unconscious (i.e., automated)
processes being at play as a result of life-changing (and emotional) experi-
ences.4 Foremost among participants who were aware of the above dis-
tinction was Andrew. Reflecting on his (painful) experience of getting
divorced from his wife (i.e., “the emotionally most difficult thing”), he
noted that he is now a lot more “aware of what catches you out” when it
comes to situations that are emotionally meaningful to him. While at the
time of the relationship breakdown he could not identify what caused it,
he added:
I like to think that I’ve learned from that in terms of the way I deal with the current
situation I’m in, because no relationship’s actually perfect … you need to make those
allowances and need to do that in a conscious way to the point where … you become
unconsciously competent and you’ve got a consciously competent stage about it and
then hopefully you become unconsciously competent because you do it as a matter of
automatic reaction.

His lived experiences of getting divorced prompted him to try to be con-

sciously aware of what might surprise him emotionally. Once he had inter-
nalized the conscious practice, he felt that this practice increasingly
rendered the process unconscious (or a matter of automatic reactions). In a
similar vein, the narrative of Paula also referred to how she reacted to
work events that for her turned out to be unpleasantly painful at first.
As hinted earlier, she suffered under the new leadership of her department,
which placed higher performance expectations upon her. Feeling upset and
angry about the conduct of a performance appraisal meeting with depart-
ment leader (and an external consultant), she indicated:
I was upset and I was a little bit angry probably as well … as a result of that
[i.e., appraisal meeting]. I think I distanced myself, not consciously but subconsciously.
I distanced myself a little bit and I stopped caring as much to a point. And I just
thought: ‘Well … I can only do what I can do and if that’s not good enough for you
then I’m just going to get on with it, get my head down do my job and I’ll see.

In retrospect, she noticed that the distancing did not occur in a conscious
fashion, but rather occurred subconsciously. The interesting insight from her
narrative and others is how that distancing affected both her performance
and well-being at work. In the next theme, I will outline personally experi-
enced outcomes from the variety of emotional processes detailed above.

Work and Health-Related Outcomes of Emotional Processes

Consistent with the aforementioned analytical approach to first dissect the

data, while subsequently allowing for new connections to emerge, this
theme offers many interfaces with the preceding themes, especially the pro-
cess subthemes (i) format and (iii) mechanisms. Below I touch upon these
two processural subthemes to highlight how and why they relate to the out-
comes that individuals experience.
Starting with the focus on format of processes, engaging too much in
processing emotional information mirrors the characteristics of curvilinear
effect (i.e., too much of anything can be bad). For instance, Marie was
once “emotionally very attached” to her former employer, being “very
loyal,” hard working, and loving it. In the course of a restructuring, she
was offered a new job. However, the demands of that new job meant that
she “was too stressed” and felt “quite physically ill.” After some time, she
felt that her loyalty was not reciprocated, prompting her to feel “mis-
treated.” She felt heartbroken to the extent that she told them that she
A Qualitative Study of Emotional Intelligence 125

“cannot work for [them] anymore.” She quit the job, but swore to herself
that she will “never give that [i.e., degree of loyalty and dedication] again.”
Her case underlines that an inability to regulate how much emotional loy-
alty is exerted has detrimental consequences for her well-being. Negative
outcomes for one’s well-being as a result of processing emotional informa-
tion too much also emerged in the account of Bill. Being empathetic and
having a tendency to speak his mind in support of colleagues who “were
harshly treated,” he became somewhat of a spokesperson for his colleagues
at his former job, which was conspicuous by a bullying culture and lack of
direction. Yet, he noticed that being too receptive to other peoples’ pro-
blems did not do his “mental health very much good.” Eventually, he quit
his job to retain his sanity. However, since he developed the ability to
detach himself from work, he is now “a better manager.” Similarly, exces-
sive worry was perceived by Terry as a hindrance to career progression.
I take less risks than some other people … I think I do worry too much about what
might go wrong and spend too much time worrying about that and I think that stops
me from taking some of the risks that some people take, perhaps worrying over
nothing … I may have been more successful at work if I took a few more chances.

Therefore, processing emotional information excessively (i.e., worrying

too much) had the perceived consequence of impaired career progression in
the case of Bill. Other participants (like Kim) indicated what an absence of
excessive worrying might entail, candidly noting that she is “quite care-
free.” Kim added:
The job that I’m doing … is important to get everything processed and do all that side
of things, but I wouldn’t take it to heart or anything like: … ‘Oh, I didn’t get this work
done so this is going to be terrible.

Thus, Kim was clear that she would not allow her work to “affect [her]
personally.” Her view proffers an intriguing contrast in terms of outcomes
that participants perceived and experienced as a result of excessive proces-
sing of emotional information (or lack thereof). Further insights surfaced
when I examined the consequences of applying the specific mechanisms of
emotional processing mentioned earlier. To begin with, Emma told of her
experiences as an HR manager, in which role she repeatedly had to make
staff redundant. Applying cognitive reframing of the situation as a protec-
tive mechanism, she believed that
compartmentalizing what you do … and … realizing that: ‘This is the job, this is what
you’ve got to do, and this is not me personally doing this to somebody [i.e., make staff

Using this protective mechanism was essential to her, since she has “had
very unpleasant experiences where people have subsequently committed
suicide.” As another example, Andrew admitted that he “had spells
in … life when [he has] had anxiety and small bouts of depression.”
However, he did not experience this for a long time, noting that now he is
“quite good at managing [his] own emotional wellbeing.” His insight stands
in direct relation to his efforts to cognitively reframe meaningful life events,
so as to lessen the emotional impact of them.
As an example of attentional deployment, the case of Paula illustrates
that, by distracting her attention away from the situation that caused her
stress (i.e., not to care so much and looking at it “from an
outside … rather than personal perspective”), she realized that she started
“ticking those boxes” at work in terms of her performance. In her words,
she “was able to function better” as a result of distancing herself from
work.5 A similar yet distinct story was shared by Sandra. As the most
emotional experience she had in her life, the case of her father’s illness
some time ago highlighted that her ability to detach herself from the
situation emotionally (i.e., distracting attention focus) had consequences
that she perceived as favorable for her (and other family members). In
her words:

I didn’t show a lot of emotion because I was conscious that my dad was very worried
about his own health and he didn’t need me to be showing that I was worried and upset
as well. So I … detached myself from it and tried to occupy myself and not sit dwelling
about it because it wouldn’t do me any good and would only make me more
worried … I think that’s just how I deal with it. Brave face, it’ll be OK and try and
look for the positive and what’s going to happen … If I dwelt, I don’t think I’d come
out of my shell, I think I’d go into a corner and not come out.

Her narrative is instrumental in showing that detaching from the situa-

tion caused her pain and fear (i.e., distraction in attention focus), but had
both intra and interpersonal consequences that were perceived as desirable.
While all these examples indicate that by either cognitively transforming
the emotion elicitor or shifting one’s attention, a (perceived) desirable out-
come can be obtained, it should also be understood that an inability to
handle these mechanisms implies that individuals can suffer negative out-
comes. For instance, some noted that their inability to “switch off” some-
times causes them to wake up at night (e.g., Sally, Leanne, and Marie), or
that role-prescriptions of being a HR manager can be stressful, especially
when one has to convey negative performance feedback or, worse still, to
lay off staff (Sharon).
A Qualitative Study of Emotional Intelligence 127


This study aimed to elicit the processes that underlie EI and to examine the
perceived outcomes at the individual level that these processes can lead to.
The motivation behind this endeavor resided in the fact that advocates of
EI themselves admit that little is known about the processes associated
with EI. In addition, it has been suggested that the outcomes of high EI
have been too narrowly and prescriptively defined, especially in terms of
pro-social and health indicators, as well as performance at work
(e.g., Mayer et al., 2008). As indicated earlier, the widely-accepted defini-
tion of EI is frequently followed by the annex “to promote emotional and
intellectual growth.” However, Suddaby (2010) cautions that, when con-
structs are defined, researchers often conflate the processes underlying a
particular phenomenon and its outcomes. If these are not clearly distin-
guished, our own value judgments predetermine the nature of the outcome
(Lindebaum & Jordan, 2014). Of note, while my initial motivation for this
study partly reflected a “gap-filling” intention with regard to the first aim,
the overall data analysis suggests that a problematizing framing of the con-
tribution was necessary eventually. This is consistent with inductive
research sometimes yielding unexpected findings (Eisenhardt, 1989). This,
in turn, is mirrored in several significant theoretical and practical contribu-
tions for the management community.

Theoretical Contributions

The empirical analysis revealed rich insights into the emotional processes
that form part of the lived experiences of participants. The first significant
insight stemmed from identifying factors that are seemingly required to
initiate a deeper processing of emotional information. In other words, how
can one speak of an individual’s ability to process emotional information
at maximum capacity in the absence of courage and honesty about oneself
in the first place? As the data show, a willingness on the part of participants
to closely look at that which gave them considerable emotional pain (such
as being averse to confrontation (Sally), or suffering from stress (David)),
was essential for initiating subsequent processural reflections. These subse-
quent reflections were akin to meaningful learning experiences that ren-
dered some participants more resilient and more confident about
themselves in terms of confronting (as opposed to avoiding) painful

situations in life. Consequently, the argument that EI will lead to “emo-

tional and intellectual growth” can probably only be sustained if the factors
of process initiation are fully appreciated. This is consistent with recent stu-
dies emphasizing the need to appreciate the role of antecedent variables
and how they affect the outcomes of EI (Leonard & Harvey, 2007).
The second theoretical contribution relates to current assumptions about
the underlying processes of EI, which are potentially in need for reconsi-
deration. This contribution has several manifestations. First, the linear
assumptions linking EI and various outcomes oftentimes do not manifest
themselves in the data. In fact, the data suggest more often than not that
individuals engaged excessively in processing emotion information (most
notably in worrying, rumination, or an inability to “switch off”).6 This is
consistent with prior speculations about the presence of curvilinear effects
in EI research (Jordan et al., 2010). In fact, findings speak neatly to recent
debates in management research on the “too-much-of-a-good-thing effect”
(TMGT effect, see Pierce & Aguinis, 2013, p. 313), whereby variables often-
times accepted as having desirable outcomes for individuals (in this case,
EI) actually have undesirable ones for individuals due to patterns of curvili-
nearity. Second, the phenomenological accounts of several participants
question the sequential and hierarchical logic of the four-branch EI model.
For instance, the case of Damian underlines the complex distribution of
mastery across all four branches of the EI model (but see also Sally and
Lindsay’s accounts). Specifically, whereas he perceived skill at understand-
ing emotions (e.g., with the consequences of becoming more resilient), he
was candid that he would not use emotions in trying to influence colleagues
at work (i.e., managing others’ emotions) in fear of being regarded as angry
or upset. Intriguingly, he puts his mind and heart into performing well at
work, by which he meant that a certain amount of how he feels about spe-
cific situations at work (i.e., using emotions to facilitate thought) should be
allowed to influence how he acts. In consequence, these accounts cast some
preliminary doubt upon whether the sequential and hierarchical logic of EI
accurately represent the actual thought-processes and mastery of abilities in
the general population. It is not hard to discern that, based upon the varied
accounts provided, that considerable fluctuation may exist across indivi-
duals in their mastery of emotional abilities, something that quantitative
studies would find hard to detect. This fluctuation may also help explain
the less-than-consistent factor structure of the MSCEIT. If the participants
above were to complete the MSCEIT, following their accounts they would
be likely to do better on some branches compared to others. In conse-
quence, the consistency of the factor structure is likely to be adversely
A Qualitative Study of Emotional Intelligence 129

affected. If future research provides further evidence to this effect (i.e.,

varied mastery across all branches), then this is likely to reinforce calls for
further improvement of this psychometric instrument (see also Føllesdal &
Hagtvet, 2013).
Third, in terms of mechanisms that characterize how participants pro-
cess emotional information in emotionally challenging situations, the parti-
cipants’ accounts suggest two prominent ones, namely, the (i) “cognitive
transformation” of emotion elicitors and (ii) “changes in attention focus.”
However, these mechanisms do not correspond very closely with the vague
descriptors of EI as far as its underlying processes are concerned. The quest
for alternative constructs with better conceptual clarity and fit with the
data revealed emotion regulation as a viable option (Gross, 1998).
Particularly its emergence in management studies (e.g., Lawrence, Troth,
Jordan, & Collins, 2011) prompted me to revisit how the mechanisms of
emotion regulation are defined. While emotion regulation is defined as “the
processes by which individuals influence which emotions they have, when
they have them, and how they experience and express these emotions”
(Gross, 1998, p. 275), two specific antecedent-focused emotion regulation
mechanisms are of relevant here: (i) cognitive change and (ii) attentional
deployment (Gross, 1998). Systematically contrasting the accounts pro-
vided by participants with the literature on cognitive change suggests that,
of all its sub-categories, cognitive re-appraisal is most closely and fre-
quently reflected in the accounts of participants (see e.g., accounts by Nina,
Wayne, and David). Cognitive re-appraisal is defined as “cognitively trans-
forming the situation so as to alter its emotional impact” (Gross, 1998,
p. 284). Applied to the accounts of participants, it can be seen how they
employed re-appraisals in emotionally complex and challenging life situa-
tions, often with the aim to prevent negative effects for their health. This is
consistent with recent longitudinal experimental studies showing that re-
appraisal training leads to lower scores of self-reported negative affect and
stress in the daily lives of individuals (Denny & Ochsner, 2014). Attentional
deployment, by contrast, refers to processes of distraction, concentration,
and rumination. However, the data analysis suggests that distraction and
concentration are more prevalent mechanisms of attentional deployment.
Distraction has been characterized as focusing “attention on nonemotional
aspects of the situation” or moving “attention away from the immediate
situation altogether,” or “changing internal focus” (Gross, 1998, p. 284).
The latter can refer to individuals disengaging from unattainable goals by
shifting their attention to more manageable goals (McIntsoh, 1996). Verbal
accounts that fall into this category can be found in several accounts (see

e.g., case of Chris, Paula and Bill). In terms of concentration (e.g., whether
on work or art), its key characteristic is the “capacity to absorb cognitive
resources” (Gross, 1998, p. 284). It, too, can be applied to diverting atten-
tion to specific emotion elicitors (Gross, 1998). Grandey (2000) provides an
example of this from a previous research project, such that “sometimes … I
have to change my mood and boost my energy to teach … I have to focus
on being positive and maintaining that” (p. 99). This specific example con-
nects plausibly with the account of Wayne (i.e., focusing on having a life
and enjoying it following hospital experience), or Sally (focusing on tasks
she wished to complete at work (i.e., focusing on “this and this”) in order
to induce a positive mental attitude to do better at work).
In sum, it would appear that, when faced with emotionally challenging
situations, the emotional processes that emerged in the data are seem better
aligned with the well-defined processes in the emotion regulation literature.
Consequently, this raises questions about the current utility of ability EI as
a scientific construct at least for now. Needless to say, emotion regula-
tion is one dimension and the apex of the four-branch EI model. Therefore,
these findings should not be interpreted to connote that EI and emotion
regulation exist independent of each other, since other abilities feed into
the ability to regulate one’s emotions. However, as the data analysis has
shown, how the ability to perceive, use, and understand emotions feed into
the ability to regulate emotions is probably less consistent with current the-
orizing on EI. As discussed, currently these processes that seemingly under-
lie EI are not well understood and vaguely defined. Therefore, in the
absence of a proper understanding of how other abilities feed into the abil-
ity to regulate emotions, future research could shift attention to emotion
regulation as a construct that is both conceptually and operationally well-
defined (Lawrence et al., 2011). This could be a temporary state of affairs
until further replication studies have been conducted, and any adjustment
to EI theory and measurement have been completed. Given the close link
between better understanding the underlying processes of EI and how this
can inform intervention at work, the domain of management is ideally sui-
ted to host these further empirical efforts.
The fourth theoretical contribution is related to the outcomes that indi-
viduals experience in response to their ability to process emotional informa-
tion. Taking again the linear assumptions and curvilinear effects as a
starting point for a discussion on perceived individual outcomes, the data
offered several compelling insights. Foremost is that an inability to stop
processing emotional information (e.g., to stop worrying, to explore ways
to reduce emotional loyalty to workplace, or limit one’s empathy for
A Qualitative Study of Emotional Intelligence 131

others) beyond a certain point often resulted in negative health outcomes

(cf. cases of Marie and Bill) or impaired career progression (cf. case of
Terry). This is inconsistent with the depiction of EI as having largely linear
relationships with outcome variables, and is also at variance with the view
that being high in EI (based upon one’s ability to process emotional infor-
mation) always leads to emotional and intellectual growth. Instead, the
ability to regulate one’s emotion (e.g., to stop caring so much emotionally
about one’s work) had the surprising effect of helping Paula to do better in
terms of both her health and career progression (see also narrative of
Andrew). These findings identify a need to be more sensitive in our theo-
rizing on emotional processes and their outcomes to how emotion regu-
lation abilities (or lack thereof) influence the presence of curvilinear effects
and their manifestation in terms of individual health and various outcomes
at work (e.g., performance). Put differently, optimal functioning in terms
of one’s mental health and performance in the workplace may be matter of
“not too much EI.”

Practical Implications

Management interventions aimed at improving employee well-being or per-

formance, for instance, can harness these findings to make better and
informed decisions on what the interventions’ content should be. Since the
processes underlying EI have not been more precisely defined hitherto, it is
plausible to suggest that some EI intervention at work lacked the precision
to impart what EI might be. With the processes better defined, and an
appreciation of a wider range of outcomes, management practitioners can
now more clearly inform their interventions (Lindebaum, in press). This is
likely to render these interventions more successful, as recent examples on
training individuals in re-appraisal underline (Denny & Ochsner, 2014).

Limitations and Future Research

Qualitative research has been criticized for its lack of generalizability

(Kincheloe & Tobin, 2009). The phenomenological data analysis appreci-
ates this limitation to generalize, but importantly makes no such claims of
generalizability or context-free results (Ardley, 2011). Furthermore, I only
included participants from two public sector organizations in the UK in
this study. Thus, findings may not be representative of how participants in

private sector (or even other public sector) organizations would have
responded to the interview questions. Another potential limitation of the
in-depth interview is that of interviewer bias (Antonakis et al., 2004).
However, the involvement of a second experienced researcher served to
mitigate the effects of potential researcher bias, in addition to close adher-
ence to recommendations for phenomenological research (Hycner, 1985).
In terms of future research, several promising avenues emerged. First,
the quest for a better understanding of the role of conscious and subcon-
scious (Fiori, 2009) processes in enabling individuals to both be well and
do well at work could be one such avenue. As the analysis suggests, some
participants were conscious of the need to cognitively transform emotion
elicitors. Doing so over time gradually become more unconscious and auto-
mated. In consequence, future research could examine the antecedents of
conscious processes (and/or factors that impede them), and what processes
are at play as these become more unconscious overtime. Overall, this
should enable management researchers to better understanding the transi-
tion between conscious and subconscious processes, and the outcomes these
might produce. Second, whether or not one is willing, for instance, to use
emotion to influence others may not be an issue of ability but of emotional
display rules at work, according to which individuals feel compelled to act
(Geddes & Callister, 2007). Therefore, future research could explore the
role of display rules in inhibiting the use of emotional abilities.

The data presented suggest that the vaguely defined processes typically
associated with EI could be replaced with more clearly defined constructs
(and their underlying processes) from the emotion regulation literature,
since participants’ accounts could often be mapped onto these constructs
at least for as long as refinement of the EI construct is under way. Of
course, emotion regulation is one dimension and the apex of the four-
branch EI model. As such, other abilities feed into the ability to regulate
one’s emotions. However, as the data analysis has shown, how this occurs
is probably less consistent with current theorizing on EI. Of note, there is
already a tendency to report only the empirical data on this fourth branch
and how it relates to a variety of outcome variables (cf. Kluemper,
DeGroot, & Choi, 2013). Therefore, this study joins a body of literature
that is somewhat skeptical about ability EI as conceptualized and
A Qualitative Study of Emotional Intelligence 133

operationalized in the form of the four-branch ability model of EI

(Antonakis & Dietz, 2011; Føllesdal & Hagtvet, 2013). Of course, I am not
suggesting that we abandon the study of EI as a human ability altogether,
but the data identify several concerns that prompt a rethinking of the con-
struct as well as further psychometric refinement. In sum, for EI to be
investigated with renewed vigor and potential, this data analysis suggests a
need to go back to the drawing board in order to rethink what is meant by
being able to process emotional information at maximum capacity, what
qualifiers need to be recognized, and how to conceptually and operationally
re-define the construct. If this is realized, then scholars and practitioners in
the realm of management and beyond are likely to have a construct at
hand that assist individuals in better harnessing emotional processes to
both be and do well at work.


1. Strictly adhering to these recommendations does not correspond with the

more advantageous approach to gain an in-depth understanding of the data as pre-
sented here. This is consistent with Sanders (1982), who notes that “a precise
methodology … does not exist for phenomenological researchers” (p. 353). For
instance, I considered writing summaries of each interview toward the end of the
analytical process (as suggested by Hycner, 1985) as too late giving that these sum-
maries are essential in familiarizing oneself the data. Also, seeking respondents’
feedback was not possible due to on-going re-structuring of the organizations. This
meant that, at the time the data were analyzed, many participants would not work
in that organization/or department any more. To mitigate this step, and to further
ensure the faithfulness to the phenomenon itself (Knaack, 1984), I frequently para-
phrased the participants’ views in order to receive confirmation (or disconfirmation)
that I understood their accounts according to their perceptions. In fact, I have
sometimes purposely paraphrased incorrectly in order to elicit corrective action on
the part of participants (see Sandberg, 2000, for a similar approach to enhance the
credibility of findings).
2. One key presupposition has been the advocacy of EI on my part for many
years, albeit with a somewhat more differentiate perspective on EI compared to
other contributors of the field. This point serves to mitigate any objections that the
analysis has merely yielded confirmatory findings.
3. I recognize that greater “honesty” can be both a factor which initiates emo-
tional processes and outcome of meaningful life experiences. However, I treat it
here as the former, since subsequent important life events can be approached with a
different mental attitude, which, as Wayne suggests later, implied stopping to pre-
tend being strong when one is not.
4. Gross (1998) also suggests that emotion regulation processes can be “auto-
matic or controlled, conscious or unconscious, and may have their effects at one or

more points in the emotion generative process” (p. 275). This has relevance in con-
junction with some participant’s accounts, in which it is hinted at the perceived role
of conscious or sub/unconscious processes.
5. This quote shows that individuals can even oscillate between cognitive trans-
formation (i.e., “not caring …”) and changes in attention focus (i.e., outside vs. per-
sonal perspective) within a matter of a two sentences.
6. Rumination concerns also directing attention, but attention is channeled
toward feelings and their outcomes (Gross, 1998). This focusing is often directed
toward the negative emotion features of depression (see Gross, 1998, for more infor-
mation). As the narratives have shown, the outcomes of excessive rumination have
often been described as detrimental (e.g., waking up at night because one cannot
switch off).


Alvesson, M., & Sandberg, J. (2011). Theory development: Generating research questions
through problematization. Academy of Management Review, 36(2), 247 271.
Amis, J. M., & Silk, M. L. (2008). The philosophy and politics of quality in qualitative organi-
zational research. Organizational Research Methods, 11(3), 456 480.
Antonakis, J., Ashkanasy, N. M., & Dasborough, M. T. (2009). Does leadership need emo-
tional intelligence? The Leadership Quarterly, 20(2), 247 261.
Antonakis, J., & Dietz, J. (2011). More on testing for validity instead of looking for it.
Personality and Individual Differences, 50(3), 418 421.
Antonakis, J., Schriesheim, C. A., Donovan, J. A., Gopalakrishna-Pillai, K., Pellegrini, E. K., &
Rossomme, J. L. (2004). Methods for studying leadership. In J. Antonakis,
A. T. Cianciolo, & R. J. Sternberg (Eds.), The nature of leadership (pp. 48 70).
Thousand Oaks, CA: Sage.
Ardley, B. (2011). Marketing theory and critical phenomenology: Exploring the human side of
management practice. Marketing Intelligence & Planning, 29(7), 628 642.
Baron, R. A. (2008). The role of affect in the entrepreneurial process. Academy of
Management Review, 33(2), 328 340.
Bryman, A. (2004). Social research methods (3rd ed.). Oxford: Oxford University Press.
Caruso, D. R., & Salovey, P. (2004). The emotionally intelligent manager: How to develop and
use the four key emotional skills of leadership. San Francisco, CA: Jossey-Bass.
Cassell, C., Bishop, V., Symon, G., Johnson, P., & Buehring, A. (2009). Becoming a qualitative
management researcher. Management Learning, 40(5), 513 533.
Clarke, N. (2006). Emotional intelligence training: A case of caveat emptor. Human Resource
Development Review, 5(4), 422 441.
Creswell, J. W., & Plano Clark, V. L. (2011). Designing and conducting mixed methods research.
Thousand Oaks, CA: Sage.
Denny, B. T., & Ochsner, K. N. (2014). Behavioral effects of longitudinal training in cognitive
reappraisal. Emotion, 14(2), 425 433.
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of
Management Review, 14(4), 532 550.
Etherington, K. (2004). Becoming a reflexive researcher: Using ourselves in research. London:
Jessica Kingsley Publishers.
A Qualitative Study of Emotional Intelligence 135

Feldman Barrett, L., Mesquita, B., Ochsner, K., & Gross, J. (2007). The experience of emo-
tion. Annual Review of Psychology, 58, 373 403.
Fiori, M. (2009). A new look at emotional intelligence: A dual-process framework. Personality
and Social Psychology Review, 13(1), 21 44.
Føllesdal, H., & Hagtvet, K. (2013). Does emotional intelligence as ability predict transforma-
tional leadership? A multilevel approach. The Leadership Quarterly, 24(5), 747 762.
Gannon, N., & Ranzijn, R. (2005). Does emotional intelligence predict unique variance in life
satisfaction beyond IQ and personality? Personality and Individual Differences, 38,
1353 1364.
Geddes, D., & Callister, R. R. (2007). Crossing the line(s): A dual threshold model of anger in
organizations. Academy of Management Review, 32(3), 721 746.
Gephart, R. P. (2004). Qualitative research and the academy of management journal. Academy
of Management Journal, 47(4), 454 462.
Gooty, J., Gavin, M. B., Ashkanasy, N. M., & Thomas, J. S. (2014). The wisdom of letting go
and performance: The moderating role of emotional intelligence and discrete emotions.
Journal of Occupational and Organizational Psychology, 87(2), 392 413.
Grandey, A. (2000). Emotion regulation in the workplace: A new way to conceptualize emo-
tional labor. Journal of Occupational Health Psychology, 5(1), 95 110.
Gross, J. (1998). The emerging field of emotion regulation: An integrative review. Review of
General Psychology, 2(3), 271 299.
Hemmati, T., Mills, J. F., & Kroner, D. G. (2004). The validity of the Bar-On emotional intel-
ligence quotient in an offender population. Personality and Individual Differences, 37(4),
695 706.
Henwood, K. L., & Pidgeon, N. F. (1992). Qualitative research and psychological theorizing.
British Journal of Psychology, 83(1), 97 111.
Huy, Q. N. (1999). Emotional capability, emotional intelligence, and radical change. Academy
of Management Review, 24, 325 345.
Hycner, R. H. (1985). Some guidelines for the phenomenological analysis of interview data.
Human Studies, 8(3), 279 303.
Johnson, P., & Cassell, C. (2001). Epistemology and work psychology: New agendas. Journal
of Occupational and Organizational Psychology, 74(2), 125 143.
Jordan, P. J., Dasborough, M., Daus, C. S., & Ashkanasy, N. M. (2010). A call to context.
Industrial and Organizational Psychology, 3, 1 4.
Joseph, D. L., & Newman, D. A. (2010). Emotional intelligence: An integrative meta-analysis
and cascading model. Journal of Applied Psychology, 95(1), 54 78.
Kafetsios, K. (2004). Attachment and emotional intelligence abilities across the life course.
Personality and Individual Differences, 37(1), 129 145.
Kincheloe, J., & Tobin, K. (2009). The much exaggerated death of positivism. Cultural Studies
of Science Education, 4(3), 513 528.
Kluemper, D. H., DeGroot, T., & Choi, S. (2013). Emotion management ability: Predicting
task performance, citizenship, and deviance. Journal of Management, 39(4), 878 905.
Knaack, P. (1984). Phenomenological research. Western Journal of Nursing Research, 6(1),
107 114.
Kunnanatt, J. T. (2004). Emotional intelligence: The new science of interpersonal effectiveness.
Human Resource Development Quarterly, 15(4), 489 495.
Lawrence, S. A., Troth, A. C., Jordan, J. P., & Collins, A. L. (2011). A review of emotion
regulation and development of a framework for emotion regulation in the workplace.
In P. L. Perrewé & D. C. Ganster (Eds.), The role of individual differences in

occupational stress and well being research in occupational stress and well being (Vol. 9,
pp. 199 266. London: Emerald.
Leonard, N. H., & Harvey, M. (2007). The trait of curiosity as a predictor of emotional intelli-
gence. Journal of Applied Social Psychology, 37(8), 1914 1929.
Lincoln, Y. S. (2009). Rethinking emotional intelligence: An alternative proposal. Advances in
Developing Human Resources, 11(6), 784 791.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Newbury Park, CA: Sage.
Lindebaum, D. (2009). Rhetoric or Remedy? A critique on developing emotional intelligence.
Academy of Management Learning and Education 8(2), 225 237.
Lindebaum, D. (in press). Emotional intelligence. In A. Wilkinson & S. Johnstone (Eds.),
Encyclopedia of human resource management. Cheltenham, UK: Edward Elgar.
Lindebaum, D., & Jordan, J. P. (2014). When it can be good to feel bad and bad to feel good:
Exploring asymmetries in workplace emotional outcomes. Human Relations, 67(9),
1037 1050.
Maitlis, S., & Ozcelik, H. (2004). Toxic decision processes: A study of emotion and organiza-
tional decision making. Organization Science, 15(4), 375 393.
Martins, A., Ramalho, N., & Morin, E. (2010). A comprehensive meta-analysis of the relation-
ship between emotional intelligence and health. Personality and Individual Differences,
49(6), 554 564.
Matthews, G., Roberts, R. D., & Zeidner, M. (2004). Seven myths about emotional intelli-
gence. Psychological Inquiry, 15(3), 179 196.
Mayer, J. D., Caruso, D. R., & Salovey, P. (1999). Emotional intelligence meets traditional
standards for an intelligence. Intelligence, 27(4), 267 298.
Mayer, J. D., Roberts, R. D., & Barsade, S. G. (2008). Human abilities: Emotional intelli-
gence. Annual Review of Psychology, 59, 507 536.
Mayer, J. D., & Salovey, P. (1995). Emotional intelligence and the construction and regulation
of feelings. Applied and Preventive Psychology, 4, 197 208.
Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & D. Sluyter
(Eds.), Emotional development and emotional intelligence: Implications for educators (pp.
3 31). New York, NY: Basic Books.
Mayer, J. D., Salovey, P., & Caruso, D. R. (2000). Models of emotional intelligence. In R. J.
Sternberg (Ed.), Handbook of human intelligence (2 ed., pp. 396 420). New York, NY:
Mayer, J. D., Salovey, P., & Caruso, D. R. (2002). Mayer-Salovey-Caruso emotional intelli-
gence test. North Tonawanda, NY: MHS.
Mayer, J. D., Salovey, P., & Caruso, D. R. (2004). Emotional intelligence: Theory, findings,
and implications. Psychological Inquiry, 15(3), 197 215.
Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G. (2003). Measuring emotional intelli-
gence with the MSCEIT V2.0. Emotion, 3(1), 97 105.
McIntsoh, W. D. (1996). When does goal nonattainment lead to negative emotional reactions,
and when doesn’t it?: The role of linking and rumination. In L. L. Martin & A. Tesser
(Eds.), Striving and feeling: Interactions among goals, affect, and self-regulation
(pp. 53 77). Hillsdale, NJ: Lawrence Erlbaum Associates.
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook.
Beverly Hills, CA: Sage.
A Qualitative Study of Emotional Intelligence 137

O’Boyle, J. E. H., Humphrey, R. H., Pollack, J. M., Hawver, T. H., & Story, P. A. (2011). The
relation between emotional intelligence and job performance: A meta-analysis. Journal
of Organizational Behavior, 32(5), 788 818.
Palmer, B. R., Gignac, G., Manocha, R., & Stough, C. (2005). A psychometric evaluation of
the Mayer-Salovey-Caruso emotional intelligence test version 2.0. Intelligence, 33(3),
285 305.
Petrides, K. V., Pita, R., & Kokkinaki, F. (2007). The location of trait emotional intelligence
in personality factor space. British Journal of Psychology, 98(2), 273 289.
Pierce, J. R., & Aguinis, H. (2013). The Too-Much-of-a-Good-Thing effect in management.
Journal of Management, 39(2), 313 338.
Sandberg, J. (2000). Understanding human competence at work: An interpretative approach.
Academy of Management Journal, 43(1), 9 25.
Sanders, P. (1982). Phenomenology: A new way of viewing organizational research. Academy
of Management Review, 7(3), 353 360.
Solomon, R. (1993). The passions: Emotions and the meaning of life. Indianapolis, IN: Hackett
Publishing Company.
Suddaby, R. (2010). Editor’s comments: Construct clarity in theories of management and
organization. Academy of Management Review, 35(3), 346 357.
Thory, K. (2013). Teaching managers to regulate their emotions better: Insights from emo-
tional intelligence training and work-based application. Human Resource Development
International, 16(1), 4 21.
Winkel, D. E., Wyland, R. L., Shaffer, M. A., & Clason, P. (2011). A new perspective on psy-
chological resources: Unanticipated consequences of impulsivity and emotional intelli-
gence. Journal of Occupational and Organizational Psychology, 84(1), 78 94.
Wright, A., Murray, J. P., & Geale, P. (2007). A phenomenographic study of what it means to
supervise doctoral students. Academy of Management Learning & Education, 6(4),
458 474.
This page intentionally left blank


Patricia L. Baratta and Jeffrey R. Spence

The multidimensional structure of boredom poses unique measurement
challenges related to scale length and statistical modeling. We systemati-
cally address these concerns in two studies. In Study 1, we use item
response theory to shorten the 29-item Multidimensional State Boredom
Scale (MSBS) (Fahlman et al., 2013). In Study 2, we use structural
equation modeling to compare two theoretically consistent multidimen-
sional structures of boredom (superordinate and multivariate) with the
most commonly used, yet theoretically inconsistent, structure in boredom
research (unidimensional parallel model). Our findings provide support

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 139 172
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011007

for modeling boredom as multidimensional and demonstrate the impact

of model selection on effect sizes and significance.
Keywords: Boredom, item response theory, multidimensional,
Multidimensional State Boredom Scale, structural equation modeling


Against boredom the gods themselves fight in vain. (Nietzsche, 1888/2006, p. 48)

Boredom is an inextricable part of the human experience. Scholars and

great minds throughout the ages have recognized humanity’s struggle with
boredom. As early as the fourth century AD, Catholic monks referred to
boredom as the “noonday demon” as it was believed to strike monks at
midday, distracting them from their duties and tempting them to leave the
monastery (Wenzel, 1967). Since the introduction of the word boredom to
the English language in the nineteenth century (e.g., Dickens, 1853/2003;
Kemble, 1879), the experience of what it means to be bored has been exam-
ined and refined across many different modern definitions (e.g., Fisher,
1993; Mikulas & Vodanovich, 1993). Despite the definitional variability of
boredom in psychology, two characteristics most researchers can agree on
are that boredom is an unpleasant experience and that it is a common feel-
ing, particularly at work. A recent survey of 500 employees indicated that
over a third regretted their career choice because of boredom (Cook, 2013).
Moreover, The Guardian newspaper reported that in a sample of over 2,000
employees, half of those surveyed reported often feeling bored at work
(Smith, 2006).
Not unlike the Catholic monks of the fourth century AD, organizational
researchers recognize that boredom can be a source of distraction and
decreased productivity. Researchers have found that boredom is associated
with absenteeism, lower performance, counterproductive work behavior,
and job dissatisfaction (Bruursema, Kessler, & Spector, 2011; Drory, 1982;
Kass, Vodanovich, & Callender, 2001; van Hooff & van Hooft, 2014). The
potential for a construct like boredom to drive employee behavior and con-
tribute to organizational ineffectiveness is not surprising; over the last two
decades, organizational researchers have increasingly highlighted the
importance of emotion in shaping employee experiences and productivity
(Ashkanasy & Humphrey, 2011; Barclay, Skarlicki, & Pugh, 2005;
Loukidou, Lou-Clarke, & Daniels, 2009). Although many of the develop-
ments in organizational emotion research have defined and operationalized
Testing the Multidimensional Structure of Boredom 141

emotion using affective dimensions, such as positive and negative affect

(e.g., Beal & Ghandour, 2011; Spence, Ferris, Brown, & Heller, 2011),
researchers have begun to showcase the ability of discrete emotions in shap-
ing organizational attitudes and behavior (e.g., Barclay et al., 2005;
Härtel & Page, 2009; Spence, Brown, Keeping, & Lian, 2014). The consid-
eration of discrete emotions is theoretically valuable as it can give us a
more precise theoretical understanding of the function of emotion in the
workplace by identifying and isolating which emotions uniquely predict
behavior and which emotions are felt in response to events and situations
(Gooty, Gavin, & Ashkanasy, 2009). Analytically, discrete emotions can
account for additional variance in outcomes that is not explained by
broader affective states, increasing the predictive power of research
(e.g., Lee & Allen, 2002; Spence et al., 2014).
Boredom is a discrete emotion that offers intriguing possibilities for
organizational researchers. Specifically, research supports the idea that
boredom’s specific action tendencies (Frijda, 1986; Lazarus, 1991) are
linked to productivity behaviors and quests for stimulation. To date, bore-
dom has been linked to absenteeism, lower performance, and risk taking
and antisocial behaviors (Drory, 1982; Farmer & Sundberg, 1986; Kass
et al., 2001). Despite recognizing the importance of discrete emotions, orga-
nizational researchers wishing to study discrete emotions are impeded by
the absence of psychometrically validated instruments to measure specific
emotions (Brief & Weiss, 2002). In this regard, boredom is no exception as
a lack of a psychometrically sound and practical tool has made boredom
research difficult to conduct (Vodanovich, 2003). In this chapter, we seek
to advance organizational research on boredom by evaluating and refining
a measure of boredom that can be used by organizational researchers. We
also explain and test a theoretical framework articulating how boredom
should be measured, thereby providing organizational researchers with a
novel and rigorous method for measuring boredom.
Although psychometric rigor is an important consideration for all
research (Stanley & Spence, 2014), it is particularly important for boredom
as boredom offers unique measurement challenges as a result of its multidi-
mensional structure (Fahlman, Mercer-Lynn, Flora, & Eastwood, 2013).
Research suggests that boredom is comprised of five dimensions: disengage-
ment, low arousal, high arousal, inattention, and time perception (Fahlman
et al., 2013). Unlike unidimensional constructs, which are comprised of a
single component, multidimensional constructs have multiple distinct, yet
related, dimensions that are treated as a single theoretical concept
(Edwards, 2001). Multidimensional constructs can generate nagging

questions about the scale’s factor structure, such as “how are the dimen-
sions related to each other?” and “how should the scale be treated by
researchers attempting to use the scale?” Researchers have only recently
begun to tackle questions pertaining to boredom’s factor structure, leaving
many assumptions left untested.
In addition to these fundamental psychometric concerns, multidimen-
sional constructs pose practical challenges to organizational researchers
conducting field research. Specifically, multidimensional scales need to be
longer than unidimensional scales as each dimension requires its own sub-
scale to ensure that it is measured with sufficient validity and reliability
(Noar, 2003). However, organizational researchers are often faced with
time and efficiency constraints when trying to collect data from employees
during their workdays. Such practical constraints can lead researchers to
collapse across dimensions, omit dimensions, or opt for a less rigorous
scale if it is more practical. Moreover, long scales can lead to method bias
by inducing fatigue and careless responding (Hinkin, 1995; Schmitt &
Stults, 1985).
In this chapter, we present two studies that address these challenges.
Study 1 uses item response theory (IRT) to shorten a lengthy boredom
scale and Study 2 identifies and tests different multidimensional structures
of boredom using Edwards’ (2001) integrated analytical framework for
multidimensional constructs. The measure we use in both studies is the
Multidimensional State Boredom Scale (MSBS; Fahlman et al., 2013). The
MSBS was chosen as it is the only measure that assesses boredom along its
theoretically derived dimensions. Using a scale that is consistent with the-
ory ensures that we have consistency between the conceptual and opera-
tional definitions of boredom in our research.


Researchers agree that boredom is a complex construct with several under-

lying components (e.g., Fisher, 1993; Goldberg, Eastwood, Laguardia, &
Danckert, 2011). Theory developed by Fahlman et al. (2013) suggests that
boredom is comprised of five distinct dimensions: (1) disengagement,
(2) low arousal, (3) high arousal, (4) inattention, and (5) time perception.
They define boredom as a negative emotional state in which one experi-
ences (1) a longing to engage in satisfying activity, (2) lethargic low arousal
and/or (3) agitated high arousal, (4) an inability to focus attention, and
(5) the perception that time is moving slowly (Fahlman et al., 2013).
Testing the Multidimensional Structure of Boredom 143

Though research has identified boredom as multidimensional, little

research has focused on whether key assumptions inherent in this structure
are true of boredom. Specifically, there are untested assumptions with
respect to how the multiple dimensions of boredom relate to one another
and how the construct should be measured to reflect the proposed struc-
ture; in fact, the most common operationalizations of boredom in research
ignore its underlying dimensions. Importantly, the type of structure one
uses affects how the construct is interpreted theoretically and how it should
be modeled statistically when conducting research (Edwards, 2001; Law,
Wong, & Mobley, 1998; Mowen & Voss, 2008; Wright, Campbell,
Thatcher, & Roberts, 2012). Modeling a multidimensional construct differ-
ently than its theoretical interpretation can lead to erroneous conclusions
as different models may yield different estimates (Law et al., 1998), includ-
ing variance accounted for and the magnitude and significance of correla-
tions. In the case of boredom, a model that does not account for its five
dimensions could produce different results than a model that does.
These misspecifications have both theoretical and practical significance
for organizational research and practice. With respect to research, model-
ing boredom consistently with its interpretation is essential for testing the-
ory on how boredom relates to organizational variables. For example, if a
study shows that the relation between boredom and job satisfaction is not
significant, is this null finding a result of there actually being no effect or
boredom not being modeled properly? Consequently, failure to model bore-
dom appropriately could lead researchers to either wrongly reject or accept
theory. With respect to practice, organizations may fail to solve problems
and incur unnecessary costs by applying research based on theoretically
inappropriate models if these models yield estimates that diverge from true
scores. In the sections below, we specify how boredom can be defined and
treated as a multidimensional construct and explain how the MSBS can be
modeled with different multidimensional structures.

Boredom as a Superordinate Construct

Although not explicitly identified as such, theory suggests that boredom is

a “superordinate” construct, which is a type of multidimensional construct
(see superordinate structure in Fig. 1). More specifically, a superordinate
construct is “a general entity that is manifested or reflected by the dimen-
sions that serve as its indicators” (Williams, Edwards, & Vandenberg, 2003,
pp. 909 910). In other words, boredom is an unobservable or latent

construct that manifests itself in experiences of disengagement, low arousal,

high arousal, inattention, and time perception. Each dimension is also
latent and manifests itself in observed variables (i.e., scale items). An exam-
ple of a superordinate construct in organizational research is transforma-
tional leadership, which can be conceptualized as an overarching construct
that manifests itself in more specific facets of leadership, such as inspira-
tional motivation or intellectual stimulation (Avolio & Bass, 2002).
In structural equation modeling (SEM) software, superordinate con-
structs are depicted as second-order factors, their dimensions as first-order
factors, and the items for each dimension as observed variables (Edwards,
2001; see Fig. 1). The relation between the superordinate construct and its
dimensions can be observed in the factor loadings for each dimension,
which are essentially correlations between the dimension and the superordi-
nate construct. These factor loadings are driven by the relations among the
dimensions with higher factor loadings (i.e., stronger correlations) indicat-
ing that the dimensions share more common variance (Williams et al.,
2003). Variance that is common among all the dimensions is captured by
the superordinate construct and variance that is not common is captured
by the residual or error term for each dimension (Edwards, 2001; Johnson,
Rosen, Chang, Djurdjevic, & Taing, 2012). Because superordinate con-
structs are expected to account for most of the variance in their dimensions,
they should have high factor loadings (Johnson, Rosen, & Chang, 2011;
Wright et al., 2012). When using superordinate constructs in substantive
research, a single statistic can be computed between the superordinate con-
struct and the other construct of interest. Because a superordinate con-
struct combines multiple dimensions into a single overarching construct,
researchers have lauded it for its parsimony and theoretical utility (Ones &
Viswesvaran, 1996).

Boredom as a Multivariate Construct

Although modeling boredom as a superordinate construct is parsimonious

and is consistent with how it is generally conceptualized, combining multi-
ple dimensions into a single theoretical concept is a contentious topic in
organizational research (Edwards, 2001). Alternatively, some researchers
have proposed a “multivariate” structure wherein each dimension is treated
as being related to, but distinct from the other dimensions (see multivariate
structure in Fig. 1; Edwards, 2001). Specifically, “multivariate” constructs
are a set of correlated dimensions and provide an alternative approach to
Testing the Multidimensional Structure of Boredom 145

Fig. 1. Superordinate, Multivariate, and Unidimensional Measurement Model


modeling multiple dimensions. For example, rather than modeling transfor-

mational leadership as a higher-order factor with underlying dimensions
(i.e., superordinate), researchers can model the dimensions (e.g., individual
consideration, inspirational motivation) as a set of correlated variables.
In SEM, the dimensions of the multivariate construct are depicted as
latent variables that are allowed to covary. Unlike superordinate con-
structs, the dimensions of a multivariate structure do not belong to a
higher-order construct, but are treated as a set (i.e., by always simulta-
neously including each of the dimensions in the model). Thus, the distinc-
tion between superordinate and multivariate constructs need not be based
on the construct’s specific conceptual framework as the same construct
(e.g., boredom) can be modeled either way. Instead, it is rooted in broader
theoretical and statistical considerations, some of which we outline below.
Researchers in favor of superordinate constructs argue that they are more
parsimonious in that they refer to a single theoretical concept, rather than a
set of dimensions (multivariate structure); facilitate theory building, because
they are broader than their underlying dimensions; and are better for match-
ing broad predictors to broad outcomes (Hanisch, Hulin, & Roznowski,
1998; Johnson et al., 2011; Law et al., 1998; Ones & Viswesvaran, 1996).
Other researchers contend that the parsimony of superordinate constructs is
achieved at the expense of ambiguity in that variation in the superordinate
construct may suggest variation in one or all of its dimensions (Edwards,
2001; Hattie, 1985). Additionally, superordinate constructs can obscure
meaning as predictors/outcomes may relate differently to the superordinate
construct than they do to one or more of its dimensions (Johns, 1998). In
contrast, multivariate structures enable one to examine the relations between
the predictors/outcomes and each dimension; however, this degree of specifi-
city is thought to occur at the expense of parsimony (Edwards, 2001).
Though there is debate regarding the relative merits of each structure,
these concerns are largely testable (Edwards, 2001). For example, research-
ers can examine the extent to which superordinate constructs obscure rela-
tions between predictors and dimensions. In the present study, we sought
to empirically test these considerations by comparing how each of these
multidimensional structures performed using a multidimensional boredom
scale (i.e., MSBS).

Boredom as a Unidimensional Construct

Regardless of which structure one chooses (i.e., superordinate or multivari-

ate), theory consistently identifies boredom as having multiple dimensions.
Testing the Multidimensional Structure of Boredom 147

Yet, when using multidimensional boredom scales in substantive research,

researchers tend to collapse across dimensions by computing a total or
aggregate score, which is consistent with the interpretation that boredom is
“unidimensional” (e.g., Fahlman, Mercer, Gaskovski, Eastwood, &
Eastwood, 2009; Fahlman et al., 2013; Goldberg et al., 2011; see unidimen-
sional structure in Fig. 1). Unidimensional constructs are distinct from
multidimensional constructs in that they refer to a single theoretical con-
cept with no separable dimensions.
Ignoring the dimensions by computing a single average makes the
assumption that the items are parallel indicators. That is, each item contri-
butes equally to the total score, is of equal quality (i.e., contains the same
amount of true score and error), and can theoretically act as a substitute
for any other item on the scale (Graham, 2006). With respect to boredom,
taking an aggregate score is equivalent to claiming that an item on the
“low arousal” dimension (e.g., “I feel lonely”) is interchangeable with an
item on the “inattention” dimension (e.g., “My attention span is shorter
than usual”). Given the disparate content of items that reflect different
boredom dimensions, the assumptions necessitated by a unidimensional
parallel model would seem overly restrictive. However, this is how
researchers typically operationalize boredom when using a multidimen-
sional scale (e.g., Fahlman et al., 2009; Fahlman et al., 2013; Mercer-Lynn,
Bar & Eastwood, 2014). Whether it is out of convenience or an unaware-
ness of alternatives, computing a total score from the MSBS makes strong
theoretical and measurement assumptions that can influence statistical rela-
tions between boredom and other constructs (e.g., validity), both in terms
of statistical significance and the magnitude of effect sizes. The implications
of taking an aggregate or total score for a multidimensional scale can be
tested empirically by comparing this approach to multidimensional struc-
tures (Edwards, 2001).

Model Comparisons

Multidimensional structures and a unidimensional parallel model make

very different assumptions about the structure of boredom that can influ-
ence interpretations, findings, and conclusions; yet each of these structures
explained above can be used with the same measure. The appropriateness
and accuracy of different structures can be tested using SEM (Edwards,
2001). To assess the implications of different model structures on boredom
research, we analyzed and compared each structure using the MSBS.

Specifically, we compared each of the multidimensional structures to a uni-

dimensional parallel structure and tested the relative merits of a multivari-
ate structure over a superordinate one using Edwards’ (2001) integrative
analytical framework for evaluating multidimensional constructs (Study 2).
Our goal was to evaluate the appropriateness of each of these structures
for studying boredom and to demonstrate the potential implications these
choices could have on research findings.



In addition to questions about the multidimensional structure of the

MSBS, the length of the MSBS poses its own challenges as it contains 29
items. Organizational researchers caution against the use of lengthy mea-
sures as such measures can lead to common method bias resulting from
carelessness or fatigue (Hinkin, 1995; Schmitt & Stults, 1985). Moreover,
researchers faced with time and efficiency constraints when conducting field
research also may find it impractical to use long scales (Hinkin, 1995).
With respect to the MSBS, researchers have already opted for alternative
measures, citing scale length as their primary concern (e.g., Eastwood,
Fenske, Smilek, & Frischen, 2013); others have reduced the MSBS without
reference to the shortened scale’s psychometric properties (e.g., reliability,
content validity; Markey, Chin, Vanepps, & Loewenstein, 2014; Merrifield,
2010). For these reasons, we sought to systematically shorten the 29-item

Item Response Theory

To shorten the MSBS we used IRT to select the most optimal items. IRT
provides data about each item’s ability to discriminate between respondents
at different levels of the underlying trait (e.g., differentiating between those
with low vs. high levels of boredom). More technically, because responses
to the MSBS are made on a seven-point Likert scale, we used a polytomous
IRT model known as a graded response model (GRM; Samejima, 1969).
IRT computes the probability of an observed response, such as choosing
the response category “agree,” based on qualities of the item as well as
respondents’ standing on the latent trait being assessed. This information is
Testing the Multidimensional Structure of Boredom 149

provided on “item response category characteristics curves.” In GRM mod-

els, the probability of endorsing each response category for every item is
estimated at different levels of the underlying trait.
Item response category curves indicate the probability of selecting a spe-
cific response category for a particular item at different levels of the under-
lying trait. For example, it will indicate the probability of selecting
“somewhat agree” for the item “I feel lonely” at different levels of the
underlying trait. Peaked curves indicate that the probability of selecting
that response option changes more severely at different levels of the latent
construct. Smaller ranges and higher peaks indicate that the item provides
more information and precision about the construct than flatter and wider
curves. The steepness of the response curves correspond to an “item discri-
minability parameter” (α). The higher the discrimination parameter, the
more peaked the response curves. High discriminability parameters indicate
a greater amount of precision in an item’s ability to differentiate people at
different levels of the underlying trait. In addition, GRM models provide
“item information curves” (IIC), which are functions indicating how much
information the item provides about individuals at different levels of the
underlying trait. The higher the information function, the more informative
the item is. Information can be thought of as a signal-to-noise ratio with
more information indicating higher levels of signal to noise (Sadler &
Woody, 2004). To shorten the MSBS, we sought to retain items that were
best at discriminating between levels of boredom and that provided the
most information.



To select the best items for a shortened version of the MSBS, we conducted
GRM analysis on the MSBS and considered discriminability parameters,
item response category curves, and item information functions for each
item. We sought to retain items that were best at discriminating people at
different levels of boredom (item discrimination parameters over 2.00) and
that provided the most precise estimates of boredom (tall IIC). Due to
space restrictions, we do not present the item response category curves for
each item, just the discrimination parameters and IICs. To conduct these
analyses, we used the GRM procedure as implemented in the ltm package

in R (Rizopoulos, 2006). Because IRT requires a large sample size

(Embretson & Reise, 2000), we recruited a large sample to complete the

Participants and Procedure

The present study consisted of 1,310 undergraduate psychology students

from a mid-sized university. Participants were recruited online using the
participant pool and were compensated with course credit in exchange for
completing the MSBS. 70.5% of the sample was female and the average
age was 19.54 years.


Multidimensional State Boredom Scale

Respondents completed the 29-item MSBS (see Table 1) and responded to
each item based on their current feelings using a 7-point Likert scale (1 =
“strongly disagree,” to 7 = “strongly agree”).

Assumption of Local Independence

GRM models require that the assumption of local independence be met.

Therefore, prior to performing our IRT analyses, we tested if the data met
the assumption of local independence. Because the MSBS is a multidimen-
sional scale, each factor can be examined as a unidimensional subscale
(Edwards, 2009). To ascertain whether the data met the assumption of local
independence, we examined whether or not single factors explained the
data for each of the five dimensions. Following the recommendation of
Edwards (2009), we used a combination of exploratory and confirmatory
factor analyses, using eigenvalues and model fit indices, respectively. For
model fit, we considered the comparative fit index (CFI) and the root mean
square error of approximation (RMSEA). Results of these analyses indi-
cated that the data met the assumption of local independence with a single
factor satisfactorily explaining the relations amongst the items for each of
Testing the Multidimensional Structure of Boredom 151

Table 1. Full Multidimensional State Boredom Scale.a

Item Dimension

MSBS-1 Time is passing by slower than usual Time perception

MSBS-2 I am stuck in a situation that I feel is irrelevant Disengagement
MSBS-3 I am easily distracted Inattention
MSBS-4 I am lonely Low arousal
MSBS-5 Everything seems to be irritating me right now High arousal
MSBS-6 I wish time would go by faster Time perception
MSBS-7 Everything seems repetitive and routine to me Disengagement
MSBS-8 I feel down Low arousal
MSBS-9 I seem to be forced to do things that have no value to me Disengagement
MSBS-10 I feel bored Disengagement
MSBS-11 Time is dragging on Time perception
MSBS-12 I am more moody than usual High arousal
MSBS-13 I am indecisive or unsure of what to do next Disengagement
MSBS-14 I feel agitated High arousal
MSBS-15 I feel empty Low arousal
MSBS-16 It is difficult to focus my attention Inattention
MSBS-17 I want to do something fun, but nothing appeals to me Disengagement
MSBS-18 Time is moving very slowly Time perception
MSBS-19 I wish I was doing something more exciting Disengagement
MSBS-20 My attention span is shorter than usual Inattention
MSBS-21 I am impatient right now High arousal
MSBS-22 I am wasting time that would be better spent on something else Disengagement
MSBS-23 My mind is wandering Inattention
MSBS-24 I want something to happen but I’m not sure what Disengagement
MSBS-25 I feel cut off from the rest of the world Low arousal
MSBS-26 Right now it seems like time is passing slowly Time perception
MSBS-27 I am annoyed with the people around me High arousal
MSBS-28 I feel like I’m sitting around waiting for something to happen Disengagement
MSBS-29 It seems like there’s no one around for me to talk to Low arousal
Bolded items were retained in the shortened 15-item version of the scale.

the subscales: disengagement (CFI = .92; RMSEA = .09; eigenvalue for sin-
gle factor = 4.91); low arousal (CFI = .98; RMSEA = .11; eigenvalue for
single factor = 3.55); high arousal (CFI = .99; RMSEA = .05; eigenvalue
for single factor = 3.35); inattention (CFI = .99; RMSEA = .08; eigenvalue
for single factor = 2.67); and time perception (CFI = .99; RMSEA = .08;
eigenvalue for single factor = 3.78). We used a congeneric model and
exploratory analyses with principle axis factoring for each subscale. After
determining that the data met the assumption of local independence, we
performed the GRM analyses.

GRM Analyses

Results for the disengagement subscale indicated that items 24 (“I want
something to happen but I’m not sure what”) and 28 (“I feel like I’m sitting
around waiting for something to happen”) were the most effective at discri-
minating between levels of disengagement and provided the most informa-
tion about the underlying trait with discrimination statistics over 2.00
(Table 2). Inspection of the IIC for the disengagement items revealed that
items 24 and 28 provided more information about disengagement than the
other items as evidenced by higher functions (Fig. 2). Given that items 24
and 28 had the highest level of discrimination and IIC, we retained these

Low Arousal
GRM results for the low arousal subscale revealed that all the items had
discrimination parameters above 2.00 (Table 2); these were items 4 (“I am
lonely”), 8 (“I feel down”), 15 (“I feel empty”), 25 (“I feel cut off from the
rest of the world”), and 29 (“It seems like there’s no one around for me to
talk to”). Inspection of the IIC (Fig. 2) demonstrates that all the items (4,
8, 15, 25, and 29) are effective indicators of low arousal as each item has a
comparatively high function. As a result, we did not eliminate any of the
low arousal items.

High Arousal
Results for the high arousal subscale indicated that items 5 (“Everything
seems to be irritating me right now”), 12 (“I am more moody than usual”),
14 (“I feel agitated”), and 27 (“I am annoyed with the people around me”)
had discriminability parameters over 2.00 (Table 2). Inspection of the IIC
(Fig. 2) illustrates that items 5, 12, and 14 had considerably higher func-
tions than item 27. As a result, although item 27 met the 2.00 threshold, it
underperformed relative to the other items. Thus, we only retained items 5,
12, and 14.

The results for the inattention subscale indicated that items 16 (“It is diffi-
cult to focus my attention”) and 20 (“My attention span is shorter than
usual”) had discrimination parameters (α) over 2.00 (Table 2) and the IIC
indicated that these two items had the highest information functions of the
four items (Fig. 2). Thus, we retained items 16 and 20.
Testing the Multidimensional Structure of Boredom 153

Table 2. Graded Response Model Multidimensional State Boredom Scale

Item and Scale Item Parameter Estimates

αa β1 b β2 β3 β4 β5 β6

MSBS-2 1.53 −1.78 −.31 .22 .93 1.97 3.07
MSBS-7 1.38 −2.63 −.97 −.22 .41 1.71 3.15
MSBS-9 1.85 −1.76 −.47 .03 .49 1.53 2.61
MSBS-10 1.94 −2.05 −.85 −.33 .16 1.18 2.39
MSBS-13 1.46 −2.37 −.99 −.45 .05 1.11 2.27
MSBS-17 1.92 −1.35 −.24 .24 .61 1.40 2.40
MSBS-19 1.87 −2.30 −1.22 −.83 −.34 .69 1.77
MSBS-22 1.40 −2.46 −.88 −.25 .24 1.21 2.34
MSBS-24 2.09 −2.09 −.95 −.55 −.08 .85 1.87
MSBS-28 2.42 −1.59 −.45 −.06 .33 1.12 2.14
Low arousal
MSBS-4 2.42 −1.53 −.56 −.16 .15 .96 1.89
MSBS-8 2.99 −1.36 −.34 .08 .46 1.25 2.17
MSBS-15 3.39 −.80 .07 .39 .78 1.43 2.04
MSBS-25 2.65 −1.12 −.05 .38 .70 1.35 2.28
MSBS-29 2.40 −.90 .03 .43 .75 1.41 2.07
High arousal
MSBS-5 3.26 −1.35 −.27 .17 .58 1.35 2.24
MSBS-12 2.92 −1.40 −.33 −.08 .54 1.23 2.13
MSBS-14 3.21 −1.38 −.30 .18 .56 1.41 2.41
MSBS-21 1.66 −1.85 −.49 .13 .63 1.64 2.67
MSBS-27 2.01 −1.45 −.23 .22 .69 1.67 2.81
MSBS-3 1.85 −2.84 −1.58 −.97 −.47 .57 1.79
MSBS-16 3.51 −1.68 −.72 −.26 .10 .94 1.85
MSBS-20 2.36 −1.84 −.54 −.06 .41 1.30 2.38
MSBS-23 1.90 −2.25 −.94 −.50 −.15 .87 2.02
Time perception
MSBS-1 1.37 −1.66 −.20 .72 1.48 2.78 4.12
MSBS-6 1.80 −1.20 −.04 .43 .99 1.84 2.60
MSBS-11 2.74 −1.31 −.13 .48 1.06 1.97 3.14
MSBS-18 3.39 −.94 .01 .51 1.05 1.75 2.99
MSBS-26 2.94 −1.10 .04 .56 1.06 1.78 2.99
α = Item discrimination parameter.
β = Level of underlying trait where there is a .50 probability of responding in current
category or above. 1 = disagree, 2 = slightly disagree, 3 = neutral, 4 slightly agree, 5 = agree,
6 = strongly agree.

Item Information Curves Item Information Curves

MSBS-24 3.5 MSBS-4 MSBS-25

1.8 MSBS-28 MSBS-8 MSBS-29
1.6 MSBS-7

1.4 2.5

1.2 2.0

1.0 1.5

0.8 1.0

0.6 0.5

0.4 0.0
–2 –1 0 1 2 –2 –1 0 1 2
Ability Ability
Disengagement Low Arousal

Item Information Curves Item Information Curves

3.0 MSBS-12 MSBS-27 3.5 MSBS-16 MSBS-23

2.5 3.0


2.0 2.5

1.5 2.0

1.0 1.5
–2 –1 0 1 2 –2 –1 0 1 2
Ability Ability
High Arousal Inattention

Item Information Curves

3.5 MSBS-1 MSBS-18
3.0 MSBS-11


–2 –1 0 1 2
Time Perception

Fig. 2. Item Information Curves for MSBS Dimensions.

Time Perception
Results for the time perception subscale indicated that items 11 (“Time is
dragging on”), 18 (“Time is moving very slowly”), and 26 (“Right now it
seems like time is passing slowly”) had discrimination parameters above
2.00 (Table 2). The item information functions for the time perception
Testing the Multidimensional Structure of Boredom 155

subscale illustrate that items 11, 18, and 26 clearly have the highest func-
tions (Fig. 2). Given these results, we retained items 11, 18, and 26.

We retained 15 MSBS items for the MSBS-15 (items are bolded in
Table 1).



In Study 2, we sought to test the performance of different measurement

models for boredom using the MSBS-15 from Study 1. Following
Edwards’ (2001) framework for evaluating multidimensional constructs, we
compared the superordinate, multivariate, and unidimensional models dis-
cussed earlier. In each model, we included antecedents that have been theo-
retically and empirically linked to boredom in order to provide a basis for
our interpretation of the different model structures (e.g., Fahlman et al.,
2009; Fahlman et al., 2013; Goldberg et al., 2011; Mercer-Lynn, Flora,
Fahlman, & Eastwood, 2013). The inclusion of these constructs enabled us
to test the implications of choosing different model structures for effect size
estimates and statistical significance when predicting boredom.
Specifically, we included four relevant antecedents in each model:
depression, neuroticism, attention deficit/hyperactivity disorder (ADHD),
and patience. We chose depression as it relates to each of boredom’s five
dimensions. Depression is characterized by sadness/loss of interest and lack
of motivation (Goldberg et al., 2011) and is often accompanied by irritabil-
ity, difficulty concentrating, and perceiving time to move slowly (American
Psychiatric Association, 2013; Hawkins, French, Crawford, & Enzle, 1988).
Neuroticism overlaps with boredom’s arousal dimensions as neurotic indi-
viduals are prone to negative affect (e.g., sadness, irritability; Costa &
McCrae, 2010). ADHD is characterized by chronic attentional difficulties
(Mercer-Lynn et al., 2013), relating to the inattention dimension of bore-
dom. Thus, we expected that individuals with higher levels of these con-
structs would be more likely to feel bored. Finally, patience requires calmly
enduring frustrating or difficult situations (Schnitker, 2012), suggesting
that patient individuals should be less likely feel bored in response to their

Statistical Methodology

We used SEM to evaluate each model using AMOS 21. Predictor variables
(i.e., depression, neuroticism, ADHD, and patience) were modeled as
observed variables to ensure that lack of model fit was due to our boredom
measurement models and not to other constructs. Model comparisons were
made using the Akaike information criterion (AIC), the Bayes information
criterion (BIC), and, in accordance with standards for nested model com-
parisons, chi-square difference tests. Model fit was assessed using the CFI
and RMSEA.
Following Edwards (2001), we also considered the criterion-related
validity of each model using R2 for the superordinate and unidimensional
models and multivariate R2 for the multivariate model, which “represents
the proportion of generalized variance in a set of dependent variables,
explained by one or more independent variables” (Edwards, 2001,
p. 163).
For the superordinate model, we tested a parallel, tau-equivalent, and
congeneric model in order to incorporate different assumptions regarding
measurement models (Graham, 2006). These constraints were added at the
higher-order level (i.e., between the superordinate construct and its dimen-
sions) whereas the lower-order level (i.e., between each dimension and its
items) followed a congeneric model in each instance.

Participants and Procedure

In a single lab session, we administered the MSBS-15 from Study 1 in

addition to measures of depression, neuroticism, ADHD, and patience
to 190 undergraduate psychology students from a mid-sized university.
Participants were recruited using the participant pool and were compen-
sated with course credit. 74.7% of the sample was female and the aver-
age age was 19.17 years. Data from these measures are displayed in
Table 3.


We measured depression using the 7-item depression subscale of the
Depression Anxiety Stress Scales-21 (DASS-21; Lovibond & Lovibond,
Testing the Multidimensional Structure of Boredom
Table 3. Means, Standard Deviations, Reliabilities, and Correlations for all Measures.
Mean S.D. 1 2 3 4 5 6 7 8 9

1. Disengagement 3.59 1.77 (.82)

2. Low arousal 3.11 1.60 .70** (.90)
3. High arousal 3.03 1.54 .56** .73** (.89)
4. Inattention 3.96 1.75 .54** .52** .61** (.85)
5. Time perception 2.93 1.60 .45** .47** .44** .42** (.93)
6. Depression .84 .76 .55** .78** .52** .45** .34** (.91)
7. Neuroticism 2.94 .78 .48** .59** .63** .46** .35** .55** (.86)
8. ADHD 2.15 .71 .48** .51** .39** .52** .28** .54** .51** (.82)
9. Patience 3.46 .87 −.12 −.13 −.18* −.16* −.12 .00 −.20** −.15* (.68)

*p < .05; **p < .01.


1995). Participants responded to each item based on their feelings over the
past week using a 4-point scale (0 = “did not apply to me at all,” to 3 =
“applied to me very much, or most of the time”).

We measured neuroticism using the 10-item neuroticism subscale of the
International Personality Item Pool (Goldberg, 1992). Participants
responded to each item using a 5-point scale (1 = “very inaccurate,” to 5 =
“very accurate”).

We measured ADHD using the 9-item inattention subscale of the Adult
ADHD Self-rating Scale (Kessler et al., 2005). Participants responded to
each item based on their feelings over the last six months using a 5-point
scale (0 = “never,” to 4 = “very often”).

We measured patience using the 3-item daily hassles patience subscale of
the 3-Factor Patience Questionnaire (Schnitker, 2012). Participants
responded to each item using a 5-point scale (1 = “not like me at all,” to 5
“very much like me”).

Superordinate Models

The results for boredom as a superordinate model are displayed in Fig. 3.

The parallel, tau-equivalent, and congeneric superordinate models fit the
data well. The CFI values were .90, .93, and .94 for the parallel, tau-
equivalent, and congeneric models, respectively. The RMSEA values were
.09, .08, and .08 for parallel, tau-equivalent, and congeneric models, respec-
tively. These values are in accordance with criteria stipulating that CFI
values above .90 indicate acceptable fit (Hooper, Coughlan, & Mullen,
2008) and RMSEA values between .08 and .10 indicate mediocre fit
(MacCallum, Browne, & Sugawara, 1996). Chi-square difference tests
revealed that the congeneric model fit the data significantly better than did
the tau-equivalent model (Δχ 24 = 23:63, p < .001) and that the tau-equivalent
model fit the data significantly better than did the parallel model
Testing the Multidimensional Structure of Boredom 159

Fig. 3. Superordinate Model Results.

(Δχ 24 = 70:80, p < .001). The AIC and BIC values for the parallel, tau-
equivalent, and congeneric models were 475.17, 412.36, and 396.74, and
608.29, 558.48, and 555.84, respectively, with lower values indicating better
fit (Kline, 2005). Overall, the congeneric model satisfied model fit conven-
tions and provided the best fit to the data.
The r2 value between boredom and each dimension for the parallel
model was .61. For the tau-equivalent model, the r2 values between bore-
dom and disengagement, low arousal, high arousal, inattention, and time
perception were .66, .92, .72, .52, and .44, respectively. For the congeneric
model, the r2 values between boredom and disengagement, low arousal,
high arousal, inattention, and time perception were .66, .96, .71, .46, and
.27, respectively. Overall, as the models became less restrictive, boredom
accounted for less variance in time perception and inattention, whereas dis-
engagement remained relatively constant. In addition, boredom accounted

for more variance in low arousal and high arousal in the least restrictive
models (i.e., tau-equivalent and congeneric) compared to the most restric-
tive model (i.e., parallel).
The R2 values relating boredom to depression, neuroticism, ADHD, and
patience were .67, .72, and .72 for the parallel, tau-equivalent, and congene-
ric models, respectively. For all three models, the standardized regression
coefficients relating boredom to each predictor indicated that boredom was
significantly positively related to depression and neuroticism, but did not
have a statistically significant relation with patience. We present these
results below.

Parallel Model
The results for the parallel model indicated that boredom was significantly
positively related to depression (β = .45, p < .001), neuroticism (β = .34,
p < .001), and ADHD (β = .15, p < .05), but not statistically significantly
related to patience (β = −.10, p = .06).

Tau-Equivalent Model
The results for the tau-equivalent indicated that boredom was significantly
positively related to depression (β = .55, p < .001), and neuroticism (β =
.32, p < .001), but not significantly related to ADHD (β = .10, p = .09), or
patience (β = −.09, p = .06).

Congeneric Model
The results for the congeneric model indicated that boredom was signifi-
cantly positively related to depression (β = .58, p < .001), and neuroticism
(β = .30, p < .001), but not significantly related to ADHD (β = .08, p =
.15), or patience (β = −.09, p = .06).
Overall, the superordinate model fit the data well with the congeneric
model providing the best fit to the data. When comparing across each
model, the results of the tau-equivalent and congeneric models were con-
sistent. This is evident when considering the statistical significance and
magnitude of the standardized regression coefficients as well as the var-
iance accounted for in boredom. The parallel model had slightly different
results. The amount of variance accounted for in boredom was lower
and, although ADHD was statistically significant in the parallel model, it
was not statistically significant in the tau-equivalent and congeneric
Testing the Multidimensional Structure of Boredom 161

Multivariate Structural Model

The multivariate structural model fit the data well (Δχ 2120 = 195:52, p <
.001; CFI = .97; RMSEA = .06) with the CFI and RMSEA values comply-
ing with conventions for good model fit (Hu & Bentler, 1999). The multi-
variate structural model fit the data significantly better than did the
superordinate congeneric model (Δχ 221 = 103:22, p < .001), which provided
the best fit to the data of the superordinate models. The AIC and BIC
values for the multivariate structural model were 335.52 and 562.81, respec-
tively, and tended to be lower than those of the superordinate models, indi-
cating better fit (Kline, 2005), with the exception of the BIC values for the
congeneric and tau-equivalent models, which were 555.84 and 558.48,
The multivariate R2 value relating the predictor variables to the dimen-
sions as a set was .87. This value is substantially higher than the R2 values
for any of the superordinate models, which ranged from .67 to .72. The R2
values relating each dimension to the predictor variables were .45, .74, .50,
.40, and .17 for disengagement, low arousal, high arousal, inattention, and
time perception, respectively. Below, we present the results for each predic-
tor variable.

Depression and Boredom Dimensions

Depression was significantly positively related to disengagement (β =
.39, p < .001), low arousal (β = .67, p < .001), high arousal, (β = .29,
p < .001), inattention (β = .20, p < .05), and time perception (β = .20,
p < .05).

Neuroticism and Boredom Dimensions

Neuroticism was significantly positively related to disengagement (β = .21,
p < .05), low arousal (β = .24, p < .001), high arousal (β = .49, p < .001),
inattention (β = .19, p < .05), and time perception (β = .20, p < .05).

ADHD and Boredom Dimensions

ADHD was significantly positively related to disengagement (β = .20, p <
.05), and inattention (β = .35, p < .001), but not statistically significantly
related to low arousal (β = .04, p = .47), high arousal (β = −.01, p = .88),
or time perception (β = .08, p = .37).

Patience and Boredom Dimensions

Patience was not statistically significantly related to disengagement (β =
−.05, p = .44), low arousal (β = −.08, p = .09), high arousal (β = −.09, p =
.12), inattention (β = −.08, p =.21), or time perception (β = −.08, p = .25).
Overall, the multivariate model fit the data significantly better than did
the superordinate model and accounted for substantially more variance in
boredom. With respect to statistical significance, the results of the multi-
variate model were relatively consistent with the superordinate model in
that depression and neuroticism were significant predictors of boredom
whereas patience was not; ADHD was significantly related to some dimen-
sions, but not others. The magnitude of these relations varied in that some
predictors were more strongly related to some dimensions than others.

Unidimensional Parallel Model

The results for boredom as a unidimensional parallel model are displayed

in Fig. 4. The fit of this model was not acceptable according to model fit
conventions. The CFI and RMSEA values were .68 and .16, respectively,
and do not comply with conventions for fit criteria as CFI values below .90
indicate less than acceptable fit (Hooper et al., 2008) and RMSEA values
above .10 indicate poor fit (MacCallum et al., 1996). Chi-square difference
tests indicated that the unidimensional parallel model fit the data

Fig. 4. Unidimensional Parallel Model Results.

Testing the Multidimensional Structure of Boredom 163

significantly worse than did the multivariate model (Δχ 254 = 804:96, p <
.001) and the superordinate congeneric model (Δχ 233 = 701:74, p < .001),
which provided the best fit of the three congeneric models. The AIC and
BIC values were 1032.48 and 1084.43, respectively, and are substantially
higher than the AIC and BIC values for the multivariate and superordinate
models, indicating worse fit (Kline, 2005). Overall, the unidimensional par-
allel model was not an acceptable fit to the data and performed worse than
each of the multidimensional models.
The R2 value relating boredom to depression, neuroticism, ADHD, and
patience was .64. The standardized coefficients relating boredom to each
predictor variable indicated that boredom was significantly positively
related to depression (β = .48, p < .001), neuroticism (β = .31, p < .001),
ADHD (β = .13, p < .05), and patience (β = −.10, p < .05).
Overall, the unidimensional model did not comply with model fit conven-
tions and accounted for considerably less variance than did the multidimen-
sional models. Boredom’s relations with each of the predictor variables were
somewhat consistent with the superordinate models. Namely, depression was
the strongest predictor and ADHD the weakest; however, in the unidimen-
sional parallel model, patience was a significant predictor of boredom.


These findings suggest that it is appropriate to model boredom as either a

superordinate or multivariate structure, both of which are consistent with
the interpretation that boredom is multidimensional. Model fit statistics
indicated that both structures fit the data well and, when modeled as super-
ordinate, boredom demonstrated strong relations with each of its dimen-
sions. When comparing across the two models, although some model
comparison indices (i.e., chi-square difference tests and AICs) favored the
multivariate model in terms of model fit, others (i.e., BICs) favored the
superordinate model, highlighting the appropriateness of both structures.
However, the multivariate model had higher criterion-related validity than
each of the superordinate models (multivariate R2 = .87; R2 for superordi-
nate models ranged from .67 to .72).
With respect to the statistical significance of boredom’s relations with
the predictor variables, the results were highly consistent when comparing
the superordinate and multivariate models. Boredom as a superordinate
construct was significantly positively related to depression and neuroticism
and non-statistically significantly related to patience. Likewise, for the

multivariate model, depression and neuroticism were significantly positively

related to each dimension whereas patience was not. However, there were
inconsistencies when considering the magnitude of these relations in that in
the multivariate model indicated that some predictors were more strongly
related to some dimensions than others. For example, the standardized
regression coefficients between depression and each of the dimensions ran-
ged from .20 to .67. This variation was not captured by the single regres-
sion coefficient between depression and the superordinate construct of
boredom (e.g., β = .58 for the congeneric model), suggesting that the super-
ordinate construct obscured the relations between depression and the
dimensions. Neuroticism, on the other hand, demonstrated comparable
coefficients with each dimension with the exception of high arousal, making
the multivariate results highly similar to those of the superordinate model.
Thus, the parsimony achieved by modeling boredom as superordinate
obscured the relations between predictors and dimensions in some cases,
but not others.
When comparing the unidimensional parallel model to the superordinate
models, the statistical significance and magnitude of the standardized
regression coefficients were mostly consistent. In fact, estimates were just as
likely to diverge within model type (e.g., superordinate parallel vs. congene-
ric) as they were between model types (i.e., superordinate vs. unidimen-
sional). Though these estimates were highly similar, model fit indices
indicated that the unidimensional parallel model was a poor fit to the data,
suggesting that the assumptions inherent in this model are not applicable to
a multidimensional scale, such as the MSBS.


Boredom is a discrete emotion that is frequently experienced at work and is

theoretically linked to employee performance, productivity, and deviance
(Bruursema et al., 2011; Drory, 1982; Smith, 2006). However, organiza-
tional research on boredom has been impeded by measurement challenges
resulting from boredom’s multidimensional structure. Specifically, multidi-
mensional scales tend to be longer, making them impractical for field
research and increasing their susceptibility to some forms of response bias
(Hinkin, 1995; Noar, 2003). Additionally, there are many untested assump-
tions with respect to how multidimensional boredom scales should be
treated. We sought to address these specific challenges by providing
Testing the Multidimensional Structure of Boredom 165

organizational researchers with a shortened scale and tests of different mea-

surement model structures. In Study 1, we used IRT to select the best per-
forming items from the 29-item MSBS to form a shortened 15-item scale.
In Study 2, we used the MSBS-15 to test and compare different structures
of boredom using Edwards’ (2001) framework for multidimensional
constructs. We tested a theoretically consistent structure of boredom
(superordinate) against a viable alternative multidimensional structure
(multivariate) and against the most commonly used structure in research
(unidimensional parallel model).
The results from Study 2 provide support for current theory defining
boredom as multidimensional. Specifically, we found that modeling bore-
dom as either a multivariate or superordinate construct demonstrated good
model fit. In addition, when modeled as superordinate, boredom had
strong factor loadings with each dimension, suggesting that it is meaningful
to treat boredom’s dimensions as a single theoretical concept. Though each
model fit the data comparably well, the findings highlight different benefits
for each approach.
First, our findings highlight the advantages of multivariate structures
in that boredom had substantially higher criterion-related validity when
modeled as multivariate than when modeled as superordinate. This indi-
cates that researchers might lose out on explained variance by opting for
a superordinate structure. The benefits of choosing a superordinate struc-
ture, however, may outweigh these costs. Specifically, our findings under-
score the parsimony inherent in modeling boredom as superordinate. We
were able to easily compute a single regression coefficient between bore-
dom and its predictors when it was modeled as superordinate, but not
when it was modeled as multivariate. Instead, for the multivariate model,
we had to examine the relation between each predictor and a single
dimension. Researchers who theorize about multidimensional constructs
at a broad level may find it difficult to interpret their results when the
construct is operationalized as a set of dimensions (i.e., multivariate)
compared to when the dimensions are treated as a single theoretical con-
cept (i.e., superordinate). When using a multivariate model, for example,
if researchers hypothesize that boredom should significantly predict job
satisfaction, is the underlying theory supported when only some of the
dimensions show a statistically significant relation with job satisfaction?
Thus, superordinate constructs may be preferable for addressing general
questions about boredom, whereas multivariate constructs may be prefer-
able for addressing specific questions tailored to each dimension
(Edwards, 2001).

In addition, our findings address claims that the parsimony inherent in

superordinate constructs distorts meaning by obscuring relations between
predictors and dimensions. When only considering statistical significance,
our analyses revealed that the relations between the predictors and dimen-
sions from the multivariate model were entirely captured when boredom
was modeled as superordinate. When considering the magnitude of these
relations, the superordinate construct obscured the relations for some pre-
dictors (e.g., depression), but not others (e.g., neuroticism, patience). These
findings suggest that the more parsimonious superordinate model was asso-
ciated with the added cost of obscuring meaning some of the time.
Finally, our findings shed light on the practice of operationalizing bore-
dom as an aggregate or total score, which is equivalent to a unidimensional
parallel model in SEM. Surprisingly, we found that the results were rela-
tively consistent when comparing the unidimensional parallel model to
each of the superordinate models. However, the model fit indices caution
against operationalizing boredom as a unidimensional parallel model
(i.e., single aggregate score). The fit of the unidimensional parallel model
was far from acceptable (CFI = .68 and RMSEA =.16), indicating that the
model assumptions imposed by a unidimensional parallel structure are not
appropriate when using the MSBS.


By addressing measurement challenges specific to boredom, our study

makes several important contributions to help stimulate organizational
research on boredom. First, using IRT, we reduced the lengthy MSBS from
29 to 15 items. To date, the MSBS is the only measure that assesses bore-
dom along its five dimensions and is therefore an invaluable tool for organi-
zational researchers. The MSBS-15 is an important step in the development
of this tool as it helps alleviate concerns regarding common method bias
associated with lengthy measures like the 29-item MSBS (Hinkin, 1995;
Schmitt & Stults, 1985). We also expect that the MSBS-15 will be more fea-
sible for use in field research as it will be less of a burden for participants
and researchers faced with time and space constraints. The existence of a
shortened scale also makes it less likely that researchers will compromise the
validity of the scale by omitting dimensions or important items, as has been
the case previously with the MSBS (e.g., Markey et al., 2014).
In addition to systematically pruning the MSBS, this study is the first to
systematically examine and compare multidimensional structures for
Testing the Multidimensional Structure of Boredom 167

boredom. Our findings provide empirical support for theory suggesting

that boredom is multidimensional and provide evidence regarding
the appropriateness of alternative structures. In the past, researchers have
often operationalized boredom by taking a total or aggregate score
(e.g., Fahlman et al., 2009; Fahlman et al., 2013). Based on the current
results, the assumptions required by a single aggregate score do not appear
to be statistically defensible. We present alternative structures and demon-
strate how researchers can use SEM to model boredom as a multidimen-
sional construct. These results provide researchers with a foundation for
measuring and modeling boredom in their future research.
Finally, this research contributes to the multidimensional construct
debate by addressing concerns regarding the utility of multidimensional
constructs. Specifically, our study supports claims that superordinate con-
structs are theoretically meaningful and parsimonious (e.g., Ones &
Viswesvaran, 1996) and indicates that though they sometimes distort rela-
tions between predictors and dimensions (Johns, 1998), this is not always
the case. Moreover, consistent with past research, our findings demonstrate
that multivariate structures are advantageous in that they are associated
with higher criterion-related validity (Edwards, 2001).

Limitations and Future Research

Although our research possesses a number of strengths, there are some lim-
itations. First, though we did manage to substantially reduce the number
of items in the MSBS, some may argue that the MSBS-15 is still too long
for some research purposes. However, when considering scale length, it is
important for researchers to take into account whether or not the construct
being measured is multidimensional. Multidimensional scales need to be
longer as each dimension requires its own subscale to ensure that it is mea-
sured with sufficient validity. Given theory stipulating that boredom is
comprised of five dimensions (Fahlman et al., 2013) and that a typical uni-
dimensional scale in organizational research has between three and five
items (Hinkin, 1995), it is reasonable for a measure of boredom to contain
15 items.
One potential limitation of Study 2 is that we relied on self-report data,
raising the possibility that our results are due to common method variance
(Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). However, we believe
that our choice to use self-report measures should not negate the contribu-
tion of our findings. Specifically, some of the measures we used either

required participants to respond based on their current feelings (e.g., bore-

dom) or their feelings over the last week (i.e., depression) and six months
(i.e., ADHD). Had we relied on other-report data, it is unlikely that these
individuals would have been able to precisely estimate participants’ internal
states, especially when these feelings were restricted to specific time frames.
Although we also measured traits (i.e., neuroticism and patience), which
can be assessed using other-reports, research has shown that self- and
other-reports of personality are highly correlated (Barrick, Patton, &
Haugland, 2000). This indicates that self-reports are a sufficient and reli-
able means of measuring personality. Additionally, each measure we used
had a different scale format, which has been suggested to reduce common
method variance (Podsakoff et al., 2003). For these reasons, we do not
believe that our choice to use self-report should undermine the contribution
of our findings.
A second limitation of Study 2 is that, although SEM language suggests
a causal relation between predictor and outcome variables, the single-time
study design precludes us from making conclusions about causality. We
should note, however, that the goal of the present research was to evaluate
and compare different structures for boredom rather than to make causal
inferences between boredom and other constructs.


French philosopher Roland Barthes said, “It cannot be helped: boredom is

not simple” (1973/1975, p. 25). Many of the complications that arise from
the study of boredom stem from its multidimensional structure, which
necessitates more elaborate theory and more sophisticated measurement
techniques. In this chapter, we address theoretical, statistical, and practical
issues related to boredom: (a) we tested the theoretical proposition that
boredom is a multidimensional construct, (b) we demonstrated the effect of
modeling boredom using different structures on model fit, effect sizes, and
statistical significance, and (c) we shortened the MSBS by selecting the
most optimal items according to IRT.


American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders
(5th ed.). Washington, DC: American Psychiatric Association.
Testing the Multidimensional Structure of Boredom 169

Ashkanasy, N. M., & Humphrey, R. H. (2011). Current emotion research in organizational

behavior. Emotion Review, 3, 214 224.
Avolio, B. J., & Bass, B. M. (2002). Manual for the multifactor leadership questionnaire (Form
5X). Redwood City, CA: Mindgarden.
Barclay, L. J., Skarlicki, D. P., & Pugh, S. D. (2005). Exploring the role of emotions in injus-
tice perceptions and retaliation. Journal of Applied Psychology, 90, 629 643.
Barrick, M. R., Patton, G. K., & Haugland, S. N. (2000). Accuracy of interviewer judgments
of job applicant personality traits. Personnel Psychology, 53, 925 951.
Barthes, R. (1975). The pleasure of the text. (R. Miller, Trans.). New York, NY: Hill and
Wang. (Original work published 1973).
Beal, D. J., & Ghandour, L. (2011). Stability, change, and the stability of change in daily
workplace affect. Journal of Organizational Behavior, 3, 526 546.
Brief, A. P., & Weiss, H. M. (2002). Organizational behavior: Affect in the workplace. Annual
Review of Psychology, 53, 279 307.
Bruursema, K., Kessler, S. R., & Spector, P. E. (2011). Bored employees misbehaving: The
relationship between boredom and counterproductive work behaviour. Work & Stress,
25, 93 107.
Cook, D. (2013). Many regret their career choice. Benefits Pro. Retrieved from http://www.
benefitspro.com/2013/08/01/many-regret-their-career-choices. Accessed on November
11, 2014.
Costa, P. T., & McCrae, R. R. (2010). NEO personality inventory-3: Interpretive report. Lutz,
FL: Psychological Assessment Resources.
Dickens, C. (2003). In. N. Bradbury (Ed.), Bleak house. London: Penguin Books. (Original
work published 1853).
Drory, A. (1982). Individual differences in boredom proneness and task effectiveness at work.
Personnel Psychology, 35, 141 151.
Eastwood, J. D., Fenske, M. J., Smilek, D., & Frischen, A. (2013). The role of acute boredom
in problem gambling. Retrieved from http://www.gamblingresearch.org/content/
research.php?appid=2815. Accessed on March 12, 2014.
Edwards, J. R. (2001). Multidimensional constructs in organizational behavior research: An
integrative analytical framework. Organizational Research Methods, 4, 144 192.
Edwards, M. C. (2009). An introduction to item response theory using the needs for cognition
scale. Social and Personality Compass, 3/4, 507 529.
Embretson, S. E., & Reise, S. P. (2000). Item response theory for psychologists. Mahwah, NJ:
Lawrence Erlbaum Associates.
Fahlman, S. A., Mercer, K. B., Gaskovski, P., Eastwood, A. E., & Eastwood, J. D.
(2009). Does a lack of life meaning cause boredom? Results from psychometric, longitu-
dinal, and experimental analyses. Journal of Social and Clinical Psychology, 28,
307 340.
Fahlman, S. A., Mercer-Lynn, K. B., Flora, D. B., & Eastwood, J. D. (2013). Development
and validation of the Multidimensional State Boredom Scale. Assessment, 20, 68 85.
Farmer, R., & Sundberg, N. D. (1986). Boredom proneness: The development and correlates
of a new scale. Journal of Personality Assessment, 50, 4 17.
Fisher, C. D. (1993). Boredom at work: A neglected concept. Human Relations, 46, 395 417.
Frijda, N. H. (1986). The emotions. Cambridge: Cambridge University Press.
Goldberg, L. R. (1992). The development of markers for the big-five factor structure.
Psychological Assessment, 4, 26 42.

Goldberg, Y. K., Eastwood, J. D., Laguardia, J., & Danckert, J. (2011). Boredom: An emo-
tional experience distinct from apathy, anhedonia, or depression. Journal of Social and
Clinical Psychology, 30, 647 666.
Gooty, J., Gavin, M., & Ashkanasy, N. M. (2009). Emotions research in OB: The challenges
that lie ahead. Journal of Organizational Behavior, 30, 833 838.
Graham, J. M. (2006). Congeneric and (essentially) tau-equivalent estimates of score reliabil-
ity: What they are and how to use them. Educational and Psychological Measurement,
66, 930 944.
Hanisch, K. A., Hulin, C. L., & Roznowski, M. (1998). The importance of individuals’ reper-
toires of behaviors: The scientific appropriateness of studying multiple behaviors and
general attitudes. Journal of Organizational Behavior, 19, 463 480.
Härtel, C. E., & Page, K. M. (2009). Organizational citizenship behavior and workplace deviance:
The role of affect and cognitions. Journal of Managerial Psychology, 24, 237 253.
Hattie, J. (1985). Methodology review: Assessing unidimensionality of tests and items. Applied
Psychological Measurement, 9, 139 164.
Hawkins, W. L., French, L. C., Crawford, B. D., & Enzle, M. E. (1988). Depressed affect and
time perception. Journal of Abnormal Psychology, 97, 275 280.
Hinkin, T. R. (1995). A review of scale development practices in the study of organizations.
Journal of Management, 21, 967 988.
Hooper, D., Coughlan, J., & Mullen, M. R. (2008). Structural equation modelling: Guidelines
for determining model fit. The Electronic Journal of Business Research Methods, 6,
53 60.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling: A
Multidisciplinary Journal, 6, 1 55.
Johns, G. (1998). Aggregation or aggravation? The relative merits of a broad withdrawal con-
struct. Journal of Organizational Behavior, 19, 453 462.
Johnson, R. E., Rosen, C. C., & Chang, C. H. (2011). To aggregate or not to aggregate: Steps
for developing and validating higher-order multidimensional constructs. Journal of
Business and Psychology, 26, 241 248.
Johnson, R. E., Rosen, C. C., Chang, C. H. D., Djurdjevic, E., & Taing, M. U. (2012).
Recommendations for improving the construct clarity of higher-order multidimensional
constructs. Human Resource Management Review, 22, 62 72.
Kass, S. J., Vodanovich, S. J., & Callender, A. (2001). State-trait boredom: Relationship to
absenteeism, tenure, and job satisfaction. Journal of Business and Psychology, 16,
317 327.
Kemble, F. A. (1879). Records of a girlhood. New York, NY: Henry Holt & Co.
Kessler, R. C., Adler, L., Ames, M., Demler, O., Faraone, S., Hiripi, E., … Walters, E. E.
(2005). The world health organization adult ADHD Self-report scale (ASRS): A short
screening scale for use in the general population. Psychological Medicine, 35, 245 256.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New
York, NY: The Guilford Press.
Law, K. S., Wong, C. S., & Mobley, W. M. (1998). Toward a taxonomy of multidimensional
constructs. Academy of Management Review, 23, 741 755.
Lazarus, R. S. (1991). Emotion and adaptation. New York, NY: Oxford University Press.
Lee, K., & Allen, N. J. (2002). Organizational citizenship behavior and workplace deviance:
The role of affect and cognitions. Journal of Applied Psychology, 87, 131 142.
Testing the Multidimensional Structure of Boredom 171

Loukidou, L., Lou-Clarke, J., & Daniels, K. (2009). Boredom in the workplace:
More than monotonous tasks. International Journal of Management Reviews, 11,
381 405.
Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the depression anxiety stress scales
(2nd ed.). Sydney: Psychology Foundation.
MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determi-
nation of sample size for covariance structure modeling. Psychological Methods, 1,
130 149.
Markey, A., Chin, A., Vanepps, E. M., & Loewenstein, G. (2014). Identifying a reliable bore-
dom induction. Perceptual & Motor Skills, 119, 237 253.
Mercer-Lynn, K. B., Bar, R. J., & Eastwood, J. D. (2014). Causes of boredom: The person,
situation, or both? Personality and Individual Differences, 56, 122 126.
Mercer-Lynn, K. B., Flora, D. B., Fahlman, S. A., & Eastwood, J. D. (2013). The measure-
ment of boredom: Differences between existing self-report scales. Assessment, 20,
585 596.
Merrifield, C. (2010). Characterizing the psychophysiological signature of boredom.
Unpublished masters’ thesis, University of Waterloo, Waterloo.
Mikulas, W. L., & Vodanovich, S. J. (1993). The essence of boredom. Psychological Record,
43, 3 12.
Mowen, J. C., & Voss, K. E. (2008). On building better construct measures: Implications of a
general hierarchical model. Psychology & Marketing, 25, 485 505.
Nietzsche, F. W. (2006). The antichrist. (H. L. Mencken, Trans.). Binghamton, NY: Vail-
Ballou Press. (Original work published 1888).
Noar, S. M. (2003). The role of structural equation modeling in scale development. Structural
Equation Modeling: A Multidisciplinary Journal, 10, 622 647.
Ones, D. S., & Viswesvaran, C. (1996). Bandwidth-fidelity dilemma in personality measure-
ment for personnel selection. Journal of Organizational Behavior, 17, 609 626.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method
biases in behavioral research: A critical review of the literature and recommended reme-
dies. Journal of Applied Psychology, 88, 879 903.
Rizopoulos, D. (2006). ltm: An R package for latent variable modeling and item response the-
ory analyses. Journal of Statistical Software, 17, 1 25.
Sadler, P., & Woody, E. Z. (2004). Four decades of group hypnosis scales: What does item-
response theory tell us about what we’ve been measuring? The International Journal of
Clinical and Experimental Hypnosis, 52, 132 158.
Samejima, F. (1969). Estimation of latent ability using response pattern of graded scores.
Psychometrika Monograph Supplement, 34, 100 114.
Schmitt, N. W., & Stults, D. M. (1985). Factors defined by negatively keyed items: The results
of careless respondents? Applied Psychological Measurement, 9, 367 373.
Schnitker, S. A. (2012). An examination of patience and well-being. The Journal of Positive
Psychology, 7, 263 280.
Smith, A. (2006). Teachers less likely to be bored in their jobs, survey finds. The Guardian.
Retrieved from http://www.theguardian.com/education/2006/jul/27/schools.uk.
Accessed on March 1, 2014.
Spence, J. R., Brown, D. J., Keeping, L. M., & Lian, H. (2014). Helpful today, but not tomor-
row: Feeling grateful as a predictor of daily organizational citizenship behavior.
Personnel Psychology, 67, 705 738.

Spence, J. R., Ferris, D. L., Brown, D. J., & Heller, D. (2011). Understanding daily citizenship
behaviors: A social comparison perspective. Journal of Organizational Behavior, 32,
547 571.
Stanley, D. J., & Spence, J. R. (2014). Expectations for replications: Are yours realistic?
Perspectives in Psychological Science, 9, 305 318.
van Hooff, M. L. M., & van Hooft, E. A. J. (2014). Boredom at work: Proximal and distal
consequences of affective work related boredom. Journal of Occupational Health
Psychology, 19, 348 359.
Vodanovich, S. J. (2003). Psychometric measures of boredom: A review of the literature.
Journal of Psychology, 137, 569 593.
Wenzel, S. (1967). The sin of sloth: Acedia in medieval thought and literature. Chapel Hill, NC:
University of North Carolina Press.
Williams, L. J., Edwards, J. R., & Vandenberg, R. J. (2003). Recent advances in causal model-
ing methods for organizational and management research. Journal of Management, 29,
903 936.
Wright, R. T., Campbell, D. E., Thatcher, J. B., & Roberts, N. (2012). Operationalizing multi-
dimensional constructs in structural equation modeling: Recommendations for IS
research. Communications of the Association for Information Systems, 30, 367 412.
This page intentionally left blank



Olof Brunninge and Anders Melander


In this chapter, we explore the impact of socioemotional and financial

wealth on the resource management of family firms. We use MoDo, a
Swedish pulp and paper firm, covering three generations of owner-
managers from 1873 to 1991, to grasp the shifting emphases on socioe-
motional and financial wealth in the management of the company.
Identifying four strategic issues of decisive importance for the develop-
ment of MoDo, we analyze the organizational values that guided the
management of these issues. We propose that financial and socioemo-
tional wealth stand for two different rationalities that infuse organiza-
tional values. The MoDo case illustrates how these rationalities go hand
in hand for extended periods of time, safeguarding both financial success
and socioemotional endowments. However, in a situation where the

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 175 213
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011008

rationalities are no longer in line with the development of the industry

context, the conflict arising between the two rationalities may have fatal
consequences for the firm in question.
Keywords: Family business; resource management; socioemotional
wealth; organizational rationalities; guiding values


Even if the majority of firms around the globe are family businesses
(Gomez-Mejia, Takács-Haynes, Núnez-Nickel, Jacobson, & Moyano-
Fuentes, 2007; Jaskiewicz, Combs, & Rau, 2014) only quite recently, the
study of family firms has emerged as a distinct field of study within man-
agement research. Within a short time-span, various specialized journals
(e.g., Family Business Review, Journal of Family Business Strategies), along
with well-reputed management and entrepreneurship journals (e.g., Journal
of Management Studies, Entrepreneurship Theory and Practice, and Journal
of Business Venturing), have published a substantial amount of high-quality
family business research. One major reason for the rapidly growing interest
in family firms is the observation that models and perspectives developed
to understand public companies do not always adequately address how
family firms manage their resources. Based on a review of resource-
management literature, we will discuss the applicability of traditional
resource-management theories to family firms and aim to identify particu-
lar challenges facing family firms when managing their resources over time.
We propose that the socioemotional wealth (SEW) perspective (Berrone,
Cruz, & Gomez-Mejia, 2012; Berrone, Cruz, Gomez-Mejia, & Larraza-
Kintana, 2010; Gomez-Mejia et al., 2007) offers a complementary view, in
order to better account for nonfinancial considerations in a family firm’s
resource management. Our key argument is that a purely financial rational-
ity focusing on profit maximization is insufficient for understanding how
family firms deploy their resources and manage strategic choices. Rather,
such firms also try to maximize their SEW (Berrone et al., 2010, 2012;
Gomez-Mejia et al., 2007). Gomez-Mejia et al. (2007, p. 106) define SEW
as “nonfinancial aspects of the firm that meet the family’s affective needs,
such as identity, the ability to exercise family influence, and the perpetua-
tion of the family dynasty.” Pursuing SEW does of course not exclude an
Family Firm Longevity and Resource Management 177

interest in financial wealth. However, it implies that SEW complements

and in some cases competes with the financial wealth rationality we
know from studies of widely held corporations.
This chapter explores the impact of SEW on the resource management
of family firms. In doing so, we present and discuss the concept of SEW
and show how family firms handle the socioemotional and financial ration-
alities in management processes.
It is undisputed that companies need to renew themselves in order to
remain successful in a changing industry context (Sirmon & Hitt, 2003;
Teece, Pisano, & Shuen, 1997). Hence, in mature family firms the ability to
review, reconstruct, and develop resources becomes an increasingly impor-
tant metaskill for ensuring long-term survival. This has been highlighted
recently in an emerging stream of literature focusing on resource manage-
ment in family firms (e.g., Sharma & Manikutty, 2005; Sirmon & Hitt,
2003) as well as in businesses in general (e.g., Ireland, Hitt, & Sirmon,
2003; Sirmon, Hitt, Ireland, & Gilbert, 2011). Based on the resource-based
view of the firm (Barney, 1991), particular emphasis has been put on the
family-related resources that are supposed to be unique to family firms
(Habbershon & Williams, 1999; Sirmon & Hitt, 2003). So far the resource-
management literature has been successful in identifying resources that
may contribute to family firm performance (Eddleston, Kellermanns, &
Sarathy, 2008; Pearson, Carr, & Shaw, 2008; Sirmon & Hitt, 2003) or
building conceptual models outlining how family firms manage their
resources (Sirmon & Hitt, 2003). With few exceptions (e.g., Salvato &
Melin, 2008), there is however little work done on empirically studying
how family firms create and manage their resource configurations over
time. Investigating the resource-management process of family firms, by
definition, requires a longitudinal research design. In addition, the manage-
ment of resources needs to be contextualized not only by considering the
immediate family and company contexts, but also including the industry
context where the resources are supposed to result in a competitive
advantage (Miller & Shamsie, 1996). So far, much of the family business lit-
erature has been focusing on internal family business dynamics (Benavides-
Velasco, Quintana-Garcıa, & Guzman-Parra, 2013). This is not surprising
as the family dynamics’ interaction with the business system constitutes the
distinctive character of family firms. However, we also need to link family
and family business processes to changes in an industry context to under-
stand why and how family businesses manage to compete and survive in
changing industries (e.g., Zellweger, Nason, Nordqvist, & Brush, 2011).

Empirically, this chapter is built upon the case of the Swedish pulp and
paper firm MoDo. The firm existed from 1873 to 1991, covering three gen-
erations of owner-managers. The long-term nature of the study will allow
us to see the shifting emphases on socioemotional and financial wealth
rationalities in the resource management of the company. Our research
aims to make the following contributions: (1) we contribute to the under-
standing of the longevity of family firms by conducting a truly processual
study of resource management in a family enterprise, accounting for both
financial and SEW rationalities, (2) we contribute to the family business lit-
erature by linking socioemotionally and financially driven rationalities in
family firms to the developments in the industry context, and (3) we contri-
bute to the literature on resource management in family firms by highlight-
ing the role of SEW and empirically studying a resource-management
process over an extended period of time.

In this frame of reference we develop a theoretical basis for understanding
the particular challenges facing family firms when managing their
resources. First, we introduce the resource-management literature, which
typically focuses on a financial wealth rationality. In the second part of the
framework, we review and evaluate the potential of the literature on SEW
to increase our understanding of resource management over time.
When trying to understand the long-term determinants of family firm
performance, the role of resources has played a central role in scholarly
efforts. This is not surprising, considering the dominance of the resource-
based view of the firm (RBV). Conceptualizing the firm as a bundle of
resources, the RBV assumes that companies can achieve sustainable compe-
titive advantage based on unique resources that need to be valuable, rare,
inimitable, and nonsubstitutable (Barney, 1991). When investigating firm
resources in a family business setting, scholars have been particularly inter-
ested in those resources that are potentially unique to family firms as they
are grounded in the interaction of family, its individual members, and the
business. Habbershon and Williams (1999) labeled these resources as the
familiness of a given firm. The familiness resources can be seen as a basis
for transgenerational wealth creation if they are used to build a portfolio of
businesses by an enterprising family (Habbershon & Pistrui, 2002). There
have been some attempts to identify typical familiness resources. For
Family Firm Longevity and Resource Management 179

instance, Sirmon and Hitt (2003) list human capital, social capital, surviva-
bility capital, patient capital, and governance structure, while Habbershon
and Williams (1999) provide an even more detailed enumeration, including,
for example, trust in family stakeholders, participation, or family routines.
Although the familiness perspective is attractive especially for understand-
ing the differences between family and nonfamily firms (Chrisman, Chua, &
Steier, 2005), the crucial issue for competitiveness on the individual firm level
is not whether the company’s unique resources are grounded in its character
as a family firm, but whether its resource configuration in its totality is
unique. Even a family business might build its competitive advantage on
resources that are not related to the family or a combination of familiness
and other resources.
A key aspect for the longevity of a business is the flexibility of its
resource configuration. The value of resources is contextual and changes
over time (Barney, 1991). Resources that have historically underpinned the
firm’s competitive advantage are not necessarily useful when the industry
environment changes, for example, through the emergence of new technolo-
gies (Anderson & Tushman, 1990). In their seminal work on dynamic cap-
abilities, Teece et al. (1997) conceptualize competitive advantage as lying in
the firm’s processes, positions, and paths. However, these characteristics of
the firm are also a potential source of inertia. Therefore, previously success-
ful resource configurations need to be adapted dynamically in order not to
turn into rigidities. The literature on dynamic capabilities defined as “the
capacity of an organization to purposefully create, extend or modify its
resource base” (Helfat, 2007, p. 1) addresses this continuous revisiting. This
changing of the resource base is easier said than done, as the very idea of
building the firm’s business on a strong set of unique resources may be
hard to combine with the need to quickly change and adapt the resource
configuration (Schreyögg & Kliesch-Eberl, 2007).
Resource management essentially implies that the resource configura-
tion, that is, the firm’s set of resources and their relationship to each other
at any given point in time, needs to be adapted to a changing business con-
text. Interestingly, most resource-management models, although implicitly
assuming that resource management aims at better pursuing opportunities
in the firm’s business context, are largely decontextualized. The main focus
lies on managing internal processes around resources. Grant (1991) pro-
poses a rationalist planning model where resource gaps are identified in the
firm’s strategy process. The firm then needs to invest in replenishing, aug-
menting, and upgrading its resource base in order to achieve a fit with its
strategic direction. Basically the process of resource building is linear,

involving a feedback loop that stands for the continuous reviewing and
updating of resources. It is implicitly taken for granted that all resource-
management decisions follow a financial rationality aiming at maximizing
the firm’s business success. The model has been adapted to a family busi-
ness context by Habbershon and Williams (1999) who sees process inter-
ventions, such as family meetings, management team meetings, and
consultant facilitations, as means for evaluating and changing resources.
Likewise studying resource management in family firms, Sirmon and Hitt
(2003) propose a more complex resource-management model that is similar
to that of Grant (1991) in being rationalist and linear (with a feedback
loop). It consists of three sequential components. First, the firm makes an
inventory of its resources, adding and shedding resources if necessary. In
the subsequent steps, resources are bundled and leveraged in order to
enhance the firm’s business model, before the process returns to its point of
departure and a new inventory begins. Sirmon and Hitt (2003) address
some differences between resource management in family firms and their
nonfamily counterparts. For example, they assume that family firm man-
agers are more knowledgeable about their firm’s resources and manage
resources with less time pressure. Still the model assumes that the genera-
tion of financial wealth is the main rationality behind family firm resource
management. Nonfinancial aspects only play a limited role as nostalgia and
emotional ties might potentially disturb financially rational resource shed-
ding decisions.
Such conceptual models of resource management are appealing as they
offer a structure for escaping from the problem of path-dependence that
lies in a firm’s resources. However, they potentially suffer from the same
weaknesses as other planning models that are essentially rational, linear,
and normative. For instance, Mintzberg’s classical criticism of strategic
planning (Mintzberg, 1990) partially also applies to resource-management
models. Even if a rational approach to resource management was desirable
from a normative perspective, how likely is it that firms being stuck in day-
to-day problems will engage in a structured resource-management process?
Will managers be able to analyze their environment in an unbiased manner,
given that their cognitive frameworks are structures that have developed
historically? (Johnson, 1987) Historical resource configurations play an
important role in shaping cognitive structures (Brunninge & Melin, 2010)
so we cannot necessarily assume that managers are able to reconsider the
appropriateness of their firm’s resources in an “objective” manner.
Moreover, normative resource-management models have limitations in
addressing how the adding and shedding of resources actually occur.
Family Firm Longevity and Resource Management 181

Sirmon and Hitt (2003) give us some guidance in stating that the involve-
ment of nonfamily members in family firms’ decision-making processes can
facilitate change. This is in line with general findings that adding cognitive
diversity to governance structures facilitates change (Brunninge, Nordqvist, &
Wiklund, 2007). In addition, the long-term perspective on businesses that
family firms often take, can make it easier for family firms to engage in the
arduous process of reconfiguring their resources (Sirmon & Hitt, 2003).
Still, this does not mean that adding and shedding resources is easily done.
The conceptual frameworks on resource management in family firms thus
need to be complemented with empirical studies that investigate how the
resource configurations of family firms actually develop over extended peri-
ods of time and context. This would allow us to understand not only how
resources are rationally managed, but also how resource configurations
evolve and change in their industry context, beyond conscious managerial
What are then the particular challenges family businesses meet when
managing their resources? The growing literature on SEW addresses this
question and tells us that family businesses tend to make decisions accord-
ing to a rationality that differs from that of firms that do not have clear
family involvement (Berrone et al., 2010, 2012; Gomez-Mejia et al., 2007).
Research applying the SEW perspective has shown that family firms some-
times put less emphasis on financial outcomes in order to maximize the
SEW of the family owners. For example, they may take financial risk,
avoid dependence on outsiders (Gomez-Mejia et al., 2007), put more
emphasis on sustainability in order to safeguard family reputation (Berrone
et al., 2010), and are reluctant to transfer ownership to outsiders as they
see family control as part of their socioemotional endowment (Zellweger,
Kellermanns, Chrisman, & Chua, 2012). According to Berrone et al. (2010)
SEW is characterized by five dimensions, being (1) family control and influ-
ence that the dominant owners want to maintain, (2) family members’ iden-
tification with the firm, meaning that the owners’ personal identity is
intimately tied to that of the firm, (3) binding social ties, leading to social
bonds and trust inside and outside the family, (4) emotional attachment,
which implies that emotional factors often interact with business factors in
the family firm, and (5) renewal of family bonds to the firm through dynas-
tic succession. For resource management, this implies that decisions about
the adding, shedding, and use of resources are at least partly guided by con-
siderations about what resource decisions will imply for existing socioemo-
tional endowments. A problematic aspect of SEW that Berrone et al.’s
(2010) suggested dimensions reveal, is that different dimensions may point

to different directions for strategic choices (Gomez-Mejia, Cruz, &

Imperatore, 2014). While the desire to maintain family control and influ-
ence (dimension one) might tempt the firm to engage in dishonest behavior,
family firm identity would rather do the opposite, as family members would
seek to avoid anything that might put the reputation of the family and its
business at stake (dimension two). Although SEW is a rationality in the
sense that it provides alternative decision-making criteria when compared
to the financial wealth rationality, it does not necessarily always point to a
clear strategic direction (Gomez-Mejia et al., 2014). Neither does it per se
deny financial considerations.
The heterogeneity of SEW types and sources makes it challenging to
empirically capture the phenomenon, particularly in quantitative studies
where SEW can be difficult to operationalize. Miller and Le Breton-Miller
(2014) criticize studies on SEW for often relying on indirect measures, such
as governance variables quantifying family involvement and then just
assuming that for instance correlations between high family involvement
and risk aversion are due to concerns about SEW. Here, in-depth qualita-
tive studies might provide a clearer understanding of the impact of SEW
on strategic choices.
Despite problems with the SEW construct, it still offers an interesting
opportunity to complement traditional resource-management models, over-
emphasizing a financial wealth rationality. Gomez-Mejia, Cruz, Berrone,
and Castro (2011) suggest that SEW should be given at least equal impor-
tance in studying decision making in family firms. The SEW literature thus
conjures up a picture of two different rationalities influencing resource-
management decisions; a financial wealth rationality that is primarily
oriented toward ensuring business success in its changing industry context,
and a SEW rationality, primarily relating to family dynamics (De Tienne &
Chirico, 2013). This suggests that resource management will become parti-
cularly challenging in choices where the two rationalities point in different
An interesting question concerning the significance of SEW is of course
under what circumstances a SEW rationality might dominate over a finan-
cial wealth rationality and vice versa. This is of course eventually a ques-
tion that is decided by the priorities set by the dominant coalition of the
family firm. De Tienne and Chirico (2013) argue that in a family firm hold-
ing a portfolio of firms, not all firms have the same level of SEW associated
with them. Even in family firms where SEW plays an important role, the
importance of the financial wealth rationality as opposed to the SEW
rationality might thus differ depending on what particular resource is
Family Firm Longevity and Resource Management 183

concerned. Sciascia, Mazzola, and Kellermanns (2014) found that generally

SEW preservation needs tend to decrease with firm age. They build their
argument on the observation that later generations in a family firm often put
a strong emphasis on creating financial wealth (Van Gils, Voordeckers, &
van den Heuvel, 2004). This might be due to the emergence of family
branches pursuing particularistic needs and showing less identification with
the family business as a whole (De Tienne & Chirico, 2013; Le Breton-
Miller, & Miller, 2013; Miller & Le Breton-Miller, 2011). On the other
hand, there is a logical argument that in some firms the importance of
SEW might increase over time (Berghoff, 2006). For instance, repeated
dynastic succession, being one dimension of SEW (Berrone et al., 2010),
might also create a tradition and expectation that the firm is passed on
from one generation of owner-managers to another as a family heirloom
(Karlsson Stider, 2000), meaning that SEW is prioritized over financial
wealth considerations. Taking a resource-management perspective,
Habbershon and Pistrui (2002) claim that some family firms concentrate
too much on the preservation of family legacy assets for emotional reasons.
Over time, family history and corporate history become more and more
entwined (Blombäck & Brunninge, 2013), implying that the reconfiguration
of resources becomes an emotionally challenging process even if it even-
tually succeeds (Salvato, Chirico, & Sharma, 2010). We conclude from this
discussion that there are contradictory arguments both for a decreasing
impact of SEW over time, as well as, for an accumulation of SEW endow-
ments over the life cycle of a family business.
Scholars being critical of the SEW perspective have remarked that the
rationality underlying SEW has already been addressed by previous litera-
ture, albeit using other labels (Miller & Le Breton-Miller, 2014). This is a
true for the extant literature on resource management. Sirmon and Hitt’s
(2003) resource-management model for instance acknowledges that nostal-
gia and emotions may have a disturbing effect on a resource-management
process that is overall guided by a financial wealth rationality. Miller, Le
Breton-Miller, and Scholnick (2008) outline the stagnation perspective on
family firms as focusing on the conservatism of family businesses. Family
firms are supposed to develop cultures that make them inflexible, resistant
to change, and adhering to family traditions (Dyer, 1986; Hall, Melin, &
Nordqvist, 2001). They risk sustainable economic wealth creation if they
get stuck to historically established business models in order to protect past
generations’ legacies (Habbershon & Pistrui, 2002). Salvato et al. (2010) see
past performance, commitment to the founder’s business, and a feeling of
personal responsibility to the founder’s business as key inhibitors of change

in family-controlled organizations. Although many scholars emphasize

inertia and traditionalism in family firms, the literature also points at open-
ings for change. For instance, Dyer (1986) distinguishes between past-
oriented and present-oriented paternalistic firms. Both types share the basic
characteristics of paternalistic businesses, namely the concentration of
power to an autocratic family leader. However, while the former empha-
sizes family legacy and sticks to old patterns, the latter type may preserve
some family tradition, by focusing on handling current needs and adapting
the firm to new challenges. Interest in the past and in family tradition does
not necessarily lead to stagnation. Salvato et al. (2010) discuss the role of
strong family business leaders who act as “family champions of continuity”
(FCC). These individuals bring about change in their firm’s resource struc-
ture while maintaining continuity by preserving timeless values, which are
rooted in the history of the family and its business. Applying a SEW per-
spective to these findings, socioemotional endowments were mainly mani-
fested in values that were sufficiently abstract to apply them to changing
contexts. Institutional theory suggests that organizations and their key
operating principles are infused with value during the early years of their
development (Selznick, 1957). Founders imprint their personal values on
the organization, a process which is apparent in family firms (Schein,
1983). If a family business is able to safeguard SEW endowments in the
form of guiding values, it may at the same time pursue changes that aim at
increasing financial wealth. Rather than facing a choice between either
emphasizing traditions or breaking new ground, firms can combine past
and present/future orientation by using the past to facilitate change, either
by providing a sense of continuity in times of turbulence (Brunninge, 2005;
Salvato et al., 2010), or by serving as a source of inspiration for using their
resources in novel ways (Brunninge, 2009; Brunninge & Melin, 2010). The
FCC concept, like Dyer’s (1986) archetypes of paternalistic leaders, empha-
sizes the role of strong individuals in either preventing or facilitating
change. Sharma and Manikutty (2005) propose that individualistic cultures
in family firms, relying on one dominant leader at a time, create a good
basis for changing resource configurations, as there is less need for
consensus-building in the family. The diverging opinions on the inertia of
family firms as well as the influence of paternalistic leaders underline the
importance of a fine-grained analysis that shows how, why, and under
which contextual influences family businesses change or stagnate (Ojala,
Lamberg, & Melander, 2010).
Overall, we conclude from our literature review that resource manage-
ment in family firms cannot be understood by applying a perspective that
Family Firm Longevity and Resource Management 185

exclusively relies on a financial rationality. Family businesses create both

financial wealth and SEW for their owners and it is far from certain that
family members in control will prioritize financial outcomes over socioemo-
tional endowments when faced with a choice between the two. At the same
time, particularly large owner families might face situations where different
individuals or owner coalitions have different priorities when it comes
to dealing with apparent conflicts between financial and socioemotional
outcomes. In our case study of MoDo, we will study the interplay over
time between considerations guided by a financial rationality and those
oriented toward SEW. What rationalities have dominated different
resource-management choices? To what extent has it been possible to com-
bine the financial and SEW rationalities? And finally, how has this related
to the interplay between internal family dynamics and developments in the
industry context during different periods of MoDo’s history?

The Resource-Based View is one of the dominating topics in management
research (Newbert, 2007), but even if the number of empirical studies is
impressive, some methodological issues still remain to be sufficiently
addressed. Two of the most prominent are the value of resources and the
understanding of the process of resource management (Rau, 2013).
Without a thorough understanding of the nature of the resource environ-
ment, the value of resources are impossible to assess (Armstrong &
Shimizu, 2007; Gibbert, 2006; Sharma & Carney, 2012), echoing the urge
for contextualized process studies (Pettigrew, 1990). Moreover, the value of
resources is low if they are not “in motion,” so without an understanding
of the processes of resource management the link between resources and
performance outcomes is weak (Rau, 2013).
These aspects have been decisive for our choice of MoDo in its indus-
trial context as our explorative case. Over a period of more than hundred
years, MoDo was a well-known company in Sweden with high-profile lea-
ders operating in an important and distinctive industrial context. A context
that today must be classified as hostile to family governance (Casson, 1999;
Glete, 1989). Moreover MoDo is an example of a family firm that we can
follow both in prosperous and less prosperous times, which is rare in
research (Armstrong & Shimizu, 2007; Zellweger et al., 2011). Further,
extensive data on the company and the industry is available.

Operational Method

In previous research, vital data on the structure of the Swedish pulp and
paper industry has been developed. Sölvell, Zander, and Porter (1993,
p. 76ff.) described the competitive advantage of the wider industry context
applying the diamond framework (Porter, 1990). On a more detailed level,
an understanding of the resource environment was developed by Melander
(1997),1 Lamberg, Näsi, Ojala, and Sajasalo (2006),2 and Järvinen, Ojala,
Melander, and Lamberg (2013). Furthermore, extensive research relates the
Swedish pulp and paper industry to international development (Lamberg,
Ojala, Peltoniemi, & Särkkä, 2012; Peterson, 1996).
The understanding of resource management in a historic case that cov-
ers a period of more than a hundred years is challenging. However,
MoDo and its leaders received a lot of public attention and extensive
data on the family and the management of the firm is available. We also
conducted interviews with three different members of the owner family.
In the appendix we have included a presentation of the data that forms
the basis of this study. As a first step of analysis, we used the data to
construct a chronological narrative of Modo’s historical development. In
the second phase, we further analyzed our data, building on Dutton,
Fahey, and Narayanan (1983) and Ocasio’s (1997, 2011) research on iden-
tifying and processing strategic issues from an attention-based perspective.
By focusing on attention given to specific issues we can understand
how resources have been valued and allocated both historically and in a
prospective view (Langley, 1999; Ocasio, 2011; Schreyögg, Sydow, &
Holtmann, 2011).
In the analysis of MoDo’s historical development, we focused on issues
that have been in focus repeatedly over time. Following Pettigrew’s state-
ment that “truth is the daughter of time” (1990, p. 271), we identified four
strategic issues that had decisive importance for the future development at
the time when they emerged. In retrospect, for example, in different histori-
cal exposés on MoDo, these issues were also viewed as decisive for the stra-
tegic direction followed for a long period of time
These issues were:

• The investment in the Husum pulp mill in 1917.

• The buildup of systematic R&D activities in organic chemicals in the
• The start of the paper machine in Husum from 1969 to 1972.
• The “third block” strategy in 1978.
Family Firm Longevity and Resource Management 187

The presentation of the strategic issues in this chapter begins with a brief
overview of MoDo’s history and the industrial context in which MoDo
operated. Thereafter we present and analyze the strategic issues, guided by
the issue framework introduced and operationalized by Melander (1997,
2005) and Melander, Melin, and Nordqvist (2011):
• Driving forces that made the issue enter the top management agenda?
• Actors involved in the management of the issue?
• Resource mobilized in the issue processing?
• The family perspective on the issue?
The family business literature focusing on the longevity of family firms
emphasizes the importance of stable guiding values both for constraining
and enabling changes in a firm’s resource configuration (Salvato et al.,
2010; Sharma & Sharma, 2011). It is thereby argued that strategic issues
offer a rich empirical setting for the study of organizational values (Bansal,
2003; Dutton & Penner, 1993). To qualify our analysis, in the third phase
we made a cross-sectional analysis in order to identify values that infused
the identified issues. The values of interest were those that repeatedly are
referred to when motivating, arguing, and justifying the identified strategic
issues. The identified values are analyzed from a resource-management per-
spective including the respective role of the financial and social-emotional
wealth dimensions.

MoDo: 1873 1991

When the founding father JC Kempe, characterized as a man of dynamism

and vigor and as an honest businessman (Ahnlund, 1917), died in 1872 he
was mourned by 11 children from three marriages. At this time MoDo was
just one of the branches of the businesses run by the business family
Kempe. As a way to manage a fair succession in 1873, all heirs received a
share of MoDo in which all JC’s businesses were incorporated in. In 1884
JC Kempe’s third son Frans was appointed managing director and took
full control of the company. At the end of the 19th century, operations
became more streamlined and the decision was made to focus on sawmill-
ing and wood pulp.
Frans Kempe was the architect behind the focused company and the
rapid expansion. He and his brother Seth closely cooperated in organizing
the first systematic research in forestry and pulp technology in 1899 and
Frans’ passion for systematic research continued (Boseaus, 1949).

From later research it is clear that Frans gradually ran MoDo on his
own and that there were several disputes between him and other close rela-
tives (Interviews; Kempe, 2006). It was at this time that the division
between two branches of the Kempe family tree emerged, named after the
two brothers, Seth and Frans who had originally collaborated so well
(Kempe, 2006). Frans remained in office until 1916 when he resigned in
anger after being criticized by the board (consisting of close relatives) for
the decision to invest heavily in the Husum pulp mill, despite the ongoing
world war. The Husum pulp mill was later to become of decisive impor-
tance for the company’s development and this first investment was followed
by many more. We therefore choose the decision to invest in the Husum pulp
mill as the first strategic issue to be further analyzed.

MoDo 1917 1949

Even though Frans was succeeded as Managing Director by his son Carl
Kempe in 1917, Frans dominated the company as chairman of the board
until 1922. That year his half-brother Seth was appointed chairman of the
board, due to Frans’ age. Seth stayed on until 1930 (Kempe, 2006). Under
Carl’s reign, the strategy was to increase the capacity in the pulp factories
close to the two main sawmills in Domsjö and Husum, combined with an
organic chemical industry, based on by-products from the production of
wood pulp. The more systematic R&D activities that began in the 1930s
showed to be an excellent platform for the expansion (Boseaus, 1949;
Valeur, 2000). R&D became the platform from which numerous products
and production innovations were developed (Valeur, 2000) and hence we
chose the buildup of systematic R&D activities in organic chemicals as the
second strategic issue.
The appointment of the bon vivant Carl was controversial in the family
and the struggle between the two branches of the family continued (Unger,
1967). In 1936, one of JC Kempe’s daughters, Lotty Kempe, married
Bruzelius, and established a foundation in her will. Lotty, a Seth ally, was
eager to tighten the ownership of MoDo and to balance family influences
(Kempe, 2006). The foundation soon became an important owner as the
dividend was reinvested in MoDo. The same year, MoDo was listed on the
stock exchange. Major reasons were to enhance trade in the share and to
become more publicly known (Gårdlund, 1951).
Family Firm Longevity and Resource Management 189

MoDo 1949 1971

In 1949, Carl Kempe was appointed chairman of the board and Erik
Kempe, a grandson to JC Kempe from his third marriage and representing
the Seth branch, succeeded Carl as MD. The reason for this change
between the two branches was that the “Frans branch” was unable to pre-
sent a suitable successor to Carl. Erik and Carl’s comanagement of the
company was not without frictions (Lyberg, 1984). In 1959, Erik Kempe
suddenly died from a heart attack and due to the lack of successors in the
family, Bengt Lyberg was appointed MD.3 Lyberg’s time in office until
1971 was characterized by a balancing of family interests. He was also sur-
rounded by family representatives in the management team. Carl remained
chairman of the board until 1965 when he was succeeded by his nephew
Matts Carlgren. Matts took over the post as MD in 1971 and stepped
down in 1985, remaining operative chairman of the board until 1990.
The Second World War opened up new opportunities. Pulp manufac-
turers in northern Sweden used their R&D departments to develop a range
of spin-off products.4 The use of pulp for manufacturing paper was almost
dimmed by all new uses arising as a consequence of the war experiments
(cf. Hernód, 1942). As a result, the route for the future was under review
(Melander, 1997).
In Sweden, the two dominating strategic alternatives5 were to integrate
pulp with a production of paper or to develop the pulp industry into an
organic chemical industry, that is, commercializing all by-products devel-
oped during the war. Of predictive importance was the decision of the lar-
gest pulp producer SCA in the mid-1950s to integrate with a production of
newsprint and cardboard paper (Gaunitz, 1979).
MoDo, an important pulp producer in 1945, made no clear choice
between the strategic alternatives, but there was an emphasis on continued
investments in organic and later petrochemicals (Carlberg & Scholander,
1989). This consistent focus on pulp technology was anchored in the exten-
sive R&D efforts. Peterson (1996) concluded that MoDo developed at least
six industry-wide technology innovations after World War II.
In 1964, Bengt Lyberg announced that the company also decided to fol-
low the first strategic alternative, and integrate the production of pulp with
a production of paper. The decision was late when compared to the major
competitors in the industry and implementation lasted. Production in
Husum did not start until 1972. This decision was the starting point for
several new investments. It was a daring investment both when it came to

the size of the machine and the choice of paper grade (Lyberg, 1984). The
initiative to build a writing paper machine in Husum therefore surfaces as the
third strategic issue to be analyzed.
The alternative chosen by MoDo was writing paper, a grade that tradi-
tionally had low volumes and was mainly produced in the south of
Sweden. An important reason for this adventurous investment was the
tempting market opportunities. In this niche, MoDo could become a domi-
nant player. The commission of the large writing paper machine in Husum
coincided with an extreme growth in demand in 1972 1974, making the
investment extremely profitable.

MoDo 1973 1990

Matts Carlgren, Carl Kempe’s nephew, took over the post as MD in 1971.
Although he stepped down in 1985, he continued exercising dominant influ-
ence as an operative chairman of the board until 1990. He was succeeded
by two nonfamily MDs during the period 1985 1990. As MoDo celebrated
its centennial in 1973, the technological developments forced the company
to decide between heavy investments in the chemical or in the pulp and
paper business. The decision was to eventually divest most of the chemical
operations and use the financial resources released to invest in a second
huge writing paper machine. The investment was adventurous as the finan-
cial bases were weak at the time.
In 1975, MoDo was still controlled by the Kempe family, but the voting
majority was weak. Matts Carlgren himself owned roughly 22% of the
votes, the Kempe foundations 20%, and other Kempe heirs 7%, that is, in
total about 50% of the company. The “Frans branch” represented by
Carlgren ran the company and the “Seth branch” controlled the Kempe
foundation. This division made family control fragile.
In the early 1970s an “elite division” started to emerge in the pulp and
paper industry. SCA had been the industry leader for long, but now Stora
Kopparberg (later STORA) emerged as a second contender. Both these
dominating competitors to MoDo had strong links to banking spheres.
SCA was closely linked to Svenska Handelsbanken and STORA belonged
to the Wallenberg sphere, including S. E. Banken.
The term “third block,” coined by Matts Carlgren in the late 1970s,
reflected the idea of a diversified holding company covering a large part of
the pulp and paper product range and hence contesting the two leaders in
the industry (Ericson, 1991). As we will see later, the implications of
Family Firm Longevity and Resource Management 191

adapting this growth-related strategy were remarkable in the decade to

come. The adoption of the “third block” strategy is therefore identified as the
fourth strategic issue to be analyzed.
The strong belief in family control of the company offered tough restric-
tions on the applicable growth methods (Croon, 2005). MoDo’s indebted-
ness increased radically in the 1970s and in 1982 it reached 71.4%, up from
54% in 1980, high above the industry average (MoDo, Annual reports
1978 1984). MoDo pursued an aggressive financing strategy that was unu-
sual at this time, including currency carry trade, offensive marketing to the
financial market, and intensive reduction of capital tied in fixed assets.
An opportunity to implement the third block strategy followed in 1983
when MoDo, after a harsh struggle with STORA, took control over
Iggesund a very profitable cardboard mill. In 1985/1986, a new opportunity
to continue the strategy emerged when Holmen, a medium-sized producer
of newsprint, tissue, and paperboard was for sale. When the acquisition of
Holmen was eventually finalized in 1988, the group was organized as a
holding company with several business areas. At this time, MoDo had
almost caught up with the two rivals in terms of sales. However, soon a sec-
ond internationally oriented acquisition wave began in the industry.
In 1985, MD Matts Carlgren commented on the future prospects of the
group. He concluded that family ownership was long term and guaranteed
the outlined strategy “in the foreseeable future.” Besides commenting on
the ongoing acquisitions he also underlined features that discerned MoDo.
One was R&D in which MoDo had been able to show a number of “firsts”
over the years and another one was the management style of the company,
“Top management will also in the future bee concentrated to a few decision-
makers that employs short and fast decision-making processes” (Matts
Carlgren in Gårdlund,1986, pp. 173 174, authors’ translation).
The acquisitions made in the 1980s, strained MoDo’s capital base. The
urge to keep the majority ownership within the Kempe family made it
impossible to make “all-share deals” as a way of financing future acquisi-
tions. Being one of the major owners, Matts Carlgren was also personally
exposed. Following the recession in the early 1990s, Matts became unable
to pay interest on the loans needed to defend his ownership (Frans Kempe
in Croon, 2005, Appendix, p. 219). As a consequence, Carlgren was forced
to sell his shares in MoDo to SCA, one of MoDo’s major competitors and
he resigned as chairman of the MoDo board in 1990. The Kempe founda-
tion refused to follow Matts’ decision to sell to SCA and the control of the
company became unclear. Commenting on the end of the family control of
MoDo, Frans Kempe jr., Carl Kempe’s grandson, concluded that “The

MoDo Group has since long been managed by a strong owner-family with an
industrial long term thinking. An owner family that had no financial resources
outside the ownership of MoDo shares. When MoDo entered the financial
‘elite division’ neither the family nor the management was prepared or had the
influential financial network needed” (Frans Kempe in Croon, 2005,
Appendix, p. 219).

MoDo in the Industrial Context

Based on Layton (1979) we can summarize the dominating forest-related

technologies in northern Sweden at the 19th and 20th centuries.
As Fig. 1 shows, the first wave of importance to MoDo was the steam-
based sawmilling. In the period 1869 1893, the value of sawmilling pro-
duction increased more than 50% in Sweden, mainly in the northern parts
of the country. As a result substantial fortunes were created, often con-
trolled by single entrepreneurs (Glete, 1987; Myhrman, 1984; Schön, 2008).
As the access of quality timber decreased and the techniques for pulp

Peak Technology

Charcoal Water-based Steam-based Pulp Integrated

ironworks Saw mills Saw mills factories Paper and
pulp mills

Ca 1800 Ca 1830 1900/1910 1950/1960


Fig. 1. Technology Waves in Northern Sweden 1700 1960. Source: Created by

authors based on Layton (1979).
Family Firm Longevity and Resource Management 193

production based on wooden fiber developed, there was a shift in the indus-
try. Sawmilling companies extended their operations to a production of
wooden pulp. In the period 1875 1920, the number of employees in the
pulp industry increased from less than a thousand to approximately 17,500
(Svenska Industri 1925, 1926). At this time, the paper industry was operat-
ing in the south of Sweden and did not follow the waves outlined in Fig. 1.
The third wave of changes in the north started in the 1950s. Between 1960
and 1990, the pulp industry doubled its annual capacity to 11.2 million
tons. During the same period, the number of pulp mills was reduced from
127 to 54 and almost the entire capacity increase was integrated with a pro-
duction of paper. Paper production capacity increased with more than
300% to 9.3 million tons at the same time as the number of mills decreased
from 76 to 54 (Svensk Industri, 1992). These production figures and espe-
cially the high export share made Sweden one of the five largest pulp and
paper countries in the world.
The decrease in both the number of companies and production units
indicates the emergence of substantially larger companies reaping the bene-
fits of economies of scale. Linked to this development was the fact that the
remaining companies more often went public and few remained with pro-
minent individuals or families as distinctive owners (Glete, 1989). A major
shake-out of individuals and family owners in the sawmill, pulp, and paper
industry took place in 1932 when the Swedish industry suffered from the
“Kreuger crash”;6 however, the decrease has gradually been evolving over
the entire period of 1873 1990 (Fagerfjäll, 2008).
MoDo followed these overall trends and developed from a sawmill com-
pany to a sawmill and pulp company, and finally to an integrated producer
of sawn timber, pulp, and paper. Between 1875 and 1890 sales of sawn tim-
ber increased more than five times (in the same period that the Swedish
production doubled). At the end of the 1880s, MoDo was number eight on
a list of the 50 largest sawmill companies in Sweden. The expansion contin-
ued in the following decades and a systematic comparison by Gårdlund
(1951, p. 54ff.) shows that MoDo with an exception of two years was con-
siderably more profitable than the average of the five most important com-
petitors in the period 1880 1913.
Production of pulp began in 1902 and at first production growth fol-
lowed the overall growth in Sweden, but in the period 1929 1938 MoDo’s
market share increased from 5.6% to 6.9%, in a time in which national
production of chemical pulp increased by 26% (Gårdlund, 1951; Streyffert,
1950). This relative increase came at a time in which Sweden’s share of
world export of pulp was about 40%.

In 1950, MoDo was one of the largest pulp and paper companies in
Sweden. However, SCA was by far the largest, with a merger of 16 inde-
pendent companies in 1929. SCAs turnover was 500 Million Swedish
Kronor almost three times higher than MoDo’s. At this time SCA domi-
nated in sawn timber, but MoDo was ahead in organic chemical products.
In the beginning of the 1970s, MoDo’s position in the Swedish pulp and
paper industry was still prominent. Table 1 illustrates MoDo’s dependence
on pulp for sale and low engagement in paper production.
In 1990, the industry in total consisted of 13 mill companies and 12 com-
panies running more than one mill. Out of the top ten in 1972, three com-
panies, Billerud, Holmen, and Fikeby, were acquired by the dominating
three and one was taken over by the government (Ncb). As a consequence,
the three dominating companies (SCA, STORA, and MoDo) produced
more than 75% of the total output in the industry in 1990.
If we compare MoDo with the market leader SCA in terms of growth of
sales, MoDo and SCA kept the relative difference. However, as illustrated
in Table 2, the differences in return of investments (RoI) became pro-
nounced after 1970. SCA had a higher average and earned money in the
entire 1960 1990 period, but MoDo suffered a loss in operating profit in
1978 1979 and 1981 1982.

Table 1. Capacity in the Swedish Pulp and Paper Industry in 1972.

Company Capacity % of Total Capacity % of Total % of
Pulp for Sale Pulp for Sale Paper Total Capacity Total
(’000 Tons) Capacity (’000 Paper (’000 Capacity
Tons) Capacity Tons)

SCA 490 (3) 8.8 760 (2) 12.7 1,250 (1) 10.8
Södra 840 15.1 120 2.0 960 8.3
Holmen 10 0.2 880 14.7 890 7.7
MoDo 710 (2) 12.7 140 (9) 2.3 850 (4) 7.3
Stora 355 6.4 475 7.9 830 7.2
Ncb 415 7.4 195 3.3 610 5.3
Billerud 230 4.1 365 6.1 595 5.1
ASSI 100 1.8 475 7.9 575 5.0
Korsnäs-Marma 290 5.2 225 3.8 515 4.5
Fiskeby 40 0.7 440 7.3 480 4.1
Ten largest 3,480 62.4 4,075 68.0 7,555 65.3
Total 5,575 100 5,993 100 11,568 100

Source: Annual Reports 1972, from the companies named in the table.
Family Firm Longevity and Resource Management
Table 2. MoDo and SCA in Terms of Sales and Profitability.
MoDo Sales MoDo Operating MoDo Profit MoDo SCA Sales SCA Operating SCA Profit SCA
in MSEK Profit in MSEK Margin (%) RoI (%) in MSEK Profit in MSEK Margin (%) RoI

1960 (RoI 435 55 12.5 12.6 623 80 12.8 10.8

1960 1962)
1970 (RoI 1,059 91 8.6 9.2 1,468 132 9.0 8.9
1969 1971)
1980 (RoI 3,808 281 7.4 3.9 6,708 736 11.0 10.5
1979 1981)
1990 (RoI 18,435 1,045 5.7 0.6 31,122 1,711 5.5 9.4
1989 1991)

Source: Modo and SCA Annual Reports: 1960 1962, 1969 1971, 1979 1981.
Note: RoI is counted as an average over three years to equal out major changes.


Issue Analyses

The Investment in the Husum Pulp Mill in 1917

On December 21, 1916 Frans Kempe, the MD of MoDo since 1884
resigned. The reason for the resignation is often said to be a dispute over
Frans’ single-minded decision to build the Husum pulp mill, the corner-
stone in the coming expansion of the company (cf. Croon, 2005). The deci-
sion to build was controversial as it was made in the middle of a world war
and in a volatile market situation. The reason why the issue became impor-
tant was not only the investment in itself. Earlier Frans had made similar
decisions without asking or discussing with relatives. His argument was
that in the long term, the decision was rational as it was in line with the
gradual shift from sawmilling to a production of pulp in the north of
Sweden. The major difference was that he now applied a long-term finan-
cial rationality, disregarding family members’ desire for short-term divi-
dends. The background was an escalating controversy between Frans
Kempe and his relatives. Frans and his half-brother Seth had since long
worked together and complemented each other. Frans focused on MoDo
and Seth (and other relatives) was managing separate forest industry busi-
nesses. However as MoDo grew, demands on dividends from family mem-
bers increased at the same time as Frans gradually viewed MoDo as “his
own possession,” to hand over to his son Carl. In fact there had been no
formal board meeting for a period of over 30 years in MoDo and Frans
did admit that his leadership was despotic (speech by Frans Kempe repro-
duced in Unger, 1967). This, together with the financial setbacks in the first
world war and Frans aging, made relatives worried about the management
of the company, and there were repeated requests for him to resign
(Kempe, 2006). Frans’s power position was fragile as he did not control the
majority of the company. However, in terms of his position as MD and
having a strong personality he managed both to appoint his son Carl as
successor and himself as chairman of the board.
The major actors in the issue were Frans and his son, the successor Carl,
and on the other side the younger half-brother Seth and his sister Lotty
Bruzelius, among other relatives. Frans was an important shareholder but
Seth and his relatives owned majority of the company. The outcome of the
issue, besides the investment in the Husum pulp mill, was the formation of
mistrust between two dominating family branches (named after “Seth” and
“Frans” Kempe) (Kempe, 2006). This mistrust later influenced other issues
in the company, such as the creation of the Kempe Foundation in 1936,
which was made to balance owner interests in the company, and the
Family Firm Longevity and Resource Management 197

succession after Carl Kempe and Erik Kempe in 1949 and 1959 (Lyberg,
1984, p. 183). Deciding about the Husum issue, Carl Kempe was guided by
a financial wealth rationality. The investment eventually secured long-term
streams of financial wealth at the sacrifice of SEW in the family. Family
relations suffered from the consequences of the Husum conflict for

The Buildup of Systematic R&D Activities in Organic Chemicals

As the pulp industry expanded, the interest for systematic research and
development increased and in the beginning of the 1930s all dominating
pulp companies started R&D departments. At MoDo, Frans Kempe‘s
entrepreneurial spirit in forestry was in line with innovation activities in
pulp technology. MoDo had an early focus on the organic chemical indus-
try based on by-products from the large-scale pulp production. MoDo was
on the forefront in this area and was one of the companies that invested
most both in production and development (Gårdlund, 1986). Boseaus
(1949) reports that the laboratory inaugurated in 1941 was well in parity
with the leading ones in the United States.
Operations were profitable during the war and forecasts about a fast-
growing market made MoDo continue to invest in organic chemistry
(Lyberg, 1984). In order to meet the international competition from oil-
based chemicals, efforts were now more focused to niche products that
often had to be oil-based. This in turn demanded even more intensified
research. Boseaus (1949) reports that MoDo invested one million SEK in
R&D which was about 2% of total sales. Comparative figures are difficult
to obtain, but the report “Svensk Skogsindustri i Omvandling” (1972,
p. 239) reports that the industry average was only 0.5% in 1963.
Carl Kempe, MD, 1917 1949 was the champion of the investments in
organic chemical technology (Gårdlund, 1986). He followed the tradition
of systematic innovation that his father Frans introduced in foresting
(Boseaus, 1949). Carl’s efforts were rewarded by wartime conditions. His
undisputed position as a successful MD made the strategic choice of con-
tinuous efforts in this business area unquestioned even after the war. It
took until 1973, six years after his death, that the chemical operations were
divested. Several observers make the interpretation that the company’s
managers had become emotionally attached to the business and clung to
the resources in the chemical industry, that is, they protected SEW endow-
ments in favor of releasing resources for other investments that would have
followed a financial wealth rationality (Croon, 2005; Interview with Karl
Kempe, 2012). A rationality relating to the resource of family ownership

had become dominant for MoDo’s research management and it was only
after much hesitation that Matts Carlgren eventually decided to divest the
chemicals business in favor of businesses that were more promising finan-
cially (Croon, 2005).

The Start of the Paper Machine in Husum in 1972

In 1964, Bengt Lyberg (MD 1959 1971) announced that the company
intended to integrate the production of pulp with the production of paper.
This happened while the company invested heavily in petrochemicals and
the hygiene sector. The decision was late compared to the major competi-
tors. Forsgren and Kinch (1970) compared the development of 17 compa-
nies in the industry between 1955 1957 and 1966 1968. MoDo, one of the
largest pulp producers, was down from position 12 (out of 17) to position
14 (out of 15).
It took until 1969 before the final choice of building a writing paper
machine in Husum was made. The choice was unexpected as the writing
paper technology was undeveloped and the market was highly fragmented.
The machine was supposed to have a capacity of 70,000 tons to be com-
pared with the average new writing paper machine in Sweden with a capa-
city of 30,000 tons.
The background to the late decision was that major financial resources
had been used in the chemical industry, together with a continuous need
for capital in the pulp industry and a costly buildup of a hygiene division.
The focus was on developing new, capital-intensive businesses, not on
divesting old businesses (Croon, 2005). In the struggle between the financial
wealth logic of renewing MoDo’s business portfolio for future competitive-
ness and a SEW rationality of preserving the portfolio established by pre-
vious generations, renewal eventually prevailed, albeit with a serious delay.

The Third Block Strategy in 1978

The successful investment in a large writing paper machine in Husum was
followed by a second one. Now market conditions had changed and MoDo
was suffering from the recession in that last years of the 1970s which was at
the same time that the major rivals, SCA and STORA, had started a
restructuring of ownership in the industry. At this time MoDo developed a
long-term strategy. Three scenarios were outlined: the first focused on
acquisitions of nearby paper factories/companies, offering expansion
opportunities and synergies in wood supply; the second focused on organic
growth in existing units; and the third focused on a merger with the largest
Family Firm Longevity and Resource Management 199

rival, SCA (Croon, 2005). All these scenarios mirrored MoDo’s traditional
belief in economies of scale
The term “third block,” refers to the first of these scenarios and this was
the strategy that was implemented. The first opportunity came when the
dominant owner of the very profitable Iggesund cardboard mill died in
1979. In the following struggle with STORA, financing became a bottle
neck and it was not until 1983 that MoDo consolidated the acquisition.
The second step was Holmen, a newsprint/hygiene company. Acquisition
of shares started in 1985 and in the following years yet another ownership
struggle took place between the Kempe family and some of Holmen’s main
owners. When the acquisition of Holmen was eventually finalized in 1988,
an integrated company, but with several rather independent business areas
was created.
Carlgren’s management in this phase followed the entrepreneurial and
sometimes even despotic style of Frans and Carl. “The pattern was that
Matts suddenly announced that he had acquired a new stock of shares and
ordered the CFO to get hold of the payment necessary” (Croon, 2005,
p. 107). Croon (2005) reported that the pace of the acquisitions was also
questioned by the Seth branch in the family. Decisions were not anchored
and the principle of absolute family control (50% of voting capacity) was
not negotiable.
As a result, Matts Carlgren personally had to take substantial loans in
order to finance his share of the Holmen acquisition and when his private
operations made a loss at the same time as stock prices decreased, the situa-
tion became impossible (Kempe, 2005). The result was a loss of the family
control of the company in 1990.
While the desire for growth could be justified from a financially oriented
logic, the risky way of funding it was entirely driven by a SEW rationality.
Carlgren was willing to expose MoDo to high financial risk in order to
preserve the ownership dominance by the family (cf. Gomez-Mejia et al.,
2007). Also on a personal level, Carlgren’s financial risk-taking was

Five Guiding Values

Following the issue analyses we now proceed with the second-order
analyses in which we identify five guiding values influencing management
choices that surface in the issue analysis. The importance and value of
these guiding values is then analyzed from a resource-management per-
spective, including the respective role of the financial and the SEW

Autocratic Leadership
Surfacing especially in issue one and four is a strong belief in patriarchic
ownership. The company is the family and the autocratic leader of the com-
pany is the obvious head of his family. The owner family is to subsume to
this hierarchical order. This principle established by Frans (and probably
JC) is carried on by Carl Kempe (Lyberg, 1984) and Matts Carlgren
(Kempe, 2006). Autocratic leadership was combined with a patriarchic car-
ing for employees. In a research thesis on labor relations, Frans Kempe
was characterized as: “If there is any manager that deserves the label ‘father
of his workers’ it is Frans Kempe …. he made MoDo to a predecessor in social
issues” (Nordström, 1993, p. 48).

Family Ownership of the Firm

The cornerstone of the autocratic leadership was absolute ownership con-
trol through the Kempe family. This was complicated as we seen in issue
one. In fact it was two opposing family branches that owned the company.
As long as there was one autocratic leader successfully running the com-
pany, the situation was manageable. In times of financial deficits or leader-
ship vacuums (e.g., the Husum crisis in 1917; when Erik Kempe died in
1959 and in the turbulent 1980s) the implications of the strict interpreta-
tion of ownership control however surfaced. An example of this appears
when the “third block strategy” (issue two) was formulated. Matts
Carlgren declared that a decisive criterion for the strategy chosen was the
possibility to ensure absolute ownership control by the Kempe family
(Croon, 2005, p. 103).

Emphasis on R&D
Frans Kempe initiated scientific research in foresting already in the 19th
century and initiated several new innovations. He was followed by Carl
Kempe who continued his scientific efforts when constructing the organic
chemical development path in the 1940s (issue two). This emphasis contin-
ued during the 1950s and 1960s. The head of R&D, Ingemar Croon com-
mented on the situation in those years with the following words: “The
years between 1965 and 1970 was a time of joy when it came to research and
development” (Croon, 2005, p. 20).

An Entrepreneurial Management Perspective

The Napoleon wars had ruined the Kempe family. JC Kempe and his
brother Wilhelm were sent to Sweden and created two impressive fortunes.
JC’s fortune eventually became the foundation of MoDo, but Wilhelm’s
Family Firm Longevity and Resource Management 201

was gradually eroded and this branch of the family eventually disappeared
in the forest industry. Here we detect the entrepreneurial spirit that contin-
ued with the remarkable growth in sawmilling and the pulp industry during
Frans Kempe’s reign, the dominating role in the pulp and organic chemical
industry during Carl Kempe’s reign, and finally the bold investments in the
writing paper industry and the construction of the third block during
Matts Carlgren’s reign. The “third block strategy” was commented by
Matts Carlgren with the following words.: “You have to be bold. Many have
failed in this industry due to cautiousness. To be cautious could sometimes be
mixed up with cowardness” (Interview with Matts Carlgren after his resigna-
tion from the MoDo board, Dagens Nyheter, 1991).

Emphasis on Economies of Scale

Sawmilling and pulp & paper are all process-oriented industries in which
economies of scale are important. However, there are always limits to the
scale possible and a balance between the scale chosen and the market
demand. As illustrated by the final words in Carl Kempe’s obituary from
1967, MoDo’s choice has always been to go for size. “Why limit an expan-
sion to 50% of the present capacity when it could be a 100%?” (Svensk
Papperstidning, 1967, no. 15 p. 485). This view was decisive both for
Frans’s decision to invest heavily in Husum in 1917, as it was when Matts
decided to invest in a huge writing paper machine in the same site in 1969.


In our theoretical frame of reference we identified two rationalities, one dri-

ven by financial wealth considerations and one driven by SEW, as poten-
tially crucial for resource management. We asked how these rationalities
influenced resource management at MoDo, to what extent they could be
combined, and how these rationalities were related to the interplay between
family dynamics and contextual developments during MoDo’s history.
In our attention-driven analyses of MoDo’s history we have identified
issues and guiding values in which a tension between the financial and
SEW rationalities have surfaced, as shown in Fig. 2.
As illustrated, the guiding values reoccurred again and again. Hence,
they resemble the “ingrained values” that Salvato et al. (2010) identified as
providers of continuity in family firms. Initially, these values were random
and tied to the personal characteristics of individuals, for example, JC and

Four The investment The buildup of The start of the

strategic in the Husum systematic R&D paper machine The “third
pulp mill in activities in organic in Husum in block” strategy
issues: 1917 chemicals 1972 in 1978
Five guiding values:
• Autocratic leadership
• Family control of the firm
• Emphasis on R&D
• An entrepreneurial
management perspective
• An emphasis on economies of

Fig. 2. Guiding Values Related to Strategic Issues.

Frans Kempe being highly entrepreneurial leaders. However, over time the
values became institutionalized. They now applied to the organization as a
whole, but they also guided the selection of future leaders. For instance,
the fact that Carl Kempe was chosen as a successor to Frans, despite his
questionable track record as a bon vivant, was partly due to his business
instinct. Selznick (1957) characterizes such institutionalizing processes as
organizational characteristics becoming infused with value and cherished
for their own sake.
Before we continue and examine the role of the guiding values, it is
important to discuss the constraining and enabling power of the identified
values, an issue often neglected in analyses of family firms. An important
determinant of this power is to what degree the values surface in the ana-
lyses of the history of decision making in the focal firm. The message in our
case is that the guiding values have permeated decision making in MoDo
over generations. We can thus talk about a high level of institutionaliza-
tion. An answer to the question of why guiding values were able to gain
this level of importance is however not answered. Succession practices pur-
sued in MoDo surface as an important structural factor driving this institu-
tionalization process. Carl Kempe succeeded his father Frans in 1917, but
Frans stayed on as chairman on the board for five years, exercising a domi-
nant leadership style. Thereafter he was succeeded by his brother Seth as
chairman, continuing the succession of guiding values for another eight
years. In the same way, Carl Kempe continued as a dominant chairman on
the board in the period 1947 1965, ensuring that the company’s guiding
values were still alive in the mind of his successors. Hence, it is not surpris-
ing that our analysis of the 1980s still identifies a lot of references to Frans’
and Carl’s management styles, and thus to the powerful guiding values of
the firm.7
Family Firm Longevity and Resource Management 203

When we take a closer look at the values that emerged and subsequently
became institutionalized as guiding values, some of them were more linked
to a SEW rationality while others mirrored a logic representing financial
wealth. We can also identify a division in more content- and process-
oriented values. The emphasis on R&D and the focus on economies of scale
represent more content-oriented values; while, for example, the autocratic
leadership and the entrepreneurial management perspective were more pro-
cess oriented.
The particular combination of guiding values became a successful lens in
MoDo; a lens that guided strategic choices over a long period. Financial
wealth was achieved by bold, entrepreneurial decisions, building on the
R&D strength of the company and materializing in huge investments aim-
ing at economies of scale. Paternalism and family ownership on the other
hand mirrored SEW considerations that also guided what strategic issues
to be pursued and most important, how management decisions were made.
Learning from MoDo‘s rise and fall we will emphasize two further
dimensions of guiding values: the importance of the process nature of the
guiding values and the constraining nature of guiding values. In the MoDo
case, we learn that the financial superiority of the firm gradually eroded
after 1945, at a time when the industry trend was rewarding owner restruc-
turings to gain size and thereby a more apparent focus on financing. In this
new industry climate, there was eventually a need for a revision of the guid-
ing values in MoDo. This revision did not take place. Instead the last patri-
arch, Matts Carlgren enforced both the enabling nature of the guiding
values as he practiced entrepreneurial management when financing bold
investments, as well as the constraining side of the value of complete family
control, when he resisted some obvious strategic alternatives. What was
probably even more important was that he, the autocratic and supreme lea-
der, was unable or unwilling to question the guiding value of autocratic lea-
dership. As a consequence a catch 22 situation emerged. The enforcement
of the autocratic leadership made the two dominating branches of the own-
ing family unable to collaborate and revise other guiding values.
Looking back at the current debate and criticism concerning the SEW
construct, the MoDo case offers some interesting additional insights.
Miller and Le Breton-Miller (2014) have addressed the heterogeneity of
SEW, having various sources, types, and outcomes. The MoDo case high-
lights the fact that in family firms with spread ownership the question
of what and whose SEW influences management choices becomes crucial.
The Kempe Carlgren family came to be characterized by two branches
that partly had conflicting interests. Moreover, Matts Carlgren, the last

owner-manager, seemed to be driven by individual interests and commit-

ments to his personal projects to a significant extent. From his personal
perspective, there was eventually a SEW rationality behind his stubborn
pursuit of the third block strategy and his diverse business venturing
project, while other family member’s SEW rationality was mainly driven by
safeguarding family control of MoDo. This does not mean that the impor-
tance of SEW decreases with the firm age or generation in control
(De Tienne & Chirico, 2013; Sciascia et al., 2014), but rather the reference
point of SEW might shift from the family, to different family branches or
even the SEW of a single individual. Also the emphasis on a financial
wealth rationality versus a SEW rationality may differ between different
groups and individuals in the family.


The aim of this chapter has been to understand the longevity of family
firms by focusing on the management of resources from the perspective of
financial and SEW rationalities, hereby following Gomez-Mejia et al.
(2011), who urge for developing further understanding of “how family own-
ers frame decisions with socialemotional wealth as a reference point” (2011,
p. 694).
The result from our explorative analyses of MoDo’s rise and fall can be
summarized in the following points.
• Resource management in family firms must be holistically understood,
taking the balancing of financial and SEW rationalities into account.
SEW endowments are not merely a factor constraining financially
rational resource management, as previous models suggest (cf. Sirmon &
Hitt, 2003). They rather represent an equally important rationality for
resource-management choices.
• Family firms develop guiding values (cf. Salvato et al., 2010; Selznick,
1957) that institutionalize as the firm matures. These guiding values can
be both content and process oriented. It is in these values that financial
and SEW rationalities become manifested, as different guiding values
reflect the financial and SEW rationalities to a different extent. The bal-
ancing process between the two rationalities is thus mirrored in the influ-
ence of guiding values on resource-management choices.
• The normative power in guiding values, that is, the level of institutionali-
zation, varies in family firms. From the MoDo case we propose that
Family Firm Longevity and Resource Management 205

succession practices are highly determinative of the level of institutionali-

zation of guiding values. Specifically, succession processes that ensure
continuity through long and overlapping tenures by key decision makers
result in deeply ingrained values.
• Contextual changes determine to what extent guiding values can be effec-
tive in maximizing both financial and SEW outcomes. In the MoDo
case, the ownership consolidation of the industry restricted the possibi-
lity for maintaining absolute family control, while being financially
• Guiding values with a processual character are of specific importance in
family firms as those values determine the management of decision mak-
ing and influence in the long run. This eventually decides whether the
governance structure, that is, the family, will enable or constrain the
development of the firm.
• SEW is a heterogeneous construct and it is far from certain that a SEW
rationality always offers clear guidance to management decisions. In
multigenerational family firms with spread ownership, even competing
SEW rationalities within the family may complicate the picture.

Financial and SEW thus stand for two different rationalities that coexist
and sometimes compete in family firms. We see the adding of this perspec-
tive with dual rationalities to the resource-management literature as the
main contribution of our chapter. The perspective can hopefully help over-
come some of the shortcomings of existing resource-management models
that put too much of a focus on a financially driven rationality only. In
addition, our article demonstrates that the two rationalities influence
resource management through a set of guiding values and that the effective-
ness of these values changes with a changing industry context. The MoDo
case shows that financial and SEW rationalities can go hand in hand for
extended periods of time, safeguarding both financial success and socioe-
motional endowments. However, in a situation where guiding values are no
longer in line with the development of the industry context, the conflict
arising between the two rationalities may have fatal consequences for
the firm.
We believe that a SEW perspective can add more to the understanding
of resource management in family firms. Of course our research is limited
by the case and the specific empirical context chosen. We therefore suggest
that future research looks at dynamics arising between financial and SEW
rationalities in other contexts. Considering the important role of guiding
values, it would also be interesting to see more empirical work on the

formation and longevity of such values in family businesses. Finally, our

research has empirically highlighted the diverse nature of SEW. The quanti-
tative studies currently dominating the family business field, have problems
in investigating this diversity as SEW and is often captured indirectly by
overly simple and thus problematic measures (Miller & Le Breton-Miller,
2014). We believe that there is a need to pay more attention to the hetero-
geneity of SEW(s) in family firms. While it is fairly obvious what financial
wealth is, this is not the case for SEW. SEW for one individual does not
have to be so for his or her siblings. The complexity of the SEW construct
is further increased by the different dimensions of SEW (Berrone et al.,
2010). We believe that through an in-depth qualitative study we have been
able to shed some light on SEW as a heterogeneous phenomenon and call
for more such studies in the future.


1. The Swedish industry context data includes more than 200 speeches, 285
annual reports, analysis of 514 meetings in trade associations, reviews of several
trade magazines, as well as interviews with industry experts.
2. In Lamberg et al., 2,620 strategic actions in 17 pulp and paper companies
from Sweden, Finland, and the United States were coded over the period
1846 2003.
3. The family alternative proposed by the Frans branch was Mats Carlgren, a
member of the board since 1955 and Carl Kempe’s nephew. Lyberg (1984) describe
the succession process in detail and reveals that had to be a nonfamily mediator to
reach a conclusion.
4. According to Valeur (2000) and Boseaus (1949) MoDo was one of the leading
companies in research and development at the time. Valeur reports that in 1947
MoDo’s R&D employed 74 out of which 11 held an academic degree.
5. According to Valeur (2000) and Boseaus (1949) MoDo was one of
the leading companies in research and development at the time. Valeur
reports that in 1947 MoDo’s R&D employed 74 out of which 11 held an academic
The strategic alternatives are extensively analyzed in Peterson (1996) and in
Melander (1997, p. 155ff.). Some of the practitioners that discussed the alternatives
in speeches and articles are Enström, (MD, SCA, 1959); Lyberg, B. (MD, MoDo,
1959); Wohlfart, (Analyst, SCPF, 1967); Carlgren (Chairman of the board of
MoDo, 1968); and Sundblad (MD, Iggesund, 1979).
6. The Kreuger crash followed the death of Ivar Kreuger and hence the end of
the Kreuger empire in 1932. As a result of the crash, several international compa-
nies went bankrupt and were taken over by banks.
7. In Brunninge and Melander (2013) this point is illustrated and elaborated.
Family Firm Longevity and Resource Management 207


Ahnlund, N. (1917). Mo och Domsjö verken: Deras ägare och utveckling intill 1873. Stockholm:
Almqvist & Wicksells boktryckeri.
Anderson, P., & Tushman, M. L. (1990). Technological discontinuities and dominant
designs: A cyclical model of technological change. Administrative Science Quarterly, 35,
604 633.
Armstrong, C. E., & Shimizu, K. (2007). A review of approaches to empirical research on the
resource-based view of the firm. Journal of Management, 33(6), 959 986.
Bansal, P. (2003). From issues to actions: The importance of individual concerns and organiza-
tional values in responding to natural environmental issues. Organization Science,
14(5), 510 527.
Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of
Management, 17, 99 120.
Benavides-Velasco, C. A., Quintana-Garcıa, C., & Guzman-Parra, V. F. (2013). Trends in
family research. Small Business Economics, 40, 41 57.
Berghoff, H. (2006). The end of family business? The Mittelstand and German capitalism in
transition, 1949 2000. Business History Review, 80, 263 295.
Berrone, P., Cruz, C., & Gomez-Mejia, L. R. (2012). Socioemotional wealth in family firms:
Theoretical dimensions, assessment approaches, and agenda for future research. Family
Business Review, 25, 258 279.
Berrone, P., Cruz, C., Gomez-Mejia, L. R., & Larraza-Kintana, M. (2010). Socioemotional
wealth and corporate responses to institutional pressures: Do family-controlled firms
pollute less? Administrative Science Quarterly, 55, 82 113.
Blombäck, A., & Brunninge, O. (2013). The dual opening to brand heritage in family business.
Corporate Communications: An International Journal, 18(3), 327 346.
Boseaus, E. (1949). Utveckling av produktion och teknik i svensk massaindustri, 1857 1939.
Industrihistorisk skriftserie, nr 4. Stockholm: Svenska cellulosa- och trämasseföreningnarna.
Brunninge, O. (2005). Organizational self-understanding and the strategy process. Strategy
dynamics in Scania and Handelsbanken. JIBS Dissertation Series No. 027. Jönköping
International Business School, Jönköping.
Brunninge, O. (2009). Using history in organizations. How managers make purposeful refer-
ence to history in strategy processes. Journal of Organizational Change Management,
22(1), 8 26.
Brunninge, O., & Melander, A. (2013). Family-firm identity across generations. In
H. Berghoff, B. Weisbrod, U. Jensen, & C. Lubinski (Eds.). History by generations:
Generational dynamics in modern history. Göttingen: Wallstein Verlag.
Brunninge, O., & Melin, L. (2010). Continuity and change. Path-dependence and transforma-
tion in two Swedish multinationals. In J. Sydow & G. Schreyögg (Eds.), The hidden
dynamics of path-dependence (pp. 94 109). Basingstoke: Palgrave.
Brunninge, O., Nordqvist, M., & Wiklund, J. (2007). Corporate governance and strategic
change in SMEs. The effects of ownership, board composition and top management
teams. Small Business Economics, 29, 295 308.
Carlberg, M., & Scholander, A. (1989). Teknisk Uppkäftighet. Stockholm: Skogsindustrierna.
Casson, M. C. (1999). Entrepreneurship and the theory of the firm. In Z. A. Acs, C. Carlsson, &
C. Karlsson (Eds.), Entrepreneurship, small and medium-sized enterprises and the macro-
economy (pp. 45 78). Cambridge: Cambridge University Press.

Chrisman, J. J., Chua, J. H., & Steier, L. (2005). Sources and consequences of distinct famili-
ness. An introduction. Entrepreneurship Theory and Practice, 29, 237 247.
Croon, I. (2005). Utveckling, Förnyelse, Omvälvning. Grodor blir prinsar och tvärtom: en vandr-
ing i en industri och ett land i förvandling. Stockholm: Spearhead Production.
De Tienne, D. R., & Chirico, F. (2013). Exit strategies in family firms: How socioemotional
wealth drives the threshold of performance. Entrepreneurship Theory and Practice,
37(6), 1297 1318.
Dutton, J. E., Fahey, L., & Narayanan, V. K. (1983). Towards understanding strategic issue
diagnosis. Strategic Management Journal, 4, 307 323.
Dutton, J. E., & Penner, W. J. (1993). The importance of organisational identity for strategic
agenda building. In J. Henry, G. Johnson, & J. J. Newton (Eds.), Strategic thinking:
Leadership and the management of change. Chichester: Wiley.
Dyer, W. G. Jr. (1986). Cultural change in family firms: Anticipating and managing business and
family transition. San Francisco, CA: Jossey-Bass.
Eddleston, K., Kellermanns, F. W., & Sarathy, R. (2008). Resource configuration in family
firms: Linking resources, strategic planning and technological opportunities to perfor-
mance. Journal of Management Studies, 45(1), 26 50.
Ericson, M. (1991). Iggesundsaffären. Rationaliteter I en strategisk förvärvsprocess. Doctoral
thesis. EFI. Stockholm School of Economics.
Fagerfjäll, R. (2008). Entrepenörer i exportindustrins barndom. I Då förändrades Sverige.
E. Giertz (Red). Lund: Studentlitteratur.
Forsgren, M., & Kinch, N. (1970). Företagets Anpassning till Förändringar i Omgivande
System. Acta Upsaliensis. Studia Oeconomiae Negotiorum nr 2. Uppsala.
Gaunitz, S. (1979). Struktur, Strategi och resurser. In G. Utterström (Ed.), SCA 50 år. Studier
kring ett storföretag och dess föregångare. Sundsvall: SCA.
Gårdlund, T. (1951). Mo och Domsjö intill 1940: Dem ekonomiska utvecklingen. Stockholm:
Almqvist & Wiksell.
Gårdlund, T. (1986). MoDo 1940 1985. Örebro: MoDos Informationsavdelning.
Gibbert, M. (2006). Generalizing about uniqueness: An essay on an apparent paradox in the
resource-based view. Journal of Management Inquiry, 15(2), 123 134.
Glete, J. (1987). Ägande och Industriell Omvandling. Stockholm: SNS Förlag.
Glete, J. (1989). Long term firm growth and ownership organization. Journal of Economic
Behavior and Organization, 12, 329 351.
Gomez-Mejia, L. R., Cruz, C., Berrone, P., & Castro, J. De. (2011). The bind that ties.
Socioemotional wealth preservation in family firms. The Academy of Management
Annals, 5(1), 653 707.
Gomez-Mejia, L. R., Cruz, C., & Imperatore, C. (2014). Financial reporting and the protec-
tion of socioemotional wealth in family controlled firms. European Accounting Review,
23(3), 302 387.
Gomez-Mejia, L. R., Takács-Haynes, K., Núnez-Nickel, M., Jacobson, K. J. L., & Moyano-
Fuentes, J. (2007). Socioemotional wealth and business risks in family-controlled firms:
Evidence from Spanish olive oil mills. Administrative Science Quarterly, 52, 106 157.
Grant, R. M. (1991). The resource-based theory of competitive advantage. Implications for
strategy formulation. California Management Review, 33(3), 114 135.
Habbershon, T. G., & Pistrui, J. (2002). Enterprising families domain: Family-influenced
ownership groups in pursuit of transgenerational wealth. Family Business Review, 15,
223 237.
Family Firm Longevity and Resource Management 209

Habbershon, T. G., & Williams, M. L. (1999). A resource-based framework for assessing the
strategic advantage of family firms. Family Business Review, 12, 1 25.
Hall, A., Melin, L., & Nordqvist, M. (2001). Entrepreneurship as radical change in the family
business: Exploring the role of cultural patterns. Family Business Review, 14(3), 193 208.
Helfat, C. E. (2007). Dynamic capabilities. In Understanding strategic change in organizations.
Malden: Blackwell.
Hernód, T. (1942). Något om Cellulosaindustrins utveckling. In R. Calissendorff, I. Högbom, &
J. Wallenberg (Eds.), Studier i Svenskt Näringsliv tillägnad Jacob Wallenberg. Mölndal:
Nordiska Bokhandeln.
Ireland, R. D., Hitt, M. A., & Sirmon, D. G. (2003). A model of strategic entrepreneurship. A
construct and its dimensions. Journal of Management, 29(6), 963 989.
Järvinen, J., Ojala, J., Melander, A., & Lamberg, J.-A. (2013). The evolution of pulp and
paper industries in Finland, Sweden and Norway 1800 2005. In J.-A. Lamberg,
J. Ojala, M. Peltoniemi, & T. Särkkä (Eds.), The evolution of global paper industry
1800 2050 (Vol. XVIII). Heidelberg: Springer World Forest Series.
Jaskiewicz, P., Combs, J. G., & Rau, S. B. (2014). Entrepreneurial legacy. Toward a theory of
how some firms nurture transgenerational entrepreneurship. Journal of Business
Venturing, 30(1), 29 49.
Johnson, G. (1987). Strategic change and the management process. Oxford: Basil Blackwell.
Karlsson Stider, A. (2000). Familjen och firman. Doctoral dissertation defended at Stockholm
School of Economics.
Kempe, F. (2005). Slutet på en familjedynasti. MoDo 1982 1991. In I. Croon (Ed.),
Utveckling, Förnyelse, Omvälvning. Grodor blir prinsar och tvärtom: en vandring i en
industri och ett land i förvandling (pp. 214 223). Stockholm: Spearhead Production.
Kempe, M. (2006). Försvinnande spår. Boken om Lotty Bruzelius. Stockholm:
Lamberg, J.-A., Näsi, J., Ojala, J., & Sajasalo, P. (2006). The evolution of competitive strategies
in global forestry industries. Netherlands: Springer.
Lamberg, J.-A., Ojala, J., Peltoniemi, M., & Särkkä, T. (2012). The evolution of global paper
industry 1800 2050. Dordrecht: Springer.
Langley, A. (1999). Strategies for theorising from process data. Academy of Management
Review, 214(4), 691 710.
Layton, I. G. (1979). Hamnar och Sjöfart I övre Norrland En studie av utvecklingen från
1700-talets mitt och dess Industriella BakgrunD. I SCA 50 år. G. Utterström (Red).
Sundsvall: SCA.
Le Breton-Miller, I., & Miller, D. (2013). Socioemotional wealth across the family firm life-
cycle: A commentary on “Family business survival and the role of boards”.
Entrepreneurship Theory and Practice, 37(6), 1391 1397.
Lyberg, B. (1959, June 25). Skogsindustrin inför en förändrad värld. Speech at 19th
Norrlandsmässan. Speech reprinted in MoDo Insikt, no22, appendix.
Lyberg, B. (1984). Vandring i gamla kvarter. FrånLusitgknopp till Ekliden Bengt Lybergs
Melander, A. (1997). Industrial wisdom and strategic change The Swedish pulp and paper
industry 1945 1990. JIBS Dissertation Series No. 001. Jönköping University.
Melander, A. (2005). Industry-wide belief structures and strategic behaviour The Swedish pulp
and paper industry 1945 1980. Scandinavian Economic History Review, 53(1), 91 119.

Melander, A., Melin, L., & Nordqvist, M. (2011). The strategic arena approach to strategy
process research. In F. Kellermanns & P. Mazzol (Eds.), Handbook of research on strat-
egy process. Northampton, MA: Edward Elger Publishing Limited.
Miller, D., Breton-Miller, L., & Scholnick, B. (2008). Stewardship vs. stagnation: An empirical
comparison of small family and non-family businesses. Journal of Management Studies,
45(1), 51 78.
Miller, D., & Le Breton-Miller, I. (2011). Governance, social identity and entrepreneurial
orientation in closely held public companies. Entrepreneurship Theory and Practice,
35(5), 1051 1076.
Miller, D., & Le Breton-Miller, I. (2014). Deconstructing socioemotional wealth.
Entrepreneurship Theory and Practice, 38(4), 713 720.
Miller, D., & Shamsie, J. (1996). The resource-based view of the firm in two environments:
The Hollywood film studios from 1936 to 1965. Academy of Management Journal, 39,
519 543.
Mintzberg, H. (1990). The design school: Reconsidering the basic premises of strategic man-
agement. Strategic Management Journal, 11, 171 195.
Myhrman, J. (1984). Hur Sverige blev rikt. Stockholm: SNS Förlag.
Newbert, S. L. (2007). Empirical research on the resource-based view of the firm: An assess-
ment and suggestions for future research. Strategic Management Journal, 28, 121 146.
Nordström, G. (1993). Mo och Domsjö AB och arbetarorganisationerna intill 1940. Research
Report No. 31. Uppsala Papers in Economic History. Department of Economic
History, Uppsala University.
Ocasio, W. (1997). Towards an attention-based view of the firm. Strategic Management
Journal, 18, 187 206.
Ocasio, W. (2011). Attention to attention. Organization Science, 22(5), 1286 1296.
Ojala, J., Lamberg, J.-A., & Melander, A. (2010). Competitive behaviour and business innova-
tions in the forestry industry: Family firms and listed firms in comparison. In
P. Fernández Pérez & M. B. Rose (Eds.), Innovation and Entrepreneurial Networks in
Europe. Routledge International Studies in Business History. New York, NY:
Pearson, A. W., Carr, J. C., & Shaw, J. C. (2008). Toward a theory of familiness: A social
capital perspective. Entrepreneurship Theory and Practice, 32, 949 969.
Peterson, C. (1996). Finsk ingenjörskonst och Svenskt Imperiebyggande. En jämförande studie av
Svensk och Finsk Skogsindustri. Stockholm: SNS Förlag.
Pettigrew, A. (1990). Longitudinal field research on change: Theory and practice. Organization
Science, 1(3), 267 292.
Porter, M. (1990). The competitive advantage of nations. Houndsmills: MacMillan.
Rau, S. (2013). Resource based view of family firms. In L. Melin, M. Nordqvist, & P. Sharma
(Eds.), The SAGE handbook of family business. London: Sage.
Salvato, C., Chirico, F., & Sharma, P. (2010). A farewell to business. Championing exit and
continuity in entrepreneurial firms. Entrepreneurship & Regional Development, 22(3),
321 348.
Salvato, C., & Melin, L. (2008). Creating value across generations in family-controlled busi-
nesses: The role of family social capital. Family Business Review, 21(3), 259 276.
Schein, E. (1983). The role of the founder in creating organizational culture. Organizational
Dynamics, 12(Summer), 13 28.
Schön, L. (2008). Entreprenörskap i den tidiga industrialiseringen. I Då förändrades Sverige. E.
Giertz (Red). Lund: Studentlitteratur.
Family Firm Longevity and Resource Management 211

Schreyögg, G., & Kliesch-Eberl, M. (2007). How dynamic can organizational capabilities be?
Toward a dual-process model of capability dynamization. Strategic Management
Journal, 28(9), 913 933.
Schreyögg, G., Sydow, J., & Holtmann, P. (2011). How history matters in organisations: The
case of path dependence. Management and Organizational History, 6(1), 81 100.
Sciascia, S., Mazzola, P., & Kellermanns, F. W. (2014). Family management and profitability
in private family-owned firms: Introducing generational stage and the socioemotional
wealth perspective. Journal of Family Business Strategy, 5, 131 137.
Selznick, P. (1957). Leadership in administration. New York, NY: Harper & Row.
Sharma, P., & Carney, M. (2012). Value creation and performance in private family firms:
Measurement and methodological issues. Family Business Review, 25(3), 233 242.
Sharma, P., & Manikutty, S. (2005). Strategic divestments in family firms. Role of family
structure and community culture. Entrepreneurship Theory and Practice, 29(3),
293 311.
Sharma, P., & Sharma, S. (2011). Drivers of proactive environmental strategy in family firms.
Business Ethics Quarterly, 21(2), 309 334.
Sirmon, D. G., & Hitt, M. A. (2003). Managing resources: Linking unique resources, manage-
ment and wealth creation in family firms. Entrepreneurship Theory and Practice, 27,
339 358.
Sirmon, D. G., Hitt, M. A., Ireland, R. D., & Gilbert, B. A. (2011). Resource orchestration to
create competitive advantage. Breadth, depth, and life cycle effects. Journal of
Management, 37(5), 1390 1412.
Sölvell, Ö., Zander, I., & Porter, M. E. (1993). Advantage Sweden. Stockholm: Norstedt
Streyffert, T. (1950). Sveriges Skogar och Skogsindustrier. Stockholm: Kooperative Förbundets
Sundblad, L. G. (1979, March 19). Skogsnäringens läge och alternativa strategier. Speech at
Svensk Papperstidning. (1967). 15 page 48 Obituary on Carl Kempe.
Svensk Skogsindustri i Omvandling. (1972). Skogsindustriernas Samarbetsutskott. Stockholm:
Svenska Industrin vid Kvartalsskiftet 1925. (1926). Sveriges Industriförbund. Stockholm:
Sveriges Industriförbund.
Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management.
Strategic Management Journal, 18, 509 533.
Unger, G. (1967). Carl Kempe: Den siste patriciern. Stockholm: Bonniers.
Valeur, C. (2000). Papper och massa i Ångermanland. Köping: Skogsindustriernas historiska
utskotts skriftserie.
Van Gils, A., Voordeckers, W., & van den Heuvel, J. (2004). Environmental uncertainty and
strategic behavior in Belgian family firms. European Management Journal, 22(5),
588 595.
Zellweger, T., Nason, R., Nordqvist, M., & Brush, C. (2011). Why do family firms strive for
nonfinancial goals? An organizational identity perspective. Entrepreneurship Theory and
Practice, 37(2), 229 248.
Zellweger, T. M., Kellermanns, F. W., Chrisman, J. J., & Chua, J. H. (2012). Family control
and family firm valuation by family CEOs: The importance of intentions for transge-
nerational control. Organization Science, 23(3), 851 868.


The case of MoDo 1873 1990 is based on the following sources:

• 22 manuscripts of speeches given by the CEO or the chairman of the
board in the period 1945 1990.
• Annual reports on the period 1945 1995.
• A review of the staff Magazine “MoDo Insikt” in the period 1960 1995.
• Personal interviews with seven top management representatives (con-
ducted in the period 1991 1995) and five interviews with three represen-
tatives from the two branches (“Seth” and “Frans”) of the owner family
in 2012 2014.
Below we list the most prominent secondary sources used in the research
Name of Author, Publication Title of the Source Topic and Time Covered
Date, and Relation to the
Kempe Family

1 Ahnlund (1917) Mo och Domsjö verken: MoDo’s development until the

Deras ägare och utveckling foundation 1873
intill 1873
2 Gårdlund (1951). Mo och Domsjö intill 1940: MoDos overall development
Researcher in Economic Den ekonomiska 1873 1940 mainly from a
History utvecklingen financial perspective
3 Unger (1967) Carl Kempe: Den siste Biography on Carl Kempe
patriciern (MD in MoDo 1917 1947)
4 Lyberg (1984). MoDo Vandring i gamla kvarter. Memoirs in which top
employee 1943 1972. MD, Från Lustigknopp till management issues and
1959 1972 Ekliden succession issues in the Kempe
family are described
5 Gårdlund (1986). MoDo 1940 1985 Company biography,
Researcher in Economic 1940 1985. The book ends
History with a company statement
authored by the chairman of
the board, Matts Carlgren
6 Carlberg and Scholander Teknisk Uppkäftighet Technological innovations in
(1989). Technology the Swedish pulp and paper
researchers and Industry industry. Two chapters are
analyzers based on interviews with
prominent researchers related
to MoDo
Family Firm Longevity and Resource Management 213

(Continued )
Name of Author, Publication Title of the Source Topic and Time Covered
Date, and Relation to the
Kempe Family

7 Ericson (1991). Researcher Iggesundsaffären. The views of all parties

Rationaliteter i en strategisk involved in the realization if
förvärvsprocess. Phd thesis the “third block” strategy are
8 Nordström (1993). Mo och Domsjö AB och Thorough examination of
Researcher arbetarorganisationerna MoDo’s industrial relations
until 1940. Phd thesis 1873 1940
9 Croon (2005). Chief Utveckling, Förnyelse, MoDos entire history
Operating Officer in R&D Omvälvning. Grodor blir described by a high-ranked
MoDo 1963 1976 prinsar och tvärtom: en employee
vandring i en industri och
ett land i förvandling
10 Kempe (2005). A Slutet på en familjedynasti. Memories from close
grandchild to Carl Kempe. MoDo 1982 1991 collaborator to Matts
A Frans Kempe relative Carlgren 1982 1991
11 Kempe (2006). (Margareta Försvinnande spår. Boken Biography on Lotty Kempe
is married to Carl Kempe om Lotty Bruzelius and the Kempe family (the
jr. The chairman of the founder of the Kempe
Kempe foundation. A foundation) 1873 1936
Seth-Kempe relative)
This page intentionally left blank


Magdalena Markowska, Charmine E. J. Härtel,

Ethel Brundin and Amanda Roan


Despite recognition of the centrality of emotions in entrepreneurship,

little attention has been given to role of emotions in the development of
entrepreneurial identity or enactment of entrepreneurial role. The contri-
bution of the chapter is in the development of a dynamic model of the
process leading to identification or dis-identification as an entrepreneur.
In this chapter, we develop a dynamic model of the process leading to
identification or dis-identification as an entrepreneur. We theorize that
the driver behind an individual’s decision to become an entrepreneur, and
their significant emotional experiences in the entrepreneurial role, influ-
ence the likelihood of following an identification or dis-identification
cycle. Specifically, our framework proposes that positive emotions

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 215 239
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011009

strengthen approach motivation and identification with the role, while

negative ones foster avoidance motivation and dis-identification. We
argue that contextual embeddedness can prompt transition between these
two cycles. Our theorization provides new insights into methods of ana-
lyzing the role of emotions in the entrepreneurial process, more specifi-
cally in the process of entrepreneurial identity crafting. These insights
also can be translated into studying the crafting of any professional
Keywords: Entrepreneurship; role identity; emotions theory;
motivation; contextual embeddedness


“I am doing the best profession in the world. I believe in what I do and am happy to be
able to contribute to pushing the science and development further”. These are the
words of Olga Malinkiewicz, co-owner of Saule Technologies, a company working on
creating a new generation of photovoltaic cells for converting solar energy into electri-
city. The decision to become an entrepreneur emerged after Olga’s breakthrough dis-
covery, which could revolutionalize the energy industry by reducing the cost of
generating electricity. The positive feedback received from other scientists and the
energy industry combined with her passion for physics pulled Olga into commercializ-
ing her technology and becoming a successful entrepreneur. (Kazimierczuk, 2015)

Individuals experience the world through emotional and cognitive processes

(Härtel, 2008) and they make sense of their experiences by constructing
self-stories of who they are, how they feel, and how they act (McAdams,
1996). Researchers have found that the development of an entrepreneurial
identity can depend on a positive perception of entrepreneurship as well as
positive feelings about being an entrepreneur and that emotions play a cen-
tral role in entrepreneurial processes and performance (Brundin, Pazelt, &
Shepherd, 2008; Cardon, Zietsma, Saparito, Matherne, & Davis, 2005;
Markman, Balkin, & Baron, 2002). As the above example of entrepreneur
Olga Malinkiewicz shows, the positive emotions she felt from her discovery
and reaction from others in the industry spurred her on to investing fully in
her discovery and embracing an entrepreneurial identity.
Despite the growing recognition within the entrepreneurship field of the
centrality of emotions in entrepreneurialism (Cardon, Foo, Shepherd, &
Wiklund, 2012), little attention has been given to the potential function of
emotions in the development of an entrepreneurial identity or in the
Entrepreneurial Identification and Dis-Identification 217

enactment of the entrepreneurial role (Simpson, Irvine, Balta, & Dickson,

2015). In this chapter, we respond to recent calls (e.g., Cardon et al., 2012;
Simpson et al., 2015) to extend the research on entrepreneurship and emo-
tion by developing a theoretical framework and propositions to inform
future research agendas. We draw together research on emotions, identity,
and entrepreneurship to develop a theoretical model of the processes lead-
ing to identification or dis-identification as an entrepreneur. Our frame-
work suggests first, that the driver behind entering an entrepreneurial role
plays a key role in a person’s motivation orientation (i.e., an approach or
avoidance orientation) toward entrepreneurship. An approach motivation
is generally associated with positive emotions (Elliot, Eder, & Harmon-
Jones, 2013) and thus we predict it will support a positive emotions feed-
back loop, which we argue is critical to the emergence and reinforcement of
developing a positive entrepreneurial identity. In contrast, an avoidance
motivation is generally associated with negative emotions (Elliot et al.,
2013) and thus we predict to support a negative emotions feedback loop,
which, we theorize fosters a dis-identification cycle.
Second, our dynamic model of entrepreneurial dis/identification high-
lights the importance of contextual embeddedness, offering an explanation
of how emotions can prompt an entrepreneur to move from a dis-
identification cycle to an identification cycle and vice versa. In this process,
the entrepreneur’s mindset (growth or fixed) plays an important role. We
argue that entrepreneurs with a fixed mindset are more dependent on the
environment’s feedback in order to continue into their role than if they
have a growth mindset, which makes them less vulnerable to such feedback.
Fig. 1 presents our proposed theoretical model.
The chapter is organized as follows. First we define the entrepreneurial
identity and the context in which a positive entrepreneurial identity can
emerge. Second, with the assistance of our dynamic model of entrepreneur-
ial (dis)identification, we explicate the role of contextual embeddedness and
emotions in crafting a positive entrepreneurial identity as well as their role
in dis-identifying as an entrepreneur. We do this by discussing the role of
contextual embeddedness in entrepreneurial dis/identification. We then
demonstrate the importance of two contextual features in entrepreneurial
identity development: the driver to become an entrepreneur (push/pull fac-
tors) and the role of emotions in the perception and enactment of an entre-
preneurial role and identity. Building on this foundation of contextual
embeddedness, we explicate a cycle of entrepreneurial identification and
dis-identification and discuss the feedback loop between motivation

Dis-identification Cycle Identification Cycle

Motivation PUSH FACTORS: e.g., sees PULL FACTORS: e.g., sees

Direction: entrepreneurial role as only choice for entrepreneurial role as enabling Motivation
Avoidance career; receives negative feedback from career vision; receives positive Direction:
others when enacting entrepreneurial feedback from others when Approach
role enacting entrepreneurial role


Increases Positive
Increased Motivation to Emotions e.g.,
Negative Motivation to Engage with Passion,
Emotions e.g., Avoid/Exit Entrepreneurial Excitement
Anxiety, Shame/ Entrepreneurial Role
Guilt Role

Development of
Inhibits Entrepreneurial
Development of Identification
Entrepreneurial Contextual
Identification embeddedness

Fig. 1. Dynamic Model of Entrepreneurial Dis/Identification. Source: Created by


orientation and emotions as the key mechanism by which an entrepreneur

maintains a particular dis/identification cycle or shifts to the opposite cycle.
Finally, we discuss the theoretical and practical implications of the model
and directions for future research.


Ashforth and Sluss (2008, p. 807) argues that “identity (what something is)
and identification (the extent to which one includes that identity as a partial
definition of self) inform how individuals think, feel, and act.”
Entrepreneurial identity is an emerging area of research (e.g., Hoang &
Gimeno, 2010; Shepherd & Haynie, 2009b). It is commonly used to refer to
the construction of the identity of being an entrepreneur within an existing
set of societal conditions (Clarke & Holt, 2010). Individuals have multiple
and intersecting identities, some claimed, some socially imposed, and some
that reflect both processes (DeRue & Ashford, 2010). For example, an indi-
vidual who is visibly recognized in society as a female may claim identity or
may claim another, such as medical doctor or mother. Identities are fluid
Entrepreneurial Identification and Dis-Identification 219

and subject to identity work which entails the ongoing social and relational
exchange processes involved in the formation, maintenance, and changing
of an individual’s identities (DeRue & Ashford, 2010; Ybema et al., 2009).
Consequently, in the process of constructing an “authentic” self, indivi-
duals simultaneously engage in the process of dis-identification from a pre-
vious identity (Costas & Fleming, 2009).
Although much of the research on identity tends to focus on an indivi-
dual’s current identities, it is important to recognize that individuals have
aspirations about who they want to be and the goals they want to achieve
(Farmer, Yao, & Kung-Mcintyre, 2011; Thornborrow & Brown, 2009). By
way of illustration, entrepreneurs may claim an entrepreneurial self to
achieve desired outcomes such as gaining access to resources they desire or
to legitimize certain actions (Lounsbury & Glynn, 2001).
Although interest in entrepreneurial identity is growing and the process
of the structuring of an entrepreneurial identity is well recognized in the lit-
erature (Fauchart & Gruber, 2011), there is still no agreed definition nor is
there consistent use of the term (e.g., Hoang & Gimeno, 2010 talk about
founder identity). Nevertheless, researchers tend to agree that entrepreneur-
ial identity is a role identity that is created by individuals in the process of
evaluative agency (Down, 2006). Being an entrepreneur can be perceived as
holding a specific role in the social structure. As a case in point, Down and
Warren’s (2008) found the metaphor of hero being used to describe an
entrepreneur. Unlike professional identity, entrepreneurial identity is often
being constructed only to achieve specific goals, such as achieving financial
support (Warren, 2004) or the creation of self-esteem (Down, 2006). Thus,
entrepreneurial identity can be constructed as a way to present oneself in a
desirable way (Lave & Wenger, 1991).
Extant research suggests that the entrepreneurial identity reflects a belief
about what it means to be an entrepreneur (Krueger, 2007). However, the
development of identity is commonly construed as a choice between
options, for instance becoming a chef or an educator (Gollwitzer &
Kirchhof, 1998). This means that to enact the identity, individuals need to
recognize their role. Schein (1978) argues that with occupational experi-
ence, individuals gradually develop clearer self-concepts of their talents and
abilities, motives, and values. Since this internalization process is embedded
in the socioeconomic context, the entrepreneur, through interactions with
others, is influenced by how tasks are defined and impressions shaped of
what is and what is not part of the entrepreneurial role (Salancik & Pfeffer,
1978). Because of the role of appraising self and others in the formation of
identity, a cognitive appraisal perspective of emotions would lead to the

proposition that entrepreneurs’ emotions will affect the course of entrepre-

neurial identity development. We thus claim that feedback about their
entrepreneurial role performance gives rise to emotional reactions, for
example, joy, happiness, anger, and frustration. Thus, we advance a dual
process perspective of entrepreneurial identity, where cognitive factors
encompassing the self-concepts about oneself in the role are key (Ashforth
& Kreiner, 1999; Ashforth & Saks, 1995), but so too are emotional factors.
Analyzing entrepreneurial identity from this “whole being” perspective of
taking in both emotional and cognitive aspects also accords with the view
of entrepreneurial identity as a meta identity (e.g., Hoang & Gimeno, 2010;
Shepherd & Haynie, 2009a), and also allows for the consideration of the
well-being of the entrepreneur.
What it is to be an entrepreneur is often expressed as a narrative. Down
and Warren (2008) found that the narratives surrounding entrepreneurial
identities often result in the use of metaphors and clichés. Their ethno-
graphic study of a small industrial firm in the United Kingdom identified
three main clichés associated with “entrepreneurial talk.” These were risk
and bravery, ambition and growth, and self-sufficiency. The authors found
that the narratives created by their research subjects were not consistent
but were similar to the discourse in the general public on entrepreneurship
or rhetoric coming out of business schools. Indeed, the entrepreneurs they
researched had low role identification, which could explain why they
emphasized who they were not. This so-called “negative identity”
(Kärreman & Alvesson, 2001), where individuals define themselves by
negation of a specific trait or characteristic, is created in order to make
sense of where certain things and others should be in an individual’s perso-
nal narratives (Fauchart & Gruber, 2011). Thus, the perception of the
desirability of the entrepreneurial role influences the shaping of an entre-
preneurial identity (cf. Goffman, 1959). The stereotypes emerging from
institutionalized values and expectations may also constrain or enable iden-
tity development.
Numerous studies indicate that individuals purposefully engage in craft-
ing identities that are important for them and shaping the role in a way
that is consistent with their perception of what it means to be in this role
(Fauchart & Gruber, 2011; Ibarra & Barbulescu, 2010; Powell & Baker,
forthcoming; Ybema et al., 2009). For example, Powell and Baker (forth-
coming) argue that while entrepreneurs purposefully engage in turning into
“who they are,” they also engage in turning into “who they want to be.”
Similarly, Lave and Wenger (1991) found that entrepreneurs use tools to
Entrepreneurial Identification and Dis-Identification 221

present themselves in a desirable and culturally appropriate way. This

means that crafting an identity requires agency on the side of the individual
(Wry, Lounsbury, & Glynn, 2011). For example, the university academics
in the study by Jain, George, and Maltarich (2009) actively engaged in
maintaining cherished identities as research scientists when faced with the
necessity to be involved in commercialization activities as well.
Next, we introduce our model, starting with the assumptions that have
guided our theorization.



Starting Assumptions

In developing our theoretical framework, we make two fundamental

assumptions. First, we assume that enacting a role and crafting a corre-
sponding identity is a contextually embedded process. Identity theory
(Stryker, 1980) accepts that identities are based on socially negotiated
meanings of roles that individuals hold in the social structure. Individuals
make sense of what it means to hold a given role and a given identity;
hence their mindsets and their emotions are likely to influence how they
perceive and how they enact the entrepreneurial identity. The crafting pro-
cess is likely to be idiosyncratic over time and across contexts as well as
across different individuals. Further the notion of contextual embeddedness
stresses that individuals do not act in a vacuum, but are socially embedded
and their actions are reciprocally influenced by their context (Dubini &
Aldrich, 1991; Welter, 2011).
Our second assumption follows the research by Powell and Baker
(forthcoming) which indicates that individuals may have single or multiple
salient identities, and the entrepreneurial identity may be just one of these
identities. As a consequence, the salience of any given identity will differ
among individuals. For example, an individual with an occupational iden-
tity may enact an entrepreneurial identity as an additional salient identity
adding it to the portfolio of currently held salient identities (e.g., a cook
and an entrepreneur), or may engage in transforming his or her occupa-
tional identity into an entrepreneurial identity to maintain only one salient

Overview of the Model

The fundamental premise of our model is that the features of contextual

embeddedness affect the salience of entrepreneurial identity and the purpo-
seful crafting of that identity by the individual (see Fig. 1). We focus on the
entrepreneur’s characteristics, in particular his or her motivation (or rea-
son) to become an entrepreneur (push or pull factors) and the entrepre-
neurial mindset (growth or fixed). We then, consider how the affective
aspects of the experience influence the salience of entrepreneurial identity
and the purposeful crafting of the entrepreneurial identity. More specifi-
cally, we distinguish between positive and negative emotions from perfor-
mance feedback on the enactment of the entrepreneurial role and
The general idea is thus that contextual embeddedness influences the
enactment of the entrepreneurial role and identity. We use the term contex-
tual embeddedness to emphasize the intersubjective experiencing of reality.
In other words, we theorize that the influence of contextual embeddedness
is visible in efforts individuals put into enacting roles and behaviors that
are socially desirable in their context. Moreover, we theorize that with
increased contextual embeddedness, the purposeful crafting of the entrepre-
neurial role increases. This means that the more entrepreneurs identify with
the entrepreneurial role, the more salient the entrepreneurial identity
becomes. Consequently the more effort they put into adopting behaviors
expected of such a role.

Entrepreneur’s Context Characteristics

In this section we consider in more detail the characteristics of entrepre-

neurs that are likely to affect whether they are likely to identify or dis-
identify as an entrepreneur.

Push and Pull Motivation to Become an Entrepreneur

The decision to engage in an entrepreneurial endeavor, we argue, is a stra-
tegic decision that in many ways impacts an individual’s perception of his
or her well-being and the perception of who they are. The decision itself
may be a result of the individual being pushed or pulled into entrepreneur-
ship (Amit & Muller, 1995; Kirkwood, 2009; Schjoedt & Shaver, 2007).
For example, if a creative chef realizes that his or her creativity and vision
cannot be realized in his/her current employment, he or she may decide to
Entrepreneurial Identification and Dis-Identification 223

enter self-employment to be able to realize their own visions and to develop

their own ideas. On the other hand, an artist may have difficulty finding a
job within existing employment opportunities and feel as if he or she is
forced to enter entrepreneurship as the only viable option to work in their
chosen field and earn money. In the first case, we talk about being “pulled
into” entrepreneurship and in the case of the artist we talk about being
“pushed into” entrepreneurship. The two different types are likely to
engage in a different way of creating their entrepreneurial identity.
Intuitively, the “pulled in” entrepreneur is likely to perceive entrepreneur-
ship as a positive vehicle for achieving career goals. In such a case, the
emerging entrepreneurial identity is likely to be a desired development
where they actively engage in shaping its content. In contrast, “pushed-in”
entrepreneurs may be more reluctant to enact an entrepreneurial identity
feeling that it does not reflect the desired development in their lives and
careers and represents high risk. Exhibit 1 illustrates this relationship.
To reiterate, individuals pulled into entrepreneurship are more likely to
experience positive emotions, like excitement, passion, and joy related to
the new career direction, while individuals feeling pushed into entrepre-
neurship may feel more negative emotions, such as anxiety, fear, and frus-
tration with the current situation (Schjoedt & Shaver, 2007). Experienced
emotions may influence the intention to enter and sustain in entrepreneurial
action. For example, positive emotions related to decision autonomy and
to the experience as such may contribute to higher identification with the
role of being an entrepreneur and subsequently to the development of a
more salient entrepreneurial identity. In contrast, if negative emotions are
related to entrepreneurial action, the entrepreneurial identity may not
emerge or be insignificant for the individual.
Consequently, we propose the following:
Proposition 1. Individuals who become entrepreneurs because they are
pulled into entrepreneurship will have a more salient entrepreneurial
identity than individuals who became entrepreneurs because they are
pushed into it.

Mindset, Perception of an Entrepreneurial Role, and Entrepreneurial Identity

The crafting of the entrepreneurial identity is influenced by the perception
of the entrepreneurial role that is negotiated in the interaction with the con-
text and shaped by the general worldview of the individual (Down &
Reveley, 2004; Lounsbury & Glynn, 2001; Wry et al., 2011). In order to
understand how entrepreneurs craft their identity, it is necessary to

Exhibit 1 The Pull and Push Factor.

Fredrik represents the pull factor. He was 16 years old when he dis-
covered his love for food. He decided to enter a culinary school to
train to become a chef, but quickly realized that working for others
had its limitations since it meant that he was not able to realize his
creativity and that he would have to follow the directions of his head
chef and the owner of the restaurant where he subsequently worked.
A desire to become an entrepreneur and to own his own restaurant,
where he could create the local modern and healthy food which was
his passion was growing within him. Encouraged by his friend and
mentor, he decided to strike out on his own and opened his own res-
taurant in 2005. He felt proud and excited as he worked on the con-
cept for the restaurant making sure that the food, the ambience, and
the décor coexisted in harmony. Fredrik talks about himself as a res-
taurateur, as a way to picture himself as the entrepreneur who uses
his wits to generate more business ideas and bring joy of the food to
health-conscious consumers. His words are reflected in his actions,
and some years later he engaged in new business activities by starting
a new successful catering business and developing a gourmet take-out
Mikael, who represents the push factor, worked as an architect in
Spain and decided to move to Sweden with his family after losing his
job four years ago. Mikael searched for an architect job for over a
year but, not speaking any Swedish, he was unsuccessful. Finally, his
family convinced him to start his own small restaurant and make a
living this way. Mikael started a small tapas bar which he runs him-
self. However, Mikael considers this just a temporary job so that he
can support his family; when asked about who he is, Mikael persis-
tently responds that he is an architect who is cooking for living.

understand their mindsets and, in particular, the tools that they implement
to manage their identities (Jain et al., 2009). More specifically, whether the
entrepreneur holds the view that skills and abilities are malleable (a growth
mindset) or are fixed (a fixed mindset) (Dweck, 1999) will shape how an
individual perceives the potential to acquire and perform a new role and
the new identity of an entrepreneur successfully.
Entrepreneurial Identification and Dis-Identification 225

This entrepreneurial mindset is what helps individuals to progress in the

entrepreneurial process (McGrath & MacMillan, 2000). However,
Achtenhagen and Johannisson (2013) claim that an entrepreneurial mindset
does not require active immersion in an entrepreneurial endeavor but is a
prerequisite for enacting an entrepreneurial identity (see Exhibit 2 for an
To elaborate, Dweck, Chiu, and Hong (1995) argues that individuals
with a growth mindset are likely to approach all challenges as learning
opportunities, not shying away from new experiences and in the face of set-
backs will not give up but be persistent and try to understand how they can

Exhibit 2 The Growth and Fixed Mindset.

Cecilia, who represents the growth mindset, is a well-known architect;
her designs have received prizes for innovativeness of form and mate-
rials used. Three years ago, she decided to open her own architecture
office. Even though she did not know what it meant to run a com-
pany, she was convinced that it was something she would learn
quickly if she put in sufficient effort. Cecilia believes that it is impor-
tant to think outside the box, to take risks and to try new things and
if they do not work, try again and again until they finally do. She
believes that practice can help achieve perfection because there is
always potential to learn. Today she is a successful entrepreneur who
not only offers interesting and bold designs but also embraces the
demands of the new role and identity, making sure that her company
attracts new customers, enters new market segments thus allowing
her company to grow.
In contrast, her former colleague, Patrick, represents the fixed
mindset. He decided to follow Cecilia’s example and opened his own
design office shortly after Cecilia made her move. However he is not
as successful as an architect as he used to be. He cannot understand
why since over the years, he has tried to retain the clean form of his
design approach. However, Patrick’s designs lack the creativity of
Cecilia’s designs. Also, Patrick has not really embraced the new role
of a business owner as well as architect. Focusing exclusively on his
architect identity, he often forgets that he also needs to attract new
customers, respond to their needs, and to create value for his com-
pany to be able to prosper.

improve their behavior and will put more effort next time around
(Blackwell, Trzesniewski, & Dweck, 2007; Dweck, 2010). On the other
hand, individuals characterized by a fixed mindset are likely to be afraid of
a negative external assessment of their competence and intelligence; and as
a result shy away from undertaking new challenging tasks and exercises.
The perception that their competence and intelligence are fixed may be
additionally inhibiting in terms of identifying with the new entrepreneurial
role. Therefore, individuals with a fixed mindset may refrain from actively
crafting an entrepreneurial identity, because they believe that they cannot
influence the level of their entrepreneurial capability and because they per-
ceive that their competence in the entrepreneurial role would be exposed to
the risk of negative assessment. Consequently, we propose that:
Proposition 2. Entrepreneurs with a growth mindset will engage in higher
levels of purposeful crafting of their entrepreneurial identity than entre-
preneurs with a fixed mindset.

Contextual Embeddedness and Purposeful Crafting of

Entrepreneurial Identity

The foregoing discussion sets the stage for understanding the emergence
and course of dis/identification as an entrepreneur. We acknowledge that
crafting an identity is an ongoing process and that entrepreneurs are active
shapers of their identities (Kira & Balkin, 2014). The crafting of an entre-
preneurial identity occurs through incorporation of different life experi-
ences and emotions that accompany these experiences and it requires
intentionality and agency from the individuals (Bandura, 2001; Wry et al.,
2011). One way in which such an identity is created is through narratives
(Ibarra & Barbulescu, 2010), where an identity emerges as a result of an
“internalized and evolving story that results from a person’s selective
appropriation of past, present and future” (McAdams, 1996, p. 486). Thus,
the development of an entrepreneurial identity could be conceptualized as a
specific type of evolving self-narrative with a specific kind of meaning (an
“entrepreneurial” meaning) that imbues the entrepreneur’s life experiences
with meaning, and integrates what could be disparate life episodes and
events into a coherent life trajectory that can be understood by themselves
and by others (Villanueva & Markowska, 2013).
In the desire to create a meaningful and legitimate image, individuals,
and entrepreneurs in particular engage in a process that combines, on the
Entrepreneurial Identification and Dis-Identification 227

one hand, the “internal strivings” with “external prescriptions” and, on the
other hand, balances the self-presentation of the role with how the role is
received by others (Ybema et al., 2009). More specifically, Ybema et al.
(2009) suggest that:
Identity formation can be conceptualized as a complex, multifaceted process which pro-
duces a socially negotiated temporary outcome of the dynamic interplay between inter-
nal strivings and external prescriptions, between self-presentation and labeling by
others, between achievement and ascription and between regulation and resistance.
(Ybema et al., 2009, p. 301)

This means that the identity crafting is influenced by contextual embedd-

edness (Alvesson & Willmott, 2002) and that the meaning of the role is
negotiated between different actors in interaction occurring in a socioeco-
nomic context. Subsequently, individuals develop a “culturally appropriate
self” who would transmit values, cognitions, and emotions deemed appro-
priate in their community (Lave & Wenger, 1991; Van Maanen, 1998).
Individuals engage in active storytelling to present their identity in a par-
ticular way (McAdams, 1996), to validate their claims, and present the
desired identities (Ashforth, 2001; Van Maanen, 1998). The more the entre-
preneurs identify with their new roles and the more these identities become
salient (describing who they are and who they want to become), the more
effort these individuals will put into creating a convincing story and convin-
cing identity (see Exhibit 3 for an illustration). For example, an aspiring
chef who dreams of becoming a restaurateur will seek to create behaviors
and tell stories about themselves and their passion for cooking by eliciting
the characteristics and/or similarities of recognized chefs and emphasizing
any similarities or connections they have with their role models. The more
this individual is embedded in the professional chefs community and shares
the language, routines and practices of this group, the more likely this aspir-
ing chef-restaurateur is to engage in developing an entrepreneurial identity
that fits into the perception of who an entrepreneurial restaurateur is and
how they act. In other words, the extent to which entrepreneurs imbue their
self-narratives with cognitive connotations and affective meanings related to
their entrepreneurial role, the more developed the story and the more salient
the identity for that entrepreneur. This is to a high degree an emotional pro-
cess, where the story is told, appeals to the emotions of the audience, such
as feelings of being part of that story. Hence, we propose:
Proposition 3. Higher entrepreneurial identity salience will be associated
with higher levels of purposeful crafting and this relationship will be
mediated in part by the cognitive-affective contextual embeddedness.

Exhibit 3 The Cognitive-Affective Contextual Embeddedness.

When Rene Barbier started his adventure with Priorat, he thought he
would produce wines for himself, his family, and friends. He quickly
realized that there was something unique about the Priorat wines; his
knowledge of winemaking and the wine market allowed him to see an
opportunity for bringing the Priorat wine to the market. As he
started to see himself more as an entrepreneur, he began to realize
that he would not be able to make the breakthrough alone. He
tapped into his social network and set about creating a collective
identity of the region using bootstrapping techniques (e.g., jointly
producing wine and sharing resources but selling under different
names) to produce and market Priorat wines. Part of the marketing
approach was to tell stories about himself, his friends, and the wines.
He actively engaged in crafting his entrepreneurial identity emphasiz-
ing initially the similarity of the Priorat wine to the recognized wine
regions, such as Bordeaux and/or Rioja. Later, when the wine had
been recognized he started to stress the uniqueness and distinctiveness
of his wines. His knowledge of the wine market, the close cognitive-
affective embeddedness provided him with the content and context in
which to create his entrepreneurial identity.

Emotion and the Enactment of an Entrepreneurial Identity

As stated above, individuals experience the world through emotional and

cognitive processes (Härtel, 2008) and constructing self-stories of who they
are, how they feel, and how they act (McAdams, 1996). Research also indi-
cates that human well-being depends on the relative number of positive
emotions to negative emotions an individual experiences over time (Härtel,
2008). Besides the influence of experienced affective events on well-being,
specific individual and organizational strategies to proactively and reac-
tively account for affective events influence an individual’s well-being out-
comes (Down & Reveley, 2004; Härtel, 2008; Kira & Balkin, 2014).
Welpe, Spörrle, Grichnik, Michl, and Audretsch (2012) found that
emotions are central to the exploitation of ideas, with fear reducing
exploitation and joy and anger increasing exploitation. Similarly, Hayton
and Cholakova (2012) argue that affect influences entrepreneurs’
Entrepreneurial Identification and Dis-Identification 229

development of ideas such that positive emotions help entrepreneurs to

take in and process information with novel and creative thinking as an
outcome. Positive and negative emotions also influence opportunity eva-
luation, as well as risk perceptions and risk preferences (Foo, 2011).
According to Foo (2011), entrepreneurs induced to anger and happiness
showed lower risk estimates for a start-up whereas entrepreneurs induced
to fear and hope showed higher risk estimates. So too do emotions play a
role in entrepreneurs’ responses to failing projects, impacting the indivi-
dual entrepreneur’s ability to learn and bounce back (Jenkins, Wiklund, &
Brundin, 2014; Shepherd, 2003).
In developing this model, we follow Cardon et al.’s (2012) suggestion
that entrepreneurial emotions refer “to the affect, emotions, moods, and/or
feelings of individuals or a collective that are antecedent to, concurrent
with, and/or a consequence of the entrepreneurial process, meaning the
recognition/creation, evaluation, reformulation, and/or the exploitation of
a possible opportunity” (p. 3). The cognitive aspect of the definition allows
us to link emotions to the mindset that an entrepreneur holds. Emotion
and cognition are not easily separable concepts since emotions influence
and are influenced by motivation and former experiences. Both are how-
ever necessary for “healthy” functioning and further development of the
entrepreneurial identity and feelings about a certain experience influence an
entrepreneur’s willingness for further exposure to similar experiences (cf.
Hayton & Cholakova, 2012).
On the other hand, an entrepreneur’s attitudes and motivations are also
shaped by personal perceptions of one’s own abilities and desirability of
the intended outcomes. In some cases, personal perceptions are incongruent
with what is socially desired and expected of an individual. When this is the
case, social pressure and social attitudes can influence the individual’s moti-
vation and goal setting and subsequently feelings about what one is doing
and being seen to be doing. For this reason, in order to understand how
individuals and entrepreneurs in particular develop their identity, it is
necessary to also understand why they do what they do and what drives
them to such behavior when it comes to emotions. Here, social pressure
and social expectations, and consequently the socioeconomic embeddedness
of the entrepreneur and his/her endeavor seem to play a vital role.
We have already argued and illustrated that the entrepreneurial identity
is partly cognitive (Krueger, 2007) where the entrepreneur develops self-
concepts about him or herself in the role and partly expressed in the entre-
preneur’s actions (Ashforth & Kreiner, 1999) where positive and negative
emotions have been shown to influence these (cf. Foo, 2011; Hayton &

Cholakova, 2012; Welpe et al., 2012 above). It follows that positive and
negative emotions influence the identity work of entrepreneurs.
In line with Barbalet (1996) positive emotion refers to when the entrepre-
neur perceives the situation as favorable and has a good feeling in general
and therefore aims higher. For example, hope is a positive emotion since it
reflects goal achievement, agency, and pathways to one’s goals (Henry,
2004; Snyder, 1995) with an “an overall perception that goals can be met”
(Henry, 2004, p. 385). Hope is also governed by a goal that is perceived as
possible to attain (Averill, Catlin, & Chon, 1990). Joy, happiness, and opti-
mism are examples of other positive emotions. So, for example, an entre-
preneur who relates positive events to internal and stable attributions and
negative events to external and unstable attributions is considered optimis-
tic (Seligman, 1998). We argue that entrepreneurs who experience positive
emotions in relation to their venture may more easily identify themselves as
On the other hand, negative emotions reflect the entrepreneur’s uncer-
tainty about his or her own capacity (Dequech, 2000). For example, fear is
regarded as a negative emotion where the individual interprets a situation as
a threat that may cause physical harm, loss, or failure and impact negatively
on personal well-being, sense of competence, or social position (Shaver,
Schwartz, Kirson, & O’Connor, 1987). Fear typically makes the individual
avoid the situation that provokes the emotion (Gooty, Gavin, & Ashkanasy,
2009). Another example of a negative emotion is sadness. This is experienced
as the aftermath to a negative situation and may lead to feelings of being
powerless or helpless (Shaver et al., 1987). An entrepreneur, who feels fear
or sadness, may thus withdraw from necessary entrepreneurial activities (cf.
Welpe et al., 2012) due to a belief that they are not competent in their role as
an entrepreneur. We argue that entrepreneurs who experience negative emo-
tions in relation to their venture may have harder times to identify them-
selves as entrepreneurs. See Exhibit 4 for an illustration.
To summarize, entrepreneurs are embedded in an emotional context
that gives rise to a range of emotions in the social interaction between the
entrepreneur and his/her environment. For example, if an entrepreneur
receives feedback about their performance that he or she perceives as posi-
tive such as customer satisfaction, they may relate this to goal achievement
and experience satisfaction and hope for the future. Should they receive
feedback about their performance that customers are not satisfied or that
their entrepreneurial idea does not hold, they may experience sadness or
fear. Since emotions are closely linked to cognition and thereby the mindset
of the entrepreneur, we argue that these emotions have an effect on their
Entrepreneurial Identification and Dis-Identification 231

Exhibit 4 Positive and Negative Emotions’ Influence on the

Entrepreneurial Identity.
Ruth and Harry have started a company together where they are
entering a relatively new market of luncheon vouchers in the British
northeast county of Cheshire. These vouchers are bought by employ-
ers and given to employees as a workplace benefit. The couple works
hard to establish themselves in the market by visiting lunch providers
and asking them to accept the vouchers and persuading employers to
buy into a scheme. While visiting pubs and restaurants Ruth receives
positive feedback on the business idea from some owners and quite
easily persuades owners to join the scheme. However, she also
receives a lot of negative feedback, especially from employers, who
believe that the English employees will not give up their traditional
lunch box in favor of restaurant vouchers. Ruth has difficulties in
taking in the negative feedback and it makes her sad and gives her
feelings of hopelessness in establishing herself as an entrepreneur.
Harry, on the other hand, is not greatly impacted by the negative
feedback; it just makes him more persistent, attributing the feedback
to employers’ lack of knowledge of the scheme and he is optimistic
about being able to educate the customer and to establish himself as
an entrepreneur.

identity work as entrepreneurs. Positive feedback or suggestions of positive

feedback give rise to positive emotions and may strengthen their role iden-
tity as entrepreneurs whereas negative feedback, or suggestions of negative
feedback, gives rise to negative emotions and may lessen their role identity
as entrepreneurs. This leads us to the following propositions:
Proposition 4a. Entrepreneurs receiving positive performance feedback
will experience positive emotions, which will be associated with higher
entrepreneurial identity salience.

Proposition 4b. Entrepreneurs receiving negative performance feedback

will experience negative emotions, which will be associated with lower
entrepreneurial identity salience.
However, drawing on the discussion above, entrepreneurs with a fixed
mindset are probably more vulnerable to negative performance feedback

since they are already in doubt about their entrepreneurial role. Therefore,
we argue that those with a fixed mindset would need more positive feed-
back in order to cultivate their entrepreneurial identity. On the other hand,
the entrepreneur with a growth mindset would either disregard such nega-
tive performance feedback or react with emotions that do not attribute
such feedback to their own role identity. We therefore propose:
Proposition 5a. The relationship between mindset and purposeful craft-
ing will be moderated by positive emotions elicited by positive perfor-
mance feedback and this relationship will be stronger for entrepreneurs
with a fixed versus growth mindset. Specifically, fixed mindset entrepre-
neurs experiencing positive emotions will show a greater increase in pur-
poseful crafting than their growth mindset counterparts.

Proposition 5b. The relationship between mindset and purposeful

crafting will be moderated by negative emotions elicited by negative
performance feedback and this relationship will be stronger for
entrepreneurs with a fixed versus growth mindset. Specifically, fixed
mindset entrepreneurs experiencing negative emotions will show a
greater decrease in purposeful crafting than their growth mindset

Although researchers agree that the entrepreneurial identity has an impor-
tant impact on the outcomes of entrepreneurs’ activity and that the content
of entrepreneurial identity may differ, emphasizing different roles in the
entrepreneurial process and different motivations for engaging in entrepre-
neurial activity (Cardon, Wincent, Singh, & Drnovsek, 2009; Fauchart &
Gruber, 2011), the triggers and the process of the emergence of and the
subsequent crafting of entrepreneurial identity have not received sufficient
attention and require further exploration. In this chapter, we argued that
emotions are an important trigger for the entrepreneurial identity. We
developed a theoretical framework that specifies the likely relationships
between the contextual features underpinning an individual’s entry into an
entrepreneurial role and the ensuing pathways for developing an entrepre-
neurial identity. In doing so, we contribute to new ways of analyzing the
role of emotions in organizational settings as well as provide suggestions
on how to approach studying emotions in understanding the emergence
Entrepreneurial Identification and Dis-Identification 233

and crafting of the entrepreneurial identity. Thus, our model contributes to

both emotion and entrepreneurship literature.

Theoretical Implications

While extant research has focused on establishing the content of entrepre-

neurial identity, for example, Cardon et al. (2009) link three entrepreneurial
roles to the content of entrepreneurial identity, and Fauchart and Gruber
(2011) point how different motivations influence the content of the identity,
we focus on other factors that influence the emergence of the entrepreneur-
ial identity. Further, we theorize that the relationship between mindset,
emotions, and entrepreneurial identity is not unidirectional but rather
bidirectional and that feedback loops occur that initiate change in each of
the factors (see Fig. 1).
Emotions are extremely important in understanding how individuals
function. We have provided a conceptual model of how emotions matter
for creating the entrepreneurial identity. Our main contribution lies in the
development of a model theorizing the interplay between context and emo-
tions in the process of crafting an entrepreneurial identity. Our theorization
provides new insights into the role of emotions in the identification process
(cf. Stets & Burke, 2000). Further, in line with previous emotion research,
the model indicates that positive feedback about the role identity leads to
positive emotions for the individual which reinforces the role identity
whereas negative feedback on the role identity leads to negative emotions
for the individual that decreases the role identity (cf. Stryker, 1987). To
this, we add that for entrepreneurs with a fixed mindset, emotions play an
even greater role since they are dependent on reinforced positive feedback
in order to experience positive emotions about creating their entrepreneur-
ial identity whereas the entrepreneurs with an attitude of growth would
continue to enhance their entrepreneurial identity without the same depen-
dence on positive feedback. Emotions thus provide an indicator of what
cycle the identification or the dis-identification cycle the individual
grows into or grows out of the entrepreneurial role. Viewed like this, our
model can be translated into the dis/identification process of other profes-
sions as well, where the role of emotions is significant. So, for example, an
individual who is crafting toward an identity of being a dentist may experi-
ence negative performance feedback from their clients, leading to emotions
of sadness, hopelessness, and the like and starting a process where the den-
tist feels that s/he is inadequate in the profession. A circular process of this

kind, especially including a fixed mindset of the individual, may eventually

lead to a dis-identification cycle.
In our model, we can also draw connections from being pulled or pushed
into the identity, and that emotions may play out differently in these two
cases. So for instance, an individual who is pulled into an identity as for
example an entrepreneur, a teacher, or another professional identity may
not only have greater chances to grow into the identity but also to perform
in a way so that the performance feedback is positive and therefore being
more prone to experience positive emotions. These positive emotions in turn
can reinforce the salience of the identity and it becomes a positive circular
process. On the other hand, being pushed into a professional identity leads
to fewer chances to grow into the identity, with a risk of more negative feed-
back with experienced negative emotions resulting in lessening the chance of
a salience identity leading into a negative circular process. Emotions in iden-
tity building thus not only play the role as positive and negative for the indi-
vidual but also as an important instigator for the identity creation.

Practical Implications

For entrepreneurs and other professionals our model suggests that being
aware of the instigation of an identity (pull or push) and a cognitive atti-
tude (growth or fixed), combined with the role of emotions and the
embeddedness is helpful as they establish themselves as entrepreneurs and
this awareness can contribute to emotional well-being. Research on entre-
preneurs has shown that life satisfaction is reached when the entrepreneur
evaluates his/her life as satisfaction as a whole, from a longer-term perspec-
tive (Diener, Kahneman, & Helliwell, 2010). Thus, well-being for the entre-
preneur, related to positive emotional experiences, a positive identity, and
meaningful work, will have implications for the entrepreneur’s family and
immediate environment.

Future Research Directions

The dynamic model of entrepreneurial dis/identification presented here lays

the theoretical foundations for testing hypotheses about the dis/identifica-
tion process of crafting an entrepreneurial identity. Quantitative studies
can operationalize the dimensions of pull and push factors, growth versus
fixed mindsets, and emotional experiences and actions from performance
feedback. This would validate the model.
Entrepreneurial Identification and Dis-Identification 235

We recognize that there are issues in the model that may not be easily
captured with a quantitative study. Emotions influence and are also influ-
enced by motivations and former experiences. How he or she feels about a
certain experience influences the entrepreneur’s willingness to be exposed to
similar experiences. Viewed this way, emotions can influence the direction
of future action.
On the other hand, an individual’s attitudes and motivations are also
being shaped by his or her perceptions of one’s own abilities and desirabil-
ity of the intended outcomes. This however, might sometimes be incongru-
ent with what is socially desired and expected of the entrepreneur. Insofar
as the social pressure and social attitudes influence the entrepreneur’s moti-
vation and goal setting, changes are likely to occur in feelings about own
actions and what is displayed to the world. From this it is clearly visible
that in order to understand how entrepreneurs in particular form their
identity, one needs to understand why they do what they do in one way or
the other and what drives them to such behavior. Here, the social pressure
and social expectations, and consequently the social embeddedness, play a
vital role. Therefore we believe that in-depth case studies of multiple entre-
preneurs and observation of the crafting process over a prolonged period
of time would be an appropriate approach for a more complete under-
standing of the model.

Concluding Remark

The basic premise of our chapter is that both mindset and emotions play
an important role in the emergence of and the subsequent crafting of the
entrepreneurial identity. The crafting of an identity is thus not unidirec-
tional but rather bidirectional process and that feedback loops occur that
initiate change in each of the factors. This means that it is the emotional
processes that influence the emergence of the entrepreneurial identity
toward an identification or dis-identification cycle that plays an important
role in entrepreneurial success.

Achtenhagen, L., & Johannisson, B. (2013). Games in entrepreneurship education to support
the crafting of an entrepreneurial mindset. In S. de Freitas, M. Ott, M. M. Poprescu, &
I. Stanescu (Eds.), New pedagogical approaches in game enhanced learning: Curriculum
integration (pp. 20 37). Hershey, PA: Idea Group Inc. (IGI).

Alvesson, M., & Willmott, H. (2002). Identity regulation as organizational control: Producing
the appropriate individual. Journal of Management Studies, 39(5), 619 644.
Amit, R., & Muller, E. (1995). “Push” and “pull” entrepreneurship. Journal of Small Business &
Entrepreneurship, 12(4), 64 80.
Ashforth, B. E. (2001). Role transitions in organizational life: An identity-based perspective.
Mahwah, NJ: Lawrence Erlbaum Associates.
Ashforth, B. E., & Kreiner, G. (1999). How can you do it?: Dirty work and the challenge of
constructing positive identity. Academy of Management Review, 24(3), 413 434.
Ashforth, B. E., & Saks, A. M. (1995). Work-role transitions: A longitudinal examination of
the Nicholson model. Journal of Occupational and Organizational Psychology, 68(2),
157 175.
Ashforth, B. E., & Sluss, D. M. (2008). How relational and organizational identification con-
verge: Processes and conditions. Organization Science, 19, 807 823.
Averill, J. R., Catlin, G., & Chon, K. K. (1990). Rules of hope. New York, NY: Springer.
Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of
Psychology, 52, 1 26.
Barbalet, J. M. (1996). Social emotions: Confidence, trust and loyalty. International Journal of
Sociology and Social Policy, 16(9 10), 75 96.
Blackwell, L. S., Trzesniewski, K. H., & Dweck, C. S. (2007). Implicit theories of intelligence
predict achievement across an adolescent transition: A longitudinal study and an inter-
vention. Child Development, 78, 246 263.
Brundin, E., Pazelt, H., & Shepherd, D. (2008). Managers’ emotional displays and employees’
willingness to act entrepreneurially. Journal of Business Venturing, 23(2), 221 243.
Cardon, M., Foo, M., Shepherd, D., & Wiklund, J. (2012). Exploring the heart:
Entrepreneurial emotion is a hot topic. Entrepreneurship Theory and Practice, 36(1),
69 96.
Cardon, M. S., Zietsma, C., Saparito, P., Matherne, B. P., & Davis, C. (2005). A tale of
passion: New insights into entrepreneurship from a parenthood metaphor. Journal of
Business Venturing, 20(1), 23 45.
Cardon, M., Wincent, J., Singh, J., & Drnovsek, M. (2009). The nature and experience of
entrepreneurial passion. Academy of Management Review, 34(3), 511 532.
Clarke, J., & Holt, R. (2010). The mature entrepreneur: A narrative approach to entrepreneur-
ial goals. Journal of Management Inquiry, 19(1), 69 83.
Costas, J., & Fleming, P. (2009). Beyond dis-identification: A discursive approach to self-
alienation in contemporary organizations. Human Relations, 62, 353 378.
Dequech, D. (2000). Confidence and action: A comment on Barbalet. Journal of Socio-
Economics, 29, 503 515.
DeRue, D. S., & Ashford, S. J. (2010). Who will lead and who will follow? A social process of
leadership identity construction in organizations. Academy of Management Review,
35(4), 627 647.
Diener, E., Kahneman, D., & Helliwell, J. (2010). International differences in well-being.
Oxford: Oxford University Press.
Down, S. (2006). Narratives of enterprise. Crafting entrepreneurial self-identity in a small firm.
Cheltenham: Edward Elgar.
Down, S., & Reveley, J. (2004). Generational encounters and the social formation of entrepre-
neurial identity. Organization, 11(2), 233 250.
Entrepreneurial Identification and Dis-Identification 237

Down, S., & Warren, L. (2008). Constructing narratives of enterprise: Clichés and entrepre-
neurial self-identity. International Journal of Entrepreneurial Behaviour and Research,
14(1), 4 23.
Dubini, P., & Aldrich, H. (1991). Personal and extended networks are central to the entrepre-
neurial process. Journal of Business Venturing, 6(5), 305 313.
Dweck, C. (2010). Mind-sets and equitable education. Principal Leadership, 10(5), 26 29.
Dweck, C. S. (1999). Self-theories: Their role in motivation, personality, and development.
Philadelphia: Psychology Press.
Dweck, C. S., Chiu, C., & Hong, Y. Y. (1995). Implicit theories and their role in judgments
and reactions: A word from two perspectives. Psychological Inquiry, 6(4), 267 285.
Elliot, A. J., Eder, A. B., & Harmon-Jones, E. (2013). Approach and avoidance motivation
and emotion: Convergence and divergence. Emotion Review, 5, 308 311.
Farmer, S., Yao, X., & Kung-Mcintyre, K. (2011). The behavioral impact of entrepreneur
identity and prior entrepreneurial experience. Entrepreneurship Theory and Practice,
35(2), 245 273.
Fauchart, E., & Gruber, M. (2011). Darwinians, communitarians and missionaries: The role of
founder identity in entrepreneurship. Academy of Management Journal, 54(5), 935 957.
Foo, M. D. (2011). Emotions and entrepreneurial opportunity evaluation. Entrepreneurship
Theory and Practice, 35(2), 375 393.
Goffman, E. (1959). The presentation of self in everyday life. Garden City, NY: Doubleday
Gollwitzer, P. M., & Kirchhof, O. (1998). The willful pursuit of identity. In J. H. C. S. Dweck
(Ed.), Motivation and self-regulation across the life span (pp. 389 423). New York, NY:
Cambridge University Press.
Gooty, J., Gavin, M., & Ashkanasy, N. M. (2009). Emotions research in OB: The challenges
that lie ahead. Journal of Organizational Behavior, 30(6), 833 838.
Härtel, C. E. J. (2008). How to build a healthy emotional culture and avoid a toxic culture. In
C. L. Cooper & N. M. Ashkanasy (Eds.), Research companion to emotion in organiza-
tion (pp. 575 588). Cheltenham, UK: Edwin Elgar Publishing.
Hayton, J. C., & Cholakova, M. (2012). The role of affect in the creation and
intentional pursuit of entrepreneurial ideas. Entrepreneurship Theory and Practice,
36(1), 41 68.
Henry, P. (2004). Hope, hopelessness, and coping: A framework for class-distinctive cognitive
capital. Psychology and Marketing, 21(5), 375 403.
Hoang, H., & Gimeno, J. (2010). Becoming a founder: How founder role identity affects entre-
preneurial transitions and persistence in founding. Journal of Business Venturing, 25(1),
41 53.
Ibarra, H., & Barbulescu, R. (2010). Identity as narrative: Prevalence, effectiveness, and conse-
quences of narrative identity work in macro work role transitions. Academy of
Management Review, 35(1), 135 154.
Jain, S., George, G., & Maltarich, M. (2009). Academics or entrepreneurs? Investigating role
identity modification of university scientists involved in commercialization activity.
Research Policy, 38, 922 935.
Jenkins, A. S., Wiklund, J., & Brundin, E. (2014). Individual responses to firm failure:
Appraisals, grief, and the influence of prior failure experience. Journal of Business
Venturing, 29(1), 17 33.

Kärreman, D., & M. Alvesson (2001). Making newsmakers: Conversational identity at work.
Organization Studies, 22(1), 59.
Kazimierczuk, A. (2015). 31- letnia Olga Malinkiewicz zrewolucjonizujeświat. Retrieved from
www.patriot24.net. Accessed on January 27.
Kira, M., & Balkin, D. B. (2014). Interactions between work and identities: Thriving, withering,
or redefining the self? Human Resource Management Review, 24(2), 131 143.
Kirkwood, J. (2009). Motivational factors in a push-pull theory of entrepreneurship. Gender in
Management: An International Journal, 24(5), 346 364.
Krueger, N. F. (2007). What lies beneath? The experiential essence of entrepreneurial thinking.
Entrepreneurship Theory and Practice, 31(1), 123 138.
Lave, J., & E. Wenger (1991). Situated learning: Legitimate peripheral participation.
Cambridge: Cambridge University Press.
Lounsbury, M., & Glynn, M. A. (2001). Cultural entrepreneurship: Stories, legitimacy, and
the acquisition of resources. Strategic Management Journal, 22(6/7), 545 564.
Markman, G. D., Balkin, D. B., & Baron, R. A. (2002). Inventors and new venture formation:
the effects of general self-efficacy and regretful thinking. Entrepreneurship Theory and
Practice, 27(2), 149 165.
McAdams, D. P. (1996). Personality, modernity, and the storied self: A contemporary frame-
work for studying persons. Psychological Inquiry, 7(4), 295 321.
McGrath, R. G., & MacMillan, I. C. (2000). The entrepreneurial mindset: Strategies for con-
tinuously creating opportunity in an age of uncertainty. Boston, MA: Harvard Business
Powell, E. E., & Baker, T. (forthcoming). It’s what you make of it: Founder identity and
enacting strategic responses to adversity. Academy of Management Journal.
Salancik, G., & Pfeffer, J. (1978). A social information processing approach to job attitudes
and task design. Administrative Science Quarterly, 23, 224 253.
Schein, E. H. (1978). Career dynamics. Reading, MA: Addison-Wesley.
Schjoedt, L., & Shaver, K. G. (2007). Deciding on an entrepreneurial career: A test of the pull
and push hypotheses using the panel study of entrepreneurial dynamics data1.
Entrepreneurship Theory and Practice, 31(5), 733 752.
Seligman, M. E. P. (1998). Learned optimism. New York, NY: Pocket Books.
Shaver, P., Schwartz, J., Kirson, D., & O’Connor, C. (1987). Emotion knowledge. Further
exploration of a prototype approach. Journal of Personality and Social Psychology, 52,
1061 1086.
Shepherd, D. A. (2003). Learning from business failure: Propositions of grief recovery for the
self-employed. Academy of Management Review, 28(2), 318 328.
Shepherd, D., & Haynie, J. M. (2009a). Family business, identity conflict, and an
expedited entrepreneurial process: A process of resolving identity conflict.
Entrepreneurship Theory and Practice, 33(6), 1245 1264. doi:10.1111/j.1540-6520.
Shepherd, D., & Haynie, J. M. (2009b). Birds of a feather don’t always flock together: Identity
management in entrepreneurship. Journal of Business Venturing, 24(4), 316 337.
Simpson, R., Irvine, K., Balta, M., & Dickson, K. (2015). Emotions, performance and entre-
preneurship in the context of fringe theatre. Organization, 22(1), 100 118.
Snyder, C. R. (1995). Conceptualizing, measuring, and nurturing hope. Journal of Counseling
and Development, 73(3), 355 360.
Entrepreneurial Identification and Dis-Identification 239

Stets, J. E., & Burke, P. J. (2000). Identity theory and social identity theory. Social Psychology
Quarterly, 63(3) 224 237.
Stryker, S. (1980). Symbolic interactionism: A social structural version. Menlo Park, CA:
Benjamin-Cummings Publishing Company.
Stryker, S. (1987). The interplay of affect and identity: Exploring the relationships of social
structure, social interaction, self, and emotion. Presented at the meetings of the
American Sociological Association, August, Chicago.
Thornborrow, T., & Brown, A. (2009). Being regimented: Aspiration, discipline and identity
work in the British Parachute Regiment. Organization Studies, 30(4), 355 376.
Van Maanen, J. (1998). Qualitative studies of organizations. London: Sage.
Villanueva, J., & Markowska, M. (2013). Storytelling in the development of entrepreneurial
identity. 5th Conference on Rhetoric and Narratives, ESADE, Barcelona.
Warren, L. (2004). Negotiating entrepreneurial identity. Communities of practice and
changing discourses. International Journal of Entrepreneurship and Innovation, 5(2),
25 37.
Welpe, I. M., Spörrle, M., Grichnik, D., Michl, T., & Audretsch, D. B. (2012). Emotions and
opportunities: The interplay of opportunity evaluation, fear, joy, and anger as antece-
dent of entrepreneurial exploitation. Entrepreneurship Theory and Practice, 36(1),
69 96.
Welter, F. (2011). Contextualising entrepreneurship Conceptual challenges and ways for-
ward. Entrepreneurship Theory and Practice, 35(1), 165 184.
Wry, T., Lounsbury, M., & Glynn, M. A. (2011). Legitimating nascent collective identities:
Coordinating cultural entrepreneurship. Organization Science, 22(2), 449 463.
Ybema, S., Keenoy, T., Oswick, C., Beverungen, A., Ellis, N., & Sabelis, I. (2009).
Articulating identities. Human Relations, 62(3), 299 322.
This page intentionally left blank


Sandra Kiffin-Petersen


Work design has largely overlooked cognitive emotional interactions in

understanding employee motivation and satisfaction. My aim in this
chapter is to develop a conceptual model that integrates what we know
about these interactions from research on emotions and neuroscience
with traditional and emergent work design perspectives. I propose that
striving for universal goals influences how a person responds to the work
characteristics, such that an event that is personally relevant or “self-
referential” will elicit an emotional reaction that must be regulated for
optimal performance, job satisfaction, and well-being. A Self-Referential
Emotion Regulatory Model (SERM) of work design is presented.
Keywords: Work design; affective events; neuroscience

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 241 269
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011010


“present-day abstraction of work has shut out feelings from the job content.” (Herzberg,
1987, p. 120)

Work design has been defined as “the content and organization of one’s
work tasks, activities, relationships, and responsibilities” (Parker, 2014,
p. 662). Prior research has demonstrated that work design is critically
important to achieving key employee outcomes including improved perfor-
mance, motivation and job satisfaction, and reduced stress and burnout
(Grant & Parker, 2009). However, we know little about how cognitions and
emotions interact to influence these outcomes (Saavedra & Kwun, 2000;
Seo, Barrett, & Bartunek, 2004). Cognition generally refers to thought pro-
cesses such as memory, attention, planning, and problem solving in humans
(Pessoa, 2008). Emotions on the other hand are subjective experiences that
are of shorter duration, greater focus and intensity than moods or affect,
have a clear cause, person or object, and importantly, create a state of
action readiness (Frijda, 1988, 1994). There is reason to believe based on
research into emotions in organizations (e.g., Affective Events Theory,
AET: Weiss & Cropanzano, 1996) and neuroscience (Pessoa, 2008) that
interactions between cognitions and emotions are influential in how work
characteristics affect important individual outcomes.
To address this evident research gap, I develop a theoretical model pro-
posing that desirable outcomes, including increased motivation, job satis-
faction, and aspects of performance, are influenced by a self-regulatory
process involving the pleasantness of emotions evoked in response to, and
shaped by social contextual factors. A focal point of new research into
emotions in organizations should therefore, be on understanding how to
combine tasks, activities, and relationships (i.e., work design) in such a way
as to promote pleasant emotions and meaningful work (Parker, 2014).
Hedonic tone or the pleasantness of an emotion, and intensity or the degree
of activation, is often used to differentiate between discrete emotions
(Larsen & Diener, 1992; Russell, 1980). Emotions with a pleasant hedonic
tone such as joy, interest, and pride are often grouped together as positive
emotions, in contrast to those commonly labeled as negative or unpleasant
emotions (e.g., frustration, anger). In this chapter, I also use the term
“work design” over job design because it encompasses the more proactive
or self-initiating aspects of employees’ work, as well as the technical tasks
that are more typically included in a job description (Morgeson &
Campion, 2003; Parker, 2014).
Model of Work Design 243

Examining the role of emotions in work design is temporally relevant

for two reasons. First, accumulating evidence supporting the stress buffer-
ing effects of positive emotions, particularly in light of increased stress at
work (Cavanaugh, Boswell, Roehling, & Boudreau, 2000) and potentially
“mindless” tasks associated with excessive workloads (Elsbach &
Hargadon, 2006), cannot be ignored (Fredrickson, 1998, 2001, 2013a;
Vacharkulksemsuk, & Fredrickson, 2013). Positive emotions also help pro-
mote creativity and greater flexibility in problem solving, as well as having
a positive effect on intrinsic motivation and performance even when the
task is complex and difficult (Isen, 2000; Isen & Reeve, 2005). Meta-
analytic findings have also found happier individuals’ are more likely to
experience job satisfaction, fulfilling relationships, superior work perfor-
mance, and improved general health (Lyubomirsky, King, & Diener, 2005).
Frequent, long-lasting positive emotions are thought to be more influential
on long-term happiness, than infrequent high-intensity positive emotions
(Diener, Sandvik, & Pavot, 1991).
The second oft cited reason emotion should be central to work design is
because changes in the way work is organized means previous models and
theories do not necessarily reflect current realities (Grant & Parker, 2009;
Humphrey, Nahrgang, & Morgeson, 2007). There is a need to better under-
stand emotional and interpersonal tasks in service jobs, given their preva-
lence (Grandey & Diamond, 2010; Grant, Fried, Parker, & Frese, 2010;
Oldham & Hackman, 2010) and adverse effect on employees’ emotional
exhaustion (Bono & Vey, 2005; Hülsheger, & Schewe, 2011; Mesmer-
Magnus, DeChurch, & Wax, 2012; Wang, Seibert, & Boles, 2011). Ideally,
new research should go beyond motivational models of work design to
incorporate approaches that also seek to enhance the physical and mental
health of employees (Parker, 2014). My examination of the role of emo-
tions in work design in organizational settings is therefore, timely.
My purpose in this chapter is twofold. First, I aim to reconcile divergent
research streams to present some broad propositions as a first step toward
clarifying the role of emotion in work design. Our emotional reactions pro-
vide us with critical information and feedback about our behavior and
interactions with others, including colleagues and the public that I argue
has been insufficiently addressed in mainstream work design research.
Second, I seek to refine our understanding of the relationship between emo-
tions and work design by drawing on recent findings from neuroscience
(Becker, Cropanzano, & Sanfey, 2011; Cropanzano, & Becker, 2013).
Neuroscience recognizes that human actions are the result of cognitive and
emotional pathways in the brain interacting in complex ways (Lindquist,

Wager, Kober, Bliss-Moreau, & Feldman Barrett, 2012; Ochsner & Gross,
2008; Pessoa, 2008) and hence, such research may explain “how the brain
and its messengers (the neurotransmitters) respond positively to work”
(Nicholson, 2010, p. 425). Management (Waytz & Mason, 2013), leadership
development (Boyatzis, Smith, & Blaize, 2006; Waldman, Balthazard, &
Peterson, 2011), and behavioral economics (Camerer, Loewenstein, &
Prelec, 2005) have all drawn from neuroscience research methods.
The conceptual model I propose is different from traditional motiva-
tional theories of work design for several reasons. First, it is embedded
within an AET framework (Weiss & Cropanzano, 1996) whilst also draw-
ing on theories of cognitive appraisal (e.g., Lazarus, 1991; Roseman,
Spindel, & José, 1990) and emotion regulation (Gross, 1998, 2013), to
afford a more central role for cognitive emotional interactions in work
design. In line with AET, emotion regulation processes determine, in part,
how perceived work characteristics translate into emotional reactions and
work attitudes, such as job satisfaction. Emotion regulation is conceptua-
lized as a proactive internal process directed at regulating either the magni-
tude or duration of an emotional response during work activities, tasks,
and relationships (Gross, Sheppes, & Urry, 2011). Positive emotions can
become self-energizing and self-sustaining via an upward spiral, as well
as counteracting negative downward spirals (Garland et al., 2010). Second,
I draw on recent findings from neuroscience and motivational theories to
highlight how the self-referential nature of an event (i.e., its importance to
the individual’s goals, motives, or concerns, Frijda, 1994) is critically
important in evoking an individual’s emotional response. My approach
therefore departs somewhat from traditional models of work design in that
I assert that the pleasantness of emotional experiences is an integral influ-
ence, parallel to key psychological states (Barrick, Mount, & Li, 2013;
Hackman & Oldham, 1980), on key individual outcomes.


The Self-Referential Emotion Regulatory Model (SERM) of work design

I propose is presented in Fig. 1. I initially discuss how AET and neu-
roscience provide an overarching framework from which we can under-
stand how cognitions and emotions interact to influence important work
design outcomes. I then elaborate on each of the main components of
Model of Work Design
Self-Referential Goals Cognitive-Emotional Emotion Regulation Individual Outcomes
(communion, status, Interaction (psychological, behavioral,
autonomy, achievement) (event-appraisal relationship) attitudinal)

Work Characteristics Pleasant Emotions
P1 P7a Organizational
Task identity Happiness
citizenship behaviors
Task significance Pride
P2 Increased motivation
Skill variety Excitement
Emotional Job Demands P3
Duration Affective Events Cognitive Appraisal
Frequency appraisal-reappraisal P7b
Emotional exhaustion
Unpleasant Emotions Counterproductive work
Social Contextual Factors P4 Boredom P8 behaviors
Peer job evaluation Frustration Absenteeism
Co-worker emotions Anger Work stress
Social support


Fig. 1. Self-Referential Emotion Regulatory Model of Work Design.


Fig. 1, beginning with a review of the major motivational approaches to

work design to establish the self-referential nature of work goals as criti-
cally important. Next, I consider the cognitive emotional interaction
involved in the event appraisal relationship. Finally, I explore the role of
emotion regulation in managing intrapersonal fluctuations in emotions
together with emergent perspectives of work design that establish the per-
son’s emotional state can be an “energizing” or “exhausting” pathway
through which motivation and key individual outcomes are affected.
Throughout this chapter, I draw on research in neuroscience to inform my
arguments and to demonstrate how such studies complement our current
methods of studying emotions in organizations, with a particular focus on
work design.

AET, Neuroscience, and Work Design

AET is directly applicable to work design (Fisher & Ashkanasy, 2000;

Parker, Wall, & Cordery, 2001; Saavedra & Kwun, 2000), because it offers
a framework for understanding how stable work characteristics (e.g., skill
variety, task identity, task significance, feedback, autonomy: Hackman &
Oldham, 1980) lead to emotional reactions, which in turn, impact on work
attitudes (e.g., job satisfaction) and behavior (e.g., task performance)
(Weiss & Cropanzano, 1996). Employees’ emotional experiences influence
their judgment, but they are not the same as job satisfaction. Rather, emo-
tions act as a potential mediating mechanism between work characteristics,
and job attitudes and behavior (Fisher & Ashkanasy, 2000; Weiss &
Cropanzano, 1996). Weiss and Cropanzano stress the need therefore, to
study employees’ emotional reactions separately to job satisfaction.
According to AET, stable work characteristics do not predict changes in
emotional states, but rather, work features make more or less likely the
occurrence of affective events or incidents during a working day (Weiss &
Beal, 2005). An affective event is defined as “an incident that stimulates
appraisal of and emotional reaction to a transitory or ongoing job-related
agent, object or event” (Basch & Fisher, 2000, p. 37). Cognitive appraisal
theory (Lazarus, 1966; Roseman, Spindel, & José, 1990) offers an explana-
tion as to how particular workplace events elicit a positive or negative emo-
tional reaction (Lazarus, 1966). In a broad sense, events that facilitate goal
achievement and lead to the fulfillment of employees’ interests elicit posi-
tive emotions, such as happiness, pride, and excitement. Conversely, events
or situations that obstruct goal achievement and are perceived as
Model of Work Design 247

threatening or harmful can lead to negative emotions in employees, such as

frustration, boredom, or anger. Cognitive appraisal theory therefore is
important in explaining how and why the same event can elicit a different
emotional reaction from employees. AET therefore provides a macro psy-
chological framework for understanding one way in which cognitive and
emotional processes interact to jointly influence employees’ job satisfaction
and behavior.
Studies in neuroscience can discern additional insights about cognitive
emotional interactions relevant to work design, because they do not rely
solely on self-report measures as the only source of information about emo-
tions (Davidson, Pizzagalli, Nitschke, & Kalin, 2003). There is increased
interest in the use of neuroscience to study brain activity largely due to the
emergence of new, readily available methods, including functional magnetic
resonance imaging (fMRI) and instruments that measure brain activity and
other physiological responses, such as quantitative electroencephalocardio-
gram (qEEG) (Waldman et al., 2011). These tools help to mitigate pro-
blems associated with self-report measures, while at the same time they can
help explain complex relationships between the mind, physiological
responses, and behavior (Harmon-Jones & van Honk, 2012). Since human
behavior is an aggregate function of complex dynamic cognitive emotional
interactions of brain areas (Pessoa, 2008), it is potentially problematic to
conceptualize these two processes as separate systems. Complex behaviors
will likely involve multiple brain systems working in conjunction with each
other, as the person appraises and reappraises their tasks and relationships
and how they feel about performing them in a particular context. Indeed,
Pessoa (2008, p. 151) has argued that in the brain “emotion and motivation
are closely linked as both depend on the relationship between the organism
and its environment.”
Neuroscience and mainstream approaches to work design also share a
common focus on the importance of human goals and self-regulation. Both
evolutionary psychology (Nicholson, 2010) and work psychology (Parker,
2014) elevate human goals and self-regulation to a prominent role in their
theorizing. And yet, emotions per se are yet to be given center stage in
work design models. The nature or type of goals and the strategies to
achieve them are ordered by human organisms based on the essential pay-
offs they anticipate receiving as a result of achieving them. This reflects a
temporal dimension of both time and place to goal importance. It also
affords an element of self-determination essential to the survival and adap-
tive functions of humans. Research using event-related fMRI has shown
that the anticipation of rewards alone is sufficient to elicit increased

self-reported happiness and to activate the nucleus accumbens, whereas

anticipated punishments by itself evoked neither of these reactions
(Knutson, Adams, Fong, & Hommer, 2001). This finding opens up the
possibility that an anticipated reward may be sufficient stimulus to initiate
an upward emotion spiral.
Neuroscience’s focus on the holistic and adaptive nature of the human
condition should also be an important consideration in new emerging per-
spectives on work design. Emotions are critical to an individual’s survival
in modern organizations because they signal where and what the person
should focus their attention on (Compton, 2003). For example, using
fMRI, Fossati et al. (2003) have shown that when subjects made evaluative
judgments about positive and negative emotional words they thought
applied to themselves there was an increase in the right dorsomedial pre-
frontal cortex associated with this self-related emotional processing, com-
pared to if they believed the words related to others. This effect occurred
regardless of the valence of the emotion, indicating that the “self” is an
important point of reference in activating an emotional response (i.e., self-
Advances in neuroscience have shown that a more complex relationship
exists between cognition and emotion in humans, whereby many more
parts of the brain are involved in the processing of emotion than previously
thought (Dalgleish, Dunn, & Mobbs, 2009). Because of this close connec-
tion (Izard, 2009), Pessoa (2008) has questioned whether the classical dis-
tinction between cognition and emotion within psychology and
neuroscience should be retained. The difficulty confronting organizational
researchers however, is that when psychological theories of emotion are
integrated with the findings from affective neuroscience they suggest emo-
tion may be generated through multiple routes for example, AET and
the cognitive appraisal theory suggest that emotions can be derived from
primary appraisals of current events or situation but brain imaging studies
show that there are also simultaneous “automatic” or primal conditioned
responses that generate emotions (Dalgleish et al., 2009; Izard, 1993). For
example, Waugh, Wager, Fredrickson, Noll, and Taylor (2008) have shown
using fMRI that low-resilient people under threat exhibit greater anterior
insula activation in anticipation of the threat and take longer to recover
than highly resilient individuals. Negative prior expectations can influence
subsequent emotional responses, even for supposedly neutral stimuli.
Conclusions from neuroscience suggest at the very least, that we now
need to move beyond “cognition” versus “emotion” to recognize cognition
and emotion jointly and equally contribute to complex behaviors, such as
Model of Work Design 249

increased satisfaction, motivation, and performance from work design

interventions. “Two-factor models in which cognitive and affective pro-
cesses engage in a tug-of-war for the control of behavior” are too simplistic
(Ochsner & Gross, 2008, p. 157). Emotions signal what is important to the
individual, a factor that is often overlooked within mainstream work design
research. Given our enhanced understanding of cognitive emotional inter-
actions through AET and neuroscience it is important to now carefully
consider traditional motivational theories of work design. Motivational
approaches generally suggest, consistent with Frijda’s (1988) Law of
Concern, that how employees think about their work is determined by their
underlying goals, motives, or concerns, a concept I refer to herewith as self-
referential goals.

Self-Referential Goals: Motivational Approaches to Work Design

Motivational approaches to work design typically encompass, inter alia,

scientific management principles (Taylor, 1911), Sociotechnical Systems
(STS) Theory (Trist & Bamforth, 1951), Motivator-Hygiene Theory
(Herzberg, 1966; Herzberg, Mausner, & Snyderman, 1959), Job
Characteristics Model (Hackman & Oldham, 1980), Demand-Control
Model (DCM) (Karasek, 1979), and the Social Information Processing per-
spective (Salancik & Pfeffer, 1978). It is not my intention here to present a
detailed review of these theories, since these have previously been exten-
sively examined by well-known scholars in the field of work design
(e.g., see Ambrose & Kulik, 1999; Morgeson & Campion, 2003; Parker,
2014; Parker et al., 2001). Rather, my discussion focuses on how these
approaches conceptualize the role of emotion in understanding motivation
and work design outcomes.
Early industrial psychologists became concerned with work design fol-
lowing the large-scale adoption of scientific management principles
(Taylor, 1911) that typically lead to improved efficiencies and performance
at the expense of interesting and meaningful jobs. Breaking work down
into discrete tasks reduced the requirements for training and the time
involved in changing from one task to another. This generally led to
improved efficiencies as workers became extremely well skilled at perform-
ing a specialized task. However, scientific management principles were later
criticized for demotivating workers who were required to endlessly perform
a narrow set of tasks. Nicholson (2010) highlighted how important the per-
sonal relevance of the work is when he described an interview with a

woman machinist who found enjoyment in earning a regular wage using

her well-practiced skills in the company of her peers. The Law of Concern
(Frijda, 1988) potentially explains why the woman still enjoyed her work,
despite it being repetitive.
The STS theory (Trist & Bamforth, 1951), which was originally devel-
oped at The Tavistock Institute of Human Relations based on studies of
English coalminers, rests on two fundamental principles. The first is that
successful performance depends on the interaction between social and tech-
nical factors; and the second therefore, that optimization of the system
depends on both these processes. Whilst STS initiatives have most often
resulted in autonomous work groups (sic., self-managing or self-directed
teams) (Cummings, 1978), the theory reinforces the notion that employee
well-being is dependent on the interaction of their personal factors with the
task environment.
The Motivator-Hygiene Theory (Herzberg, 1966; Herzberg et al., 1959)
was developed from engineers and accountants descriptions of “sequences
of events” that lead them to feel particularly good or bad about their
work. Motivators were found to be important for job satisfaction (i.e.,
achievement, recognition, work itself, responsibility, and advancement) and
dissatisfiers or hygiene factors to unhappiness or dissatisfaction (i.e., salary,
supervision, and working conditions). A major contribution of Herzberg
and colleagues is that they distinguished between events that cause positive
and negative emotions in employees (Weiss & Cropanzano, 1996), a tenet
consistent with an asymmetrical view of positive and negative emotions
(Dasborough, 2006; Gable, Reis, & Elliot, 2000). Motivator-hygiene theory
has increased in acceptance (Ambrose & Kulik, 1999; Sachau, 2007)
because it recognizes employees must engage emotionally with their work
as well as intellectually, and that positive emotions are important for
employee well-being (Sachau, 2007).
The Job Characteristics Model (JCM) (Hackman & Oldham, 1980) is
one of the most well-known and cited theories of job design. The major
premise underpinning this theory is that the core job characteristics of skill
variety, task significance, task identity, autonomy, and feedback directly
influence employees’ psychological states, which, in turn, influences their
job satisfaction and motivation, and to a lesser degree, performance
(Parker & Wall, 1998). When certain characteristics are present, the job is
expected to have greater motivating potential, which has a positive effect
on the critical psychological states of experienced meaningfulness, responsi-
bility, and knowledge of results, which in turn, have a positive effect
on job satisfaction, intrinsic motivation, and task performance. Recent
Model of Work Design 251

meta-analytic results found experienced meaningfulness to be the most

important of the three psychological states in mediating between the core
job dimensions and outcomes (Humphrey et al., 2007).
Hackman and Oldham (1980, p. 60) originally suggested that “job-
holders will experience a positive, self-generated affective ‘kick’ when
they perform well and this internal reinforcement serves as an incentive
for continued good performance.” High levels of the five core job charac-
teristics should therefore make particular affective events more or less
likely to occur. To better understand this relationship, Saavedra and
Kwun (2000) investigated managers’ perceptions of the five core job
characteristics and their affective experiences during a normal working
week. The authors found that task significance, autonomy, and feedback
accounted for 18% of the variance in enthusiasm; whilst task significance
and autonomy were associated with reduced fatigue, and task feedback
with increased relaxation. They concluded that certain work characteris-
tics serve to “energize, reinforce and maintain work behavior” (p. 143).
In contrast, skill variety was associated with increased nervousness and
reduced relaxation, suggesting that too much variety may be stressful for
some employees. Employees’ cognitive appraisal of their work was there-
fore differentially related to their affective experiences, suggesting that
individuals may interpret work characteristics differently from that
originally proposed by Hackman and Oldham. Humphrey (2000) pro-
vided a convincing argument that the five job dimensions may also have
an effect on emotions experienced during service jobs, a proposition
requiring further research.
The Theory of Purposeful Behavior developed by Barrick et al. (2013)
explains how personality traits interact with the JCM core job dimensions
to influence satisfaction, performance, citizenship, counterproductive work
behaviors, and withdrawal. A key proposal in the model is that there are
two central self-regulatory cognitive processes involved including purpose-
fulness, or goal directedness to the person’s behavior, and experienced
meaningfulness, or the meaning the person derives from engaging in work
activities. Distilled from previous research, including the self-determination
theory (Deci & Ryan, 1985), Barrick et al. identified four higher order fun-
damental goals that employees are motivated to strive to meet in their
work including communion, status, autonomy, and achievement. Striving
to meet these goals should also therefore be significant in determining the
individual’s emotional reaction to events because the goals establish how
self-referential work tasks, activities, and relationships are to the individual.
Ohly and Schmitt (2015) drew similar conclusions to Barrick et al. in noting

the importance of overarching goals or values as a metainfluence on

employees’ appraisal of events since they provide important clues as to
expected payoffs anticipated from the work.
The importance of fundamental self-referential goals in understanding
motivation is consistent with findings in affective neuroscience that particu-
lar brain areas are activated due to the significance of the stimulus with
respect to the person’s goals and needs at work (Ochsner & Gross, 2008). I
include Barrick et al.’s (2013) four fundamental work goals in Fig. 1, to
denote the self-referential importance of these goals for emotion activation
pathways. The four goals are communion or relatedness, status striving or
getting ahead, autonomy or discretion, and achievement. I also acknowl-
edge here that personality traits (Barrick et al., 2013; Weiss & Cropanzano,
1996) are also a potential influence on the occurrence of affective events
through goal striving, as shown by the dashed lines in Fig. 1. However,
given my focus in this chapter is on central interactions between cognitions
and emotional states rather than traits, an extensive review of the influence
of personality is beyond its scope but is important for future research. This
leads to my first two broad propositions:
Proposition 1. Fundamental work goals of communion, status, auton-
omy, and achievement that are self-referential to the individual will
result in them reporting more affective events associated with striving to
meet that particular goal.

Proposition 2. High work characteristics of skill variety, skill identity,

task significance, autonomy, and feedback will result in the individual
reporting more positive affective events, than if these characteristics
were low.
Parker et al. (2001) argue in their Elaborated Job Characteristics model
that the traditional work characteristics in work design research are insuffi-
cient for modern workplaces and that some characteristics, such as emo-
tional demands, are more salient to some contexts. The significant growth
in service sector employment has highlighted the importance of the emo-
tional demands of work, and concomitant emotional labor. Hochschild
(1983) argued in The Managed Heart, that jobs higher in emotional
demands (e.g., airline attendants, debt collector) require a form of emotion
regulation (Gross, 1998) that she termed “emotional labor” that is deleter-
ious to the employee’s health. Emotional labor, according to Hochschild
(1983, p. 7), involves the “management of feeling to create a publicly obser-
vable facial or bodily display.” Several meta-analyses have concluded that
Model of Work Design 253

frequent usage of certain types of emotional labor (i.e., surface acting) is

associated with increased emotional exhaustion (i.e., Bono & Vey, 2005;
Hülsheger, & Schewe, 2011; Mesmer-Magnus et al., 2012; Wang et al.,
2011). Researchers have shown particular interest in studying whether
autonomy can mitigate some of the negative effects of emotional labor
(e.g., Grandey, 2000; Johnson & Spector, 2007). Emotional job demands
have typically been conceptualized and measured in service jobs by asses-
sing the duration of the interactions, frequency, intensity, and variety of
emotional displays with customers (Brotheridge & Lee, 2003; Grandey,
2000, 2003; Morris & Feldman, 1996, 1997). However, emotional job
demands do not only apply to organizational outsiders but also have been
shown to occur in interactions with insiders including coworkers and super-
visors (Grandey, Kern, & Frone, 2007). Given my focus on the role of emo-
tion in work design it seems prudent therefore, to include emotional job
demands along with work characteristics in Fig. 1 (Parker et al., 2001).
Jobs that are high in these core emotional demands may result in more
negative affective events, which lead to my third proposition:
Proposition 3. High emotional job demands (i.e., duration, frequency,
intensity, variety of emotions) involved in interactions with organiza-
tional insiders and outsiders (e.g., supervisors, coworkers, customers)
will result in the individual reporting more negative affective events,
than if these demands were low.
The Social Information Processing perspective (Salancik & Pfeffer,
1978) of work design challenged a core tenet of the JCM by suggesting that
employees’ perceptions of job characteristics are shaped by how they feel
about their work and also the influence of their peers’ evaluation of the
work. Salancik and Pfeffer stressed that the context influences how features
of the work impact on employees’ perceptions, a notion echoed in recent
reviews of work design (i.e., Morgeson & Humphrey, 2008; Nicholson,
2010). The words used by others can have a powerful influence on how an
employee perceives their job and hence, their resultant satisfaction. The
emotions of others can also influence a person’s perception, and hence,
how they think and feel about their job. Barsade (2002) found that emo-
tional contagion (Hatfield, Cacioppo, & Rapson, 1994) in a work group
subsequently impacted on individual attitudes and willingness to cooperate,
irrespective of the pleasantness of the emotion or its intensity.
At about the same time as Salancik and Pfeffer, Karasek (1979) devel-
oped the DCM, which posited that some work characteristics create higher
psychological demands on the employee (e.g., workload) and when these

are combined with low control (e.g., autonomy) the person experiences
mental strain. Morgeson and Humphrey (2006) identified social support,
which can include emotional support and informational aids, as one social
characteristic that has an important influence on employees’ ability to do
their work effectively and for them to remain in good health. The most
relevant perspectives that incorporate social support include an extension
to the aforementioned DCM, the Demand-Control-Support (DCS) model
(Karasek & Theorell, 1990), and the Job Demands Resources (JDR) model
(Bakker & Demerouti, 2007). Research into the potential buffering effects
of social support on psychological strain using the DCS model has gener-
ally been inconclusive, with Bakker and Demerouti (2007) arguing that
social support instead acts more directly on employee well-being, by affect-
ing their engagement in the work.
According to the JDR model (Bakker & Demerouti, 2007) various work
characteristics impact on employee well-being by reducing the energy the
employee has to deal with various aspects of the work. These job demands
may include physical, psychological, or organizational aspects of the work.
Each job also carries with it some resources, so named because they tend to
replenish the employees’ energy and are beneficial to their well-being. Job
resources help to diminish burnout and add to the employees’ engagement
in their work. Van den Broeck, Vander Elst, Dikkers, De Lange, and De
Witte (2012) suggested that humor may be one resource that helps employ-
ees to deal with the negative aspects of their work environment and job
characteristics, because humor generates positive emotions, such as satis-
faction and happiness. The positive effect of humor on work engagement
was greatest when role conflict and social support were low. Positive emo-
tions associated with humor may act to mediate the effects of job charac-
teristics on employee well-being. Albeit, an alternative conceptual role for
emotion influencing employee engagement is posited, rather than as a
response to perceived work characteristics.
Relational perspectives of work design that incorporate social character-
istics of the work have emerged in response to the increased interdepen-
dence deemed critical to organizational effectiveness in modern
organizations (Grant & Parker, 2009). The importance of the context in
understanding how work should be organized is also emphasized within
evolutionary psychology (Nicholson, 2010). Social embeddedness shapes
employees’ experiences of their role, tasks, and activities, and subsequent
behaviors. This approach recognizes that social interactions and relation-
ships with customers, clients, patients, and coworkers are integral to
employees’ motivation and performance. The increased use of teams and
Model of Work Design 255

network forms of organizing has stimulated a renewed interest in emphasiz-

ing the social context of work and in understanding how this interdepen-
dence relates to work design. Here, I suggest a general proposition about
how selected social contextual factors may impact on the occurrence of
affective events as follows:
Proposition 4. Social contextual factors of positive peer job evaluations,
positive coworker emotions, and social support will result in the indivi-
dual reporting more positive affective events and less negative affective
events, than if these factors were absent.

Cognitive Emotional Interactions: Affective Events and the Cognitive

Appraisal Theory

Based on AET and the cognitive appraisal theory discussed earlier, one
emerging line of emotions research has sought to identify specific categories
of affective events, in order to understand the events that are appraised
positively and hence, lead to employees experiencing positive emotions and
reducing the frequency of negative emotions (e.g., Basch & Fisher, 2000;
Kiffin-Petersen, Murphy, & Soutar, 2012; Ohly & Schmitt, 2015). These
affective events are important because their accumulative effects ultimately
also influence employee’s judgments about their work, including for
example their job satisfaction (Ashkanasy & Humphrey, 2011). Basch and
Fisher (2000) found that goal achievement, positive recognition, and
actions of colleagues (e.g., friendly, helpful, supportive, and competent
behavior) were responsible for the majority of positive emotional events
(63%) in a sample of hotel employees. Acts of colleagues (e.g., back-
stabbing, not cooperating) and management were the main causes of nega-
tive emotions (59%) in the same sample. Contrary to Herzberg et al.’s
(1959) original findings, interpersonal relationships with colleagues and
customers were a frequent source of happiness and affection for service
employees (Basch & Fisher, 2000). However, goal achievement, receiving
recognition, involvement in decision making, challenging tasks, and pro-
blem solving were also associated with specific positive emotions.
Investigating the causes of positive emotions among sales employees,
Kiffin-Petersen et al. (2012) found that solving the customer’s problem was
the most common reason why these employees experienced a positive affec-
tive “uplift” during a normal working day. This finding is consistent with
Basch and Fisher’s (2000) involvement in challenging tasks being associated

with enthusiasm. Ohly and Schmitt (2015) recently developed a comprehen-

sive taxonomy of four positive and seven negative affective event clusters
that could provide a common point of reference for future research in this
area. Their two affective event categories of agency (e.g., goal attainment,
overload) and communal (e.g., conflicts and communication issues) are
broadly consistent with Barrick et al.’s (2013) universal work goals. Whilst
these types of taxonomies are helpful in indicating “peaks” and “lows” at
work; research is needed to identify work events that elicit less intense
but potentially more frequent positive emotions as these are thought to
have a greater influence on overall happiness (Diener et al., 1991). Based
on my discussion, I put forward two overarching propositions about the
event appraisal relationship:
Proposition 5. Affective events that are appraised as fulfilling self-
referential goals will elicit more frequent pleasant or positive emotions.

Proposition 6. Affective events that are appraised as blocking self-

referential goals will elicit more frequent unpleasant or negative

Emotion Regulation: Emotions as “Energizing” and “Exhausting”

Ashkanasy (2003) has stressed the importance of studying emotions at

intra- and inter-personal levels because of the variation associated with the
neurophysiological processes that underlie emotional experiences.
Understanding intrapersonal fluctuations in emotional states is particularly
important for employees’ general health and well-being (Xanthopoulou,
Bakker, & Ilies, 2012). In particular, it is important to understand how to
foster positive emotions, which have a positive influence on personal well-
being through their broaden-and-build effect (Fredrickson, 1998, 2001,
2013a). Within-person studies of emotional fluctuations and performance
also offer important avenues for understanding how work characteristics
ultimately may translate into improved performance (Fisher & Noble,
2004). Nicholson (2010) suggested that part of the self-regulation of the self
involves controlling one’s emotions in order to take care of them within the
wider context of our social environment, including, for example, surviving
in a work context. Gross (2013) observed therefore that a key concern of
emotion regulation research is how to cultivate emotions that are helpful
and at the same time be able to manage negative harmful emotions.
Model of Work Design 257

Using self-reported measures of emotions, Diefendorff, Richard, and

Yang (2008) found the negative emotions of annoyance, frustration, and
anger were most commonly reported in situations that require emotion reg-
ulation. In contrast, Fredrickson (2013a) identified 10 positive emotions
that people appear to feel more often, with joy, gratitude, serenity, interest,
hope, and pride being some of the most common ones applicable to the
workplace. Understanding the specific appraisal pattern that triggers a spe-
cific emotion, whether it is positive or negative, is a line of research that
has the potential to inform work design (e.g., see Fredrickson, 2013a and
Kiffin-Petersen et al., 2012), particularly if also connected to the overarch-
ing self-referential goals of the individual.
Gross (2013) believes that new research methods such as experience sam-
pling, fMRI, electroencephalography, and deep brain stimulation will help
us to discern specific underlying psychological and physiological mechan-
isms, which will enable us to better influence and manage our emotions in
real time. By way of example, imaging data has shown differences between
reappraisal and suppression forms of emotional regulation, with the former
decreasing amygdala/insular activity over time and the latter increasing it
(Ohira et al., 2006). This finding provides a potential explanation for why
certain forms of emotional control may be more emotionally exhausting, as
evidenced by meta-analytical studies of emotional labor cited earlier.
Practically speaking, the findings are also of significance since they suggest
that teaching employee’ reappraisal strategies, rather than suppressing their
emotions, may be more effective in reducing emotional exhaustion. Kühn,
Haggard, and Brass (2014) have also shown using fMRI that different
areas of the brain are involved in the emotional self-control depending on
whether we choose to suppress an emotion, or are instructed to do so by an
external party. Future research is needed to examine how these different
strategies aimed at regulating emotions affect important outcomes.
According to Parker (2014), designing work for motivation is a neces-
sary part of the story, but it is insufficient for the complexities associated
with modern day patterns of working. New theories of job design have
emerged (Corsun & Enz, 1999; Grandey & Diamond, 2010; Grant &
Parker, 2009) including proactivity (Crant, 2000; Parker, 1998) in an
attempt to meet this need. Emotional reactions have been directly linked
to proactive behaviors (Weiss & Cropanzano, 1996). Employees may
engage in proactive activities as part of their normal work day. Proactive
employees take control and seek to make things happen, so proactivity
typically involves them engaging in activities that are “self-starting,
change-oriented, and future-focused” (Parker, Bindl, & Strauss, 2010,

p. 828). “Energized to” is considered to be an important direct affective

pathway that influences the setting of goals and individual persistence in
striving toward achieving a range of proactive goals (Parker et al., 2010).
Research has also found that a high level of positive emotion on one work
day leads to higher levels of proactivity on the same and the next work
day among a group of civil service employees (Fritz & Sonnentag, 2009).
Regulating positive emotions therefore, also has a potentially important
motivational role influencing employees’ willingness to engage in proactive
behaviors at work.
Emerging work design perspectives envision a more dynamic, interactive
role for emotions in motivating employees, in which positive emotions are
more critical in driving intrinsic motivation and achieving proactive goals.
Ashby, Isen, and Turken (1999) believe that positive emotions also have an
effect on the performance of various cognitive tasks including problem sol-
ving in particular, because increases in these emotions increase dopamine
release in the anterior cingulate. These perspectives more closely align
themselves with emerging research from affective neuroscience, which
emphasizes the close neural connection between emotion and thought in
understanding how the brain interprets and acts on cues from the external
environment. Emerging relational perspectives highlight the importance of
the social characteristics of the work and interactions with organizational
outsiders in considering how work is designed. Emotions are a potential
resource to be controlled or regulated in some way to increase employee
engagement, highlighting the need to recognize employees’ control of their
emotional experiences as an important individual factor within work design
Spector and Fox (2002) theorized that an employee’s emotional state
also influences their willingness to do extra-role or voluntary behavior in
organizations. In their proposed emotion model, the person’s appraisal of
their environment leads them to experience both negative and positive emo-
tional reactions, depending on their baseline emotional state, perceptions
of control, and personality. Employees negative emotions are hypothesized
to lead to increased counterproductive work behaviors and their positive
emotions to increased organizational citizenship behaviors. Frequent sup-
pression of negative emotions is also potentially associated with increased
emotional exhaustion, absenteeism, and increased stress in service jobs
(i.e., Bono & Vey, 2005; Hülsheger, & Schewe 2011; Mesmer-Magnus
et al., 2012; Wang et al., 2011). Importantly, Spector and Fox suggested in
a similar vein to job crafting that focusing on employees’ emotional experi-
ences may be one way to effectively influence their voluntary behavior.
Model of Work Design 259

The importance of positive emotions for human well-being and optimal

flourishing is now generally accepted (Fredrickson, 1998, 2001, 2013a;
Garland et al., 2010; Lyubomirsky et al., 2005). The broaden-and-build
theory (Fredrickson, 1998) posits that positive emotions help to broaden
individual’s thought-action repertoires so they can more easily access
higher level connections between their thinking and their emotional experi-
ences. This enables greater flexibility in how the individual subsequently
behaves in a particular situation. Experiencing frequent positive emotions
over time also helps build enduring personal resources and resilience
against stress (Cohn, Fredrickson, Brown, Mikels, & Conway, 2009;
Fredrickson, Cohn, Coffey, Pek, & Finkel, 2008). Fredrickson and Losada
(2005) initially advocated a 3-to-1 ratio of positive to negative emotions for
optimal flourishing. Evidence for this critical ratio however, has been ques-
tioned because it is derived from a fundamentally erroneous mathematical
analysis (Brown, Sokal, & Friedman, 2013). Fredrickson (2013b) responded
by arguing that regardless of the mathematics, accumulating empirical evi-
dence does support the value of more positive emotions, but only up to
a point. Job crafting (Wrzesniewski & Dutton, 2001) focusing on changing
work tasks, activities, relationships, and responsibilities to increase affective
events, such as goal attainment and problem solving, that elicit positive
emotions may therefore have a beneficial effect on employees’ resilience to
Proposition 7a. Frequent pleasant emotions will result in increased
proactivity, organizational citizenship behaviors, intrinsic motivation,
and improved well-being.

Proposition 7b. Experiencing positive or pleasant emotions will act as a

buffer and reduce emotional exhaustion and counterproductive work
behaviors, and increase resilience to stress.

Proposition 8. Experiencing unpleasant emotions will increase emotional

exhaustion and counterproductive work behaviors, and reduce stress


Given the lack of research attention to work design in general (Parker,

2014); the inclusion of emotions within models of work design may help to

further our understanding of interactive jobs and their effects on customers

and employees. The SERM of work design that I have proposed here is a
first step toward integrating these divergent streams to improve our under-
standing of this complex phenomenon. The model posits that striving for
universal goals influences how a person responds to the work characteris-
tics, such that an event that is personally relevant or meets a “self-
referential goal” will elicit an emotional reaction that must be regulated for
optimal performance, proactivity, voluntary behaviors, and employee well-
being to be realized.
My review of the literature has also highlighted several significant obser-
vations pertinent to future work design research. First, problematic to the
fundamentals of AET is the growing recognition by emotion researchers
that a complex, reciprocal relationship exists between thoughts and emo-
tion, such that the reverse is also true in that emotion can also shape an
individual’s appraisal of an event (Izard, 2009). This viewpoint is reflected
in the affective neuroscience shift from a functional localization approach
to understanding brain function, to an integrated, interactive one. Behavior
is believed to emerge from the “rich, dynamic interactions between brain
networks” (Pessoa, 2008, p. 148); hence, there is a need to study emotional
and cognitive processes in concert with each other. It is recognized that the
SERM approach to work design presented here does not take account of
automatic or conditioned responses that generate emotions (Dalgleish
et al., 2009). This direct pathway to emotion would benefit from further
research using neuroscience methods that are not reliant on self-report
measure of emotions, which are essentially problematic here because such
reports rely on memory and cognitive appraisals.
Second, longitudinal studies of the relationship between job characteris-
tics, affective events, and discrete emotions are needed to investigate multi-
ple time points in order to assess how these parts of the work design puzzle
interact to achieve optimal outcomes (Saavedra & Kwun, 2000). Further to
this I would argue there is a need to also include the cognitive appraisal
theory within these efforts to provide a more complete picture of how
employees internally respond to changes in work characteristics over time
and the emotional control involved in this self-determination. Growth need
strength is one individual difference variable that has been shown to mod-
erate the relationship between the job characteristics and affect, so this may
be a consideration in future studies (Saavedra & Kwun, 2000).
Third, there is also a need to expand the list of core job characteristics
beyond those contained within the JCM. Recent research has highlighted
the motivating properties of problem solving or challenging tasks within
Model of Work Design 261

our knowledge-based economies. The clear link between this aspect of the
work and task performance may be the reason to believe this feature of the
work has strong self-referential value to employees. Several studies have
also shown that the interaction with others, such as customers or collea-
gues, may also be important in generating positive feelings at work
and therefore, of importance in considering the design of work. In this
regard, early job design theories remind us of the importance of the social
context as an influence on how others perceive their work. It is possible
that individual emotions may play a role in how others’ perceive their work
Fourth, the asymmetry of positive and negative emotional reactions to
events and how they interact with each other to determine the person’s
overall judgment about and enjoyment of their work is an area that
requires further study. Fossati et al. (2003) did not find differences in
neural activity based on the valence of emotional words, but rather that
emotional stimuli relevant to the self were processed differently than that
related to others. Employees’ do tend to recall negative emotional events
more easily than positive (Beaumeister, Bratslavsky, Finkenauer, & Vohs,
2001; Dasborough, 2006). This negativity imbalance has implications for
understanding how both types of experiences impact on outcomes, and
that differing degrees of emotional control may be required depending on
the valence of the emotion.
Fifth, the role of specific discrete emotions in work design is an area that
is worth pursuing further. Emerging research into emotions in working life
has emphasized the importance of studying discrete emotions rather than
grouping them under the broader labels of positive and negative affect. The
circumplex model of affect (Russell, 1980), which is empirically well sup-
ported (Larsen & Diener, 1992), classifies emotions based on their hedonic
tone (pleasant unpleasant) and degree of intensity (low activation high
activation) could be utilized in a more fine-grained approach to work
design research (Saavedra & Kwun, 2000). Saavedra and Kwun found that
job characteristics did not universally affect emotions regardless of their
valence, suggesting a more nuanced approach is required to studying emo-
tional reactions as mediating mechanisms. The importance of more fre-
quent positive emotions for well-being stressed here also highlights the
need to further explore these types of emotions in relation to optimal
human flourishing,
Finally, recent reviews of work design have called for researchers to focus
on individual differences and how one person, may be affected by whole-
scale work design changes (Nicholson, 2010). This observation is also

reflected in the cognitive appraisal theory of emotion, since it is the person’s

actual appraisal of the event and working conditions that will determine
how they perceive the work and their emotional reaction to particular job
characteristics. As mentioned earlier, personality influences may interact
with the work environment factors to encourage the particular expression
or certain reactions to work characteristics. Hence, future studies into the
role of emotion in work design needs to, at a minimum, consider the context
and individual differences in theorizing about how the future of work has
changed. How self-referential goals connect with personality is an area for
future research as this might help determine the type and intensity of the
emotion that is generated and how much sustained effort the person may be
willing to invest in working toward goal attainment.


Work design is, arguably, of even greater practical importance in contem-

porary workplaces than in the postscientific management era, given
changes in the way work is organized (Morgeson & Campion, 2003). The
SERM of work design presented in this chapter is informed by the macro
and micro theories from multiple disciplines reviewed here, including emo-
tions research and recent advances in social cognitive and affective neu-
roscience. In doing so, I have endeavored to address the concerns of
Nicholson (2010, p. 423) who urged researchers to consider “the relation-
ship between the design of work and the design of man, for clearly they are
interdependent.” Along with improving the motivating potential of a job,
recent research suggests that work design should also aim at maximizing
the totality of positive emotional experiences for optimal human flourishing
and stress resilience (Vacharkulksemsuk & Fredrickson, 2013). For this
reason, I have argued for the importance of including emotions and their
interactions with cognition in future work design research.
The conceptual model proposed in Fig. 1 makes several important con-
tributions to the literature. First, the model incorporates the importance of
higher order universal goals derived from Barrick et al.’s (2013) theory of
purposeful behavior, influencing how work characteristics, together with
emotional job demands and social contextual factors influence affective
events and subsequent individual outcomes. The importance of social con-
textual factors as an influence on employees’ perceptions of their work is
given greater prominence. Second, the macrostructure of AET (Weiss &
Model of Work Design 263

Cropanzano, 1996) when combined with the micro theory of cognitive

appraisal theory provides a potential mediating mechanism that intervenes
between work characteristics and explains how individual’s process infor-
mation in relation to the meaningfulness, experienced responsibility, and
knowledge of results found within the traditional JCM (Hackman &
Oldham, 1980). Significant to the model however, is the key principle of
appraisals differing depending on the self-relevance of the specific affective
event that has occurred. Third, the model emphasizes the importance of the
cognitive emotional interaction in the appraisal and reappraisal of events
arising from work characteristics that has been overlooked in prior
research. In particular, findings from affective neuroscience in relation to
anticipatory rewards and self-inhibiting behavior have shown that differing
brain activation occurs when the person evaluates the stimulus in relation
to themselves, versus others. Finally, the negative or positive emotions that
arise from the person’s appraisal and reappraisal are expected to have dif-
ferential outcomes, highlighting the importance of studying the asymmetry
of these emotional processes separately and jointly in future research.
It is important for practice and policy to pay attention to the way work
is designed, since poorly designed jobs are still evident in contemporary
workplaces (Parker, 2014). The focus of work design initiatives on the types
of activities, degree of decision latitude, and complexity of tasks that
should be incorporated into a meaningful job remains of critical impor-
tance for contemporary jobs (Parker, 2014). I assert, based on my review of
cognitive emotional interactions involved in work design, that work
design also needs to consider the following questions: How should tasks,
activities, and relationship be combined together so as to give rise to fre-
quent, positive emotional states? How can we combine tasks together so as
to minimize the impact of negative emotions on employees’ stress and psy-
chological well-being? How does the emotion regulation of positive and
negative emotions interact to influence thoughts and perceptions of the job
and ultimately, the employees’ health, psychological well-being, and perfor-
mance? Asking such questions will help to ensure that emotions are given
the prominent role in work design interventions they deserve within mod-
ern workplaces.


Ambrose, M. L., & Kulik, C. T. (1999). Old friends, new faces: Motivation in 1990s. Journal
of Management, 25, 231 292.

Ashby, F. G., Isen, A. M., & Turken, A. M. (1999). A neuropsychological theory of positive
affect and its effect on cognition. Psychological Review, 106, 529 550.
Ashkanasy, N. M. (2003). Emotions in organizations: A multilevel perspective. In
F. Dansereau & F. J. Yammarino (Eds.), Multi-level issues in organizational behavior
and strategy (Vol. 2, pp. 9 54). Research in Multi-Level Issues. Bingley, UK: Emerald
Group Publishing Limited.
Ashkanasy, N. M., & Humphrey, R. H. (2011). Current emotion research in organizational
behavior. Emotion Review, 3, 214 224.
Bakker, A. B., & Demerouti, E. (2007). The job-demands-resources model: State of the art.
Journal of Managerial Psychology, 22(3), 309 328.
Barrick, M. R., Mount, M. L., & Li, N. (2013). The theory of purposeful work behavior: The
role of personality, higher-order goals, and job characteristics. Academy of
Management Review, 38, 132 153.
Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on group beha-
vior. Administrative Science Quarterly, 47, 644 675.
Basch, J., & Fisher, C. D. (2000). Affective events-emotions matrix: A classification of work
events and associated emotions. In N. M. Ashkanasy, C. E. J. Härtel, & W. J. Zerbe
(Eds.), Managing emotions in the workplace: Theory, research and practice (pp. 36 48).
Westport, CT: Quorum Books.
Beaumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger
than good. Review of General Psychology, 5, 323 370.
Becker, W. J., Cropanzano, R., & Sanfey, A. G. (2011). Organizational neuroscience: Taking
organizational theory inside the neural black box. Journal of Management, 37,
933 961.
Bono, J. E., & Vey, M. A. (2005). Toward understanding emotional management at work:
A quantitative review of emotional labor research. In C. E. J. Härtel, W. J. Zerbe, &
N. M. Ashkanasy (Eds.), Emotions in organizational behavior (pp. 213 233). Hillsdale,
NJ: Lawrence Erlbaum Associates.
Boyatzis, R. E., Smith, M. L., & Blaize, N. (2006). Developing sustainable leaders through
coaching and compassion. Academy of Management Learning and Education, 5, 8 24.
Brotheridge, C. M., & Lee, R. T. (2003). Development and validation of the emotional labour
scale. Journal of Occupational and Organizational Psychology, 76, 365 379.
Brown, N. J. L., Sokal, A. D., & Friedman, H. L. (2013). The complex dynamics of wishful
thinking: The critical positivity ratio. American Psychologist, 68, 801 813.
Camerer, C., Loewenstein, G., & Prelec, D. (2005). Neuroeconomics: How neuroscience can
inform economics. Journal of Economic Literature, 43, 9 64.
Cavanaugh, M. A., Boswell, W. R., Roehling, M. V., & Boudreau, J. W. (2000). An empirical
examination of self-reported work stress among U.S. managers. Journal of Applied
Psychology, 85, 65 74.
Cohn, M. A., Fredrickson, B. L., Brown, S. L., Mikels, J. A., & Conway, A. M. (2009).
Happiness unpacked: Positive emotions increase life satisfaction by building resilience.
Emotion, 9, 361 368.
Compton, R. J. (2003). The interface between emotion and attention: A review of evidence
from psychology and neuroscience. Behavioral and Cognitive Neuroscience Reviews,
2(2), 115 129.
Corsun, D. L., & Enz, C. A. (1999). Predicting psychological empowerment among service
workers: The effect of support based relationships. Human Relations, 52, 205 224.
Model of Work Design 265

Crant, J. M. (2000). Proactive behavior in organizations. Journal of Management, 26,

435 462.
Cropanzano, R., & Becker, W. J. (2013). The promise and peril of organizational neu-
roscience: Today and tomorrow. Journal of Management Inquiry, 22, 306 310.
Cummings, T. G. (1978). Self-regulating work groups: A socio-technical synthesis. Academy of
Management Review, 3, 625 637.
Dalgleish, T., Dunn, B. D., & Mobbs, D. (2009). Affective neuroscience: Past, present, and
future. Emotion Review, 1, 355 368.
Dasborough, M. T. (2006). Cognitive asymmetry in employee emotional reactions to leader-
ship behaviors. The Leadership Quarterly, 17, 163 178.
Davidson, R. J., Pizzagalli, D., Nitschke, J. B., & Kalin, N. H. (2003). Parsing the subcompo-
nents of emotion and disorders of emotion: Perspectives from affective neuroscience. In
R. J. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Handbook of affective
sciences (1st ed., pp. 9 24). Oxford, UK: Oxford University Press.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human beha-
vior. New York, NY: Plenum.
Diefendorff, J. M., Richard, E. M., & Yang, J. (2008). Linking emotion regulation strategies
to affective events and negative emotions at work. Journal of Vocational Behavior, 73,
498 508.
Diener, E., Sandvik, E., & Pavot, W. (1991). Happiness is the frequency, not the intensity, of
positive versus negative affect. In F. Strack, M. Argyle, & N. Schwartz (Eds.),
Subjective well-being: An interdisciplinary perspective (Vol. 21, pp. 119 139). New
York, NY: Pergamon.
Elsbach, K. D., & Hargadon, A. B. (2006). Enhancing creativity through “mindless” work: A
framework of workday design. Organization Science, 17, 470 483.
Fisher, C. D., & Ashkanasy, N. M. (2000). The emerging role of emotions in work life: An
introduction. Journal of Organizational Behavior, 21, 123 129.
Fisher, C. D., & Noble, C. S. (2004). A within-person examination of correlates of perfor-
mance and emotions while working. Human Performance, 17, 145 168.
Fossati, P., Hevenor, S. J., Graham, S. J., Grady, C., Keightley, M. L., Craik, F., & Mayberg,
H. (2003). In search of the emotional self: An fMRI study using positive and negative
emotional words. American Journal of Psychiatry, 160, 1938 1945.
Fredrickson, B. L. (1998). What good are positive emotions? Review of General Psychology, 2,
300 319.
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology. American
Psychologist, 56, 218 226.
Fredrickson, B. L. (2013a). Positive emotions broaden and build. In P. Devine & A. Plant
(Eds.), Advances in experimental social psychology (Vol. 47, pp. 1 54). San Diego, CA:
Academic Press.
Fredrickson, B. L. (2013b). Updated thinking on positivity ratios. American Psychologist, 68,
814 822.
Fredrickson, B. L., Cohn, M. A., Coffey, K. A., Pek, J., & Finkel, S. M. (2008). Open hearts
build lives: Positive emotions, induced through loving-kindness meditation build conse-
quential person resources. Journal of Personality and Social Psychology, 95, 1045 1062.
Fredrickson, B. L., & Losada, M. F. (2005). Positive affect and the complex dynamics of
human flourishing. American Psychologist, 60, 678 686.
Frijda, N. H. (1988). The laws of emotion. American Psychologist, 43, 349 358.

Frijda, N. H. (1994). Varieties of affect: Emotions and episodes, moods, and sentiments. In
P. Ekman & R. J. Davison (Eds.), The nature of emotions: Fundamental questions
(pp. 59 67). Oxford, UK: Oxford University Press.
Fritz, C., & Sonnentag, S. (2009). Antecedents of day-level proactive behavior: A look at job
stressors and positive affect during the workday? Journal of Management, 35, 94 111.
Gable, S. L., Reis, H. T., & Elliot, A. J. (2000). Behavioral activation and inhibition in every-
day life. Journal of Personality and Social Psychology, 78, 1135 1149.
Garland, E. L., Fredrickson, B., Kring, A. M., Johnson, D. P., Meyer, P. S., & Penn, D. L.
(2010). Upward spirals of positive emotions counter downward spirals of negativity:
Insights from the broaden-and-build theory and affective neuroscience on the treatment
of emotion dysfunctions and deficits in psychopathology. Clinical Psychological Review,
30, 849 864.
Grandey, A. A. (2000). Emotion regulation in the workplace: A new way to conceptualize
emotional labor. Journal of Occupational Health Psychology, 5, 95 110.
Grandey, A. A. (2003). When “the show must go on”: Surface and deep acting as determinants
of emotional exhaustion and peer-related service delivery. Academy of Management
Journal, 46, 86 96.
Grandey, A. A., & Diamond, J. A. (2010). Interactions with the public: Bridging job design
and emotional labor perspectives. Journal of Organizational Behavior, 31, 338 350.
Grandey, A. A., Kern, J. H., & Frone, M. R. (2007). Verbal abuse from outsiders versus insi-
ders: Comparing frequency, impact on emotional exhaustion, and the role of emotional
labor. Journal of Occupational Health Psychology, 12, 63 79.
Grant, A. M., Fried, Y., Parker, S. K., & Frese, M. (2010). Putting job design in context:
Introduction to the special issue. Journal of Organizational Behavior, 31, 145 157.
Grant, A. M., & Parker, S. K. (2009). Redesigning work design theories: The rise of relational
and proactive perspectives. Academy of Management Annals, 3, 317 375.
Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of
General Psychology, 2, 271 299.
Gross, J. J. (2013). Emotion regulation: Taking stock and moving forward. Emotion, 13,
359 365.
Gross, J. J., Sheppes, G., & Urry, H. L. (2011). Emotion generation and emotion regulation:
A distinction we should make (carefully). Cognition and Emotion, 25, 765 781.
Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-Wesley.
Harmon-Jones, E., & van Honk, J. (2012). Introduction to a special issue on the neuroscience
of motivation and emotion. Motivation and Emotion, 36, 1 3.
Hatfield, E., Cacioppo, J., & Rapson, R. L. (1994). Emotional contagion. New York, NY:
Cambridge University Press.
Herzberg, F. (1966). Work and the nature of man. New York, NY: Thomas Y. Crowell.
Herzberg, F. (1987). One more time: How do you motivate employees? Harvard Business
Review, 65(5), 109 120.
Herzberg, F., Mausner, B., & Snyderman, B. B. (1959). The motivation to work. New York,
NY: Wiley.
Hochschild, A. R. (1983). The managed heart: Commercialization of human feeling. Berkeley:
CA: University of California Press.
Hülsheger, U. R., & Schewe, A. F. (2011). On the costs and benefits of emotional labor: A
meta-analysis of three decades of research. Journal of Occupational Health Psychology,
16, 361 389.
Model of Work Design 267

Humphrey, R. H. (2000). The importance of job characteristics to emotional displays. In

N. M. Ashkanasy, C. E. J., Härtel, & W. J. Zerbe (Eds.), Emotions in the workplace:
Research, theory, and practice (pp. 236 249). Westport, CT: Quorum Books.
Humphrey, S. E., Nahrgang, J. D., & Morgeson, F. P. (2007). Integrating motivational,
social, contextual work design features: A meta-analytic summary and
theoretical extension of the work design literature. Journal of Applied Psychology, 92,
1332 1356.
Isen, A. M. (2000). Positive affect and decision making. In M. Lewis & J. Haviland-Jones
(Eds.), Handbook of emotions (pp. 417 435). New York, NY: Guildford.
Isen, A. M., & Reeve, J. (2005). The influence of positive affect on intrinsic and extrinsic moti-
vation: Facilitating enjoyment of play, responsible work behavior, and self-control.
Motivation and Emotion, 29, 297 325.
Izard, C. E. (1993). Four systems for emotion activation: Cognitive and noncognitive pro-
cesses. Psychological Review, 100, 68 90.
Izard, C. E. (2009). Emotion theory and research: Highlights, unanswered questions, and
emerging issues. Annual Review of Psychology, 60, 1 25.
Johnson, H. M., & Spector, P. E. (2007). Service with a smile: Do emotional intelligence, gen-
der, and autonomy moderate the emotional labor process? Journal of Occupational
Health Psychology, 12, 319 333.
Karasek, R. A. (1979). Job demands, job decision latitude, and mental strain: Implications for
job redesign. Administrative Science Quarterly, 24, 285 308.
Karasek, R. A., & Theorell, T. (1990). Healthy work. New York, NY: Basic Books.
Kiffin-Petersen, S. A., Murphy, S. A., & Soutar, G. (2012). The problem-solving service
worker: Appraisal mechanisms and positive affective experiences during customer inter-
actions. Human Relations, 65, 1179 1206.
Knutson, B., Adams, C. M., Fong, G. W., & Hommer, D. (2001). Anticipation of increasing
monetary reward selectively recruits nucleus accumbens. The Journal of Neuroscience,
21, 1 5.
Kühn, S., Haggard, P., & Brass, M. (2014). Differences between endogenous and exogenous
emotion inhibition in the human brain. Brain Structure & Function, 219, 1129 1138.
Larsen, R. L., & Diener, E. (1992). Promises and problems with the circumplex model of
emotion. In M. S. Clark (Ed.), Review of personality and social psychology: Emotion
(Vol. 114, pp. 25 59). Newbury Park, CA: Sage.
Lazarus, R. S. (1966). Psychological stress and the coping process. New York, NY: McGraw-Hill.
Lazarus, R. S. (1991). Progress on a cognitive-motivational-relational theory of emotion.
American Psychologist, 46, 819 834.
Lindquist, K. A., Wager, T. D., Kober, H., Bliss-Moreau, E., & Feldman Barrett, L. (2012).
The brain basis of emotion: A meta-analytic review. Behavioral and Brain Sciences, 35,
121 202.
Lyubomirsky, S., King, L., & Diener, E. (2005). The benefits of frequent positive affect: Does
happiness lead to success? Psychological Bulletin, 131, 803 855.
Mesmer-Magnus, J. R., DeChurch, L. A., & Wax, A. (2012). Moving emotional labor beyond
surface and deep acting: A discordance-congruence perspective. Organizational
Psychology Review, 2, 6 53.
Morgeson, F. P., & Campion, M. A. (2003). Work design. In W. Borman, D. Ilgen, &
R. Klimoski (Eds.), Handbook of psychology: Industrial and organizational psychology
(Vol. 12, pp. 423 452). Hoboken, NJ: Wiley.

Morgeson, F. P., & Humphrey, S. E. (2006). The Work Design Questionnaire (WDQ):
Developing and validating a comprehensive measure for assessing job design and the
nature of the work. Journal of Applied Psychology, 91, 1321 1339.
Morgeson, F. P., & Humphrey, S. E. (2008). Job and team design: Toward a more integrative
conceptualization of work design. In J. J. Martocchio (Ed.), Research in personnel and
human resources management (Vol. 27, pp. 39 91). Research in Personnel and Human
Resources Management. Bingley, UK: Emerald Group Publishing.
Morris, J. A., & Feldman, D. C. (1996). The dimensions, antecedents, and consequences of
emotional labor. Academy of Management Review, 21, 986 1010.
Morris, J. A., & Feldman, D. C. (1997). Managing emotions in the workplace. Journal of
Managerial Issues, 9(3), 257 274.
Nicholson, N. (2010). The design of work An evolutionary perspective. Journal of
Organizational Behavior, 31, 422 431.
Ochsner, K. N., & Gross, J. J. (2008). Cognitive emotion regulation: Insights from social cog-
nitive and affective neuroscience. Current Directions in Psychological Science, 17(2),
153 158.
Ohira, H., Nomura, M., Ichikawa, N., Isowa, T., Iidaka, T., Sato, A., … Yamada, J. (2006).
Association of neural and physiological responses during voluntary emotion suppres-
sion. Neuroimage, 29, 721 733.
Ohly, S., & Schmitt, A. (2015). What makes us enthusiastic, angry, feeling at rest or worried?
Development and validation of an affective work events taxonomy using concept map-
ping methodology. Journal of Business & Psychology, 30, 15 35.
Oldham, G. R., & Hackman, J. R. (2010). Not what it was and not what it will be: The future
of job design research. Journal of Organizational Behavior, 31, 463 479.
Parker, S., Bindl, U., & Strauss, K. (2010). Making things happen: A model of proactive moti-
vation. Journal of Management, 36, 827 856.
Parker, S. K. (1998). Enhancing role breadth self-efficacy: The roles of job enrichment and
other organizational interventions. Journal of Applied Psychology, 83, 835 852.
Parker, S. K. (2014). Beyond motivation: Job and work design for development, health, ambi-
dexterity, and more. Annual Review of Psychology, 65, 661 691.
Parker, S. K., & Wall, T. D. (1998). Job and work design: Organizing work to promote well-
being and effectiveness. San Francisco, CA: Sage.
Parker, S. K., Wall, T. D., & Cordery, J. L. (2001). Future work design research and practice:
Towards an elaborated model of work design. Journal of Occupational and
Organizational Psychology, 74, 413 440.
Pessoa, L. (2008). On the relationship between emotion and cognition. Nature Reviews
Neuroscience, 9(2), 148 158.
Roseman, I. J., Spindel, M. S., & José, P. E. (1990). Appraisal of emotion-eliciting events: Testing
a theory of discrete emotions. Journal of Personality and Social Psychology, 59, 899 915.
Russell, J. A. (1980). The circumplex model of affect. Journal of Personality and Social
Psychology, 39, 1161 1178.
Saavedra, R., & Kwun, S. K. (2000). Affective states in job characteristics theory. Journal of
Organizational Behavior, 21, 131 146.
Sachau, D. A. (2007). Resurrecting the motivation-hygiene theory: Herzberg and the positive
psychology movement. Human Resource Development Review, 6(4), 377 393.
Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to job attitudes
and task design. Administrative Science Quarterly, 23, 224 253.
Model of Work Design 269

Seo, M., Barrett, L. F., & Bartunek, J. M. (2004). The role of affective experience in work
motivation. Academy of Management Review, 29, 423 439.
Spector, P. E., & Fox, S. (2002). An emotion-centred model of voluntary work behavior:
Some parallels between counterproductive work behavior and organizational citizen-
ship behavior. Human Resource Management Review, 12, 269 292.
Taylor, F. W. (1911). The principles of scientific management. New York, NY: W. W. Norton.
Trist, E. L., & Bamforth, K. M. (1951). Some social and psychological consequences of the
longwall method of coal-getting. Human Relations, 4, 3 38.
Vacharkulksemsuk, T., & Fredrickson, B. L. (2013). Looking back and glimpsing forward:
The broaden-and-build theory of positive emotions as applied to organizations.
Advances in Positive Organizational Psychology, 1, 45 60.
Van den Broeck, A., Vander Elst, T., Dikkers, J., De Lange, A., & De Witte, H. (2012). This
is funny: On the beneficial role of self-enhancing and affiliative humour in job design.
Psicotherma, 24(1), 87 93.
Waldman, D. A., Balthazard, P. A., & Peterson, S. (2011). Leadership and neuroscience: Can
we revolutionize the way that inspirational leaders are identified and developed?
Academy of Management Perspective, 25(1), 60 74.
Wang, G., Seibert, S. E., & Boles, T. (2011). Synthesizing what we have known and looking
ahead: A meta-analytic review of 30 years of emotional labor research. In C. E. J.
Härtel, N. M. Ashkanasy, & W. J. Zerbe (Eds.), What have we learned? Ten years on
(Vol. 7, pp. 1 45). Research on Emotion in Organizations. Bingley, UK: Emerald
Group Publishing Limited.
Waugh, C. E., Wager, T. D., Fredrickson, B. L., Noll, D. C., & Taylor, S. F. (2008). The
neural correlates of trait resilience when anticipating and recovering from threat. Social
Cognitive and Affective Neuroscience, 3(4), 322 332.
Waytz, A., & Mason, M. (2013). Your brain at work: What a new approach to neuroscience
can teach us about management. Harvard Business Review, 91(7/8), 102 111.
Weiss, H. M., & Beal, D. J. (2005). Reflections on affective events theory. In N. M.
Ashkanasy, W. J. Zerbe, & C. E. J. Härtel (Eds.), The effect of affect in organizational
settings (Vol. 18, pp. 1 21). Research on Emotion in Organizations. San Diego, CA:
Emerald Group Publishing Limited.
Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion of
affective experiences at work. In B. M. Staw & L. L. Cummings (Eds.), Research in
organizational behavior (pp. 1 74). Greenwich, CT: JAI Press.
Wrzesniewski, A., & Dutton, J. E. (2001). Crafting a job: Revisioning employees as active craf-
ters of their work. The Academy of Management Review, 26, 179 201.
Xanthopoulou, D., Bakker, A. B., & Ilies, R. (2012). Everyday working life: Explaining
within-person fluctuations in employee well-being. Human Relations, 65, 1051 1069.
This page intentionally left blank



Jim A. McCleskey


This chapter examines EI, presents a history of EI including the various

models, and a discussion of the three streams approach to classifying EI
literature. The author advocates for the efficacy of the Stream One
Ability Model (SOAM) of EI citing previous authors and literature.
The commonly used SOAM instruments are discussed in light of recent
studies. The discussion turns to alternate tests of the SOAM of EI
including Situational Judgment Tests (SJTs). Recommendations include
an analysis of SOAM instruments, a new approach to measurement, and
increased use of SJTs to capture the four-branch ability model of EI.
Keywords: Emotional intelligence; EI; ability model; MSCEIT; SJTs;
emotion regulation

New Ways of Studying Emotions in Organizations

Research on Emotion in Organizations, Volume 11, 271 293
Copyright r 2015 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN: 1746-9791/doi:10.1108/S1746-979120150000011011


The study of organizations experienced an affective revolution in the past

two decades. The discussion of emotions in organizations and workplaces
has evolved from a position of nearly “undiscussable” to a mainstream
topic in organizational behavior, management, and leadership (Ashkanasy,
Hartel, & Zerbe, 2012, p. 1). Indeed, most modern textbooks on organiza-
tional behavior, management, and leadership include a mention of emo-
tions and/or emotional intelligence (EI) if not an entire chapter on these
topics. While this revolution impacts these fields at all levels of scholarship,
the relatively recent nature of these changes contributes to an evolving and
dynamic approach to the study of emotions in organizations. The challenge
for scholars involves, not just understanding these changes, but guiding
and facilitating them within the institutions and organizations they impact.
This chapter examines the changing nature of the study of EI, advocates
for the Stream One Ability Model (SOAM) of EI, and suggests some future
directions for EI scholarship.
Recently, a group of noted scholars suggested that in psychology, theory
building is never finished (Matthews, Zeidner, & Roberts, 2012). Further,
these authors argued that good research leads to theory refinement and
modification and on to further research in a dynamic and circular process.
These scholars might have been discussing any number of constructs in the
social sciences; however, they were specifically addressing the construct of
EI. A growing body of research drawn from a variety of disciplines includ-
ing management, organizational behavior, individual psychology, industrial
psychology, sociology, and leadership suggests that EI plays a role in a
variety of organizational and work-related outcomes (McCleskey, 2013).
However, EI continues to suffer from an acute form of theoretical plural-
ism and is variously defined as an ability, a competency, or a trait in its
models and conceptions (McCleskey, 2014). The following discussion pro-
vides an overview of these models, supports an ability model approach to
EI, and reviews some recent literature on the SOAM of EI.

History of Emotional Intelligence

In 1990, Salovey and Mayer published an article that presented a frame-

work for EI. Mayer and his colleagues were the first to publish works
that referred to an EI (Mayer, DiPaolo, & Salovey, 1990; Salovey &
SOAM of EI and the Future of EI 273

Mayer, 1990). Interestingly, the concept of EI was not new. The earliest
intelligence researchers understood there was more to the construct of
intelligence than the mental abilities represented in the traditional intelli-
gence tests of the day (Riggio, 2002). In fact, Edward Thorndike defined
a social intelligence in 1920. Unfortunately, attempts at measuring the
construct were largely unsuccessful (Thorndike & Stein, 1937). As early
as 1967, Guilford argued that multiple intelligences existed including cog-
nition, memory, divergent-production factors, convergent-production abil-
ities, and evaluation. Later, Gardner (1983) presented the idea that
individuals possessed an interpersonal and intrapersonal intelligence seven
years before the publication of the Salovey and Mayer article. Sternberg
(1985) advocated the existence of multiple domains of intelligence includ-
ing fluid intelligence, crystallized intelligence, and social (practical) intelli-
gence. Salovey and Mayer originally conceived EI as a subset of social
intelligence (1990). The term EI first appeared in Leuner (1966) although
the construct presented in Leuner’s work did not resemble the concept of
social/EI under discussion here. Payne (1986) used the term in his unpub-
lished doctoral dissertation entitled “A study of emotion: developing EI;
self-integration; relating to fear, pain, and desire.” As early as 1988, Bar-
On discussed the concept of emotional and social intelligence in his
unpublished dissertation. Despite these earlier forays into the idea,
Salovey and Mayer’s published article receives credit for the initial publi-
cation of the current conception of EI. Later, EI achieved popular notori-
ety when Goleman (1995) wrote the bestseller Emotional intelligence: Why
it can matter more than IQ and cited the work of Salovey and Mayer
(1990). Goleman’s work landed EI (also called EQ in the early days) on
the cover of Time Magazine (Gibbs & Epperson, 1995). Today a variety
of EI models and approaches exist.

The Models and Classifications of EI

Several different conceptual approaches to modeling the construct of EI

exist including the ability model, the mixed models, and the trait model.
The Mayer ability model is the most commonly accepted model of EI and
is based on a four-branch approach to EI. Four basic abilities comprise the
ability model: emotion perception; emotion facilitation; emotion under-
standing; and emotion regulation (Caruso & Salovey, 2004). Jordan,
Dasborough, Daus, and Ashkanasy (2010) described the Mayer ability
model and definition as the “gold standard” for defining EI (p. 145).

Antonakis, Ashkanasy, and Dasborough (2009) agree that the future of EI

belongs to the ability model. Other authors also express a preference for
the ability model based on largely psychometric evaluation criteria
(Matthews et al., 2012). Following the recommendations of these scholars,
this chapter includes an examination of the recent ability model literature
to understand the current state and future directions of EI research. After a
brief discussion of the competing models, the chapter returns to recent lit-
erature on the SOAM.

Mixed Models of EI

Emotional and social competency models of EI are more commonly known

as mixed models of EI. Bar-On (1988) described the first mixed-model of
EI. Bar-On’s model of competencies and skills included:
the ability to be aware of, to understand, and to express oneself; the ability to be aware
of, to understand and relate to others; the ability to deal with strong emotions and con-
trol one’s impulses; and the ability to adapt to change and to solve problems of a perso-
nal or social nature. (Bar-On, 2006, p. 14)

Bar-On’s model, which he later named the Bar-On Model of Emotional-

Social Intelligence (ESI), includes the components of interpersonal skills,
adaptability, stress management, and general mood (Bar-On, 2006, p. 14).
Bar-On’s model reached farther than the original conception and showed
signs of overlap with the more general construct of personality. Other scho-
lars developed additional mixed-model approaches to EI.
Goleman and Boyatzis developed another mixed-model approach to EI
(Boyatzis, Goleman, & Rhee, 2000). Goleman’s (1995) bestseller, Emotional
Intelligence: Why It Can Matter More Than IQ, launched the popular
concept of EI and became the “touchstone for many of the controversial
issues that continue in the academic field of EI to the present day”
(McCleskey, 2014). The Salovey and Mayer ability model inspired the
Boyatzis Goleman mixed-model. However, Boyatzis and Goleman
expanded the scope of their model to encompass social and emotional com-
petencies linked to effective performance in the workplace. These included a
number of competencies sorted into four clusters: self-awareness, self-
management, social awareness, and relationship management (Boyatzis,
2009; Cherniss, 2010; Goleman, Boyatzis, & McKee, 2002). This model also
showed overlap with the more established construct of personality. An alter-
native conception views EI as a unique personality trait.
SOAM of EI and the Future of EI 275

The Trait model of EI consists of four components: well-being

(self-confidence, happ