Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
Leaders’ and Followers’ Information-Processing Styles, Constructive Thinking
and their Relationship to Implicit Followership and Leadership Theories and
Leader-Member Exchange
Matthew Walford
BPsych (Hons), M. App Psych (Organisational)
This thesis is presented in fulfilment of the requirements
for the degree of Doctor of Philosophy - Psychology
Murdoch University
2016
ii
I declare that this thesis is my own account of my research and contains, as its main
content, work which has not previously been submitted for a degree at any tertiary
educational institution
_____________________________________________
Matthew Walford
iii
Copyright Acknowledgement
I acknowledge that a copy of this thesis will be held at the Murdoch University
Library.
I understand that, under the provisions s51.2 of the Copyright Act 1968, all or part of
this thesis may be copied without infringement of copyright where such a
reproduction is for the purposes of study and research.
This statement does not signal any transfer of copyright away from the author.
Signed: …………………………………………………………...
Full Name of Degree: Doctor of Philosophy - Psychology
Thesis Title: Leaders’ and Followers’ Information Processing Styles,
Constructive Thinking and their Relationship to Implicit
Followership and Leadership Theories and Leader-Member
Exchange
Author: Matthew P.W. Walford
Year: 2016
iv
Acknowledgements
I am extremely thankful to my supervisors, Dr Graeme Ditchburn, Dr Guy
Curtis and Mrs Libby Brook. Your encouragement, guidance, friendship and support
helped me complete this research. I would particularly like to thank my wife, Kaini.
Without your love and support during these last three years, the completion of this
project would not have been possible. I promise to repay the sacrifices you have also
made out of my choice to complete my studies. I would also like to thank my
parents, Valda and Peter Walford. You have supported me throughout this journey
like you have throughout my life. I can never repay you for everything you have
done for me. I love you both. To Jay Chinnery, thank you for your friendship and
support these years. You are a true friend in all senses of the word.
I would like to offer my sincere appreciation and thanks to all of those who
supported me in any respect during the completion of the project. This includes the
kind support of the organisations and their staff who participated and supported this
research. Your engagement was incredible and exceeded my expectations by far. It
also includes my fellow Ph.D. candidates. Each one of you has provided
encouragement and support, both professionally and personally. I wish you all the
best in your research. I look forward to continuing our friendships long after your
theses have been completed.
Matthew Walford
v
Abstract
Leadership theory and research has traditionally focused on leaders at the
expense of followers. However, followers are highly active in the leadership process.
Adopting a relationship based conceptualisation of leadership, four studies are
presented that assess elements of a holistic leadership model incorporating
Cognitive-Experiential Self Theory (CEST), Implicit Leadership Theory, Implicit
Followership Theory, and Leader-Member Exchange (LMX). Using data from 352
followers, study one found that followers develop implicit ideal leadership theories
(IILTs) that are congruent with their information processing style and constructive
thinking as defined in CEST. Using data from 133 leaders, study two found that
leaders develop implicit ideal followership theories (IIFTs) that are congruent with
their information processing style and constructive thinking. Studies three and four
are multi-level research designs, which assess whether leaders’ and followers’
information processing style and constructive thinking relate to followers’ and
leaders’ LMX respectively. They also assess whether leaders’ IIFTs and followers’
IILTs moderate these relationships. Based on data from 106 followers and 26 direct-
line managers, study three found that leaders’ information processing style and
constructive thinking predict followers’ LMX ratings and that these relationships are
moderated by followers’ IILTs. Using data from 77 followers and 18 direct-line
managers, study four demonstrates that followers’ information processing style and
constructive thinking relates to leaders’ LMX ratings. However, the evidence that
these relationships are moderated by leaders’ IIFTs is not as strong as study three.
This is likely due to a low sample size. Generally, the proposed model is supported.
The findings have important implications for leadership theory and practice
vi
including training/development, employee selection and team composition decisions.
These are discussed in the relevant sections of the dissertation.
vii
CONTENTS
CHAPTER 1 Introduction and Thesis Overview.................................................... 1
Positioning the Research’s Theoretical Approach ................................................... 3
Conceptualising the Internal World of Leaders and Followers: Dual-Processing
Models ...................................................................................................................... 5
Smith and DeCoster’s (2000) Associative vs. Rule Based Processing................. 7
Kahneman’s (2003) System 1 and System 2 processing ...................................... 8
Strack and Deutsch’s (2004) Reflective-Impulsive Model (RIM) ....................... 9
Epstein’s Cognitive Experiential Self-theory ..................................................... 11
Implicit Leadership and Followership Theories ..................................................... 21
Organisation of This Dissertation ........................................................................... 24
CHAPTER 2 Literature Review and Research Model ......................................... 25
Leadership Definitions ........................................................................................... 26
Leadership Theory Classification ........................................................................... 28
Leader-Centric Theories of Leadership .............................................................. 28
Leader-Context Match Theories. ........................................................................ 32
New Leadership Theories ................................................................................... 38
Follower Centric Theories of Leadership ........................................................... 42
Relationship-Focused Perspectives of Leadership ............................................. 49
Followership ........................................................................................................... 60
Followership Theories ........................................................................................ 61
Role-Based Theories of Followership ............................................................. 61
Relationship Based Perspectives of Followership .......................................... 70
Limitations in Leadership/Followership Research: Proposal of an Integrated
Research Model ...................................................................................................... 71
Proposal of a Research Model and Outline of Chapters 3 to 6 ........................... 75
CHAPTER 3 ILT Congruency: Exploring the Relationships between Followers’
Information Processing Style, Constructive Thinking, and Implicit Leadership
Theories ..................................................................................................................... 78
Person-Supervisor Fit: The Importance of Similarity ............................................ 80
Theory Development .............................................................................................. 81
viii
Hypothesis Development: Information Processing Style and ILTs................... 83
Hypothesis Development: Constructive Thinking and ILTs .............................. 87
Method .................................................................................................................... 91
Participants.......................................................................................................... 91
Measures ............................................................................................................. 92
Procedure ............................................................................................................ 94
Results .................................................................................................................... 95
Removal of defensive and inattentive participants ............................................. 95
Common method bias ......................................................................................... 95
Exploratory Factor Analysis – ILT Measure ...................................................... 96
Data screening & assumption checking ............................................................ 100
Descriptive Statistics......................................................................................... 100
Hypothesis Testing ........................................................................................... 102
Hierarchical Regression Analyses .................................................................... 105
Discussion ............................................................................................................ 112
Information Processing Style – ILT Congruence ............................................. 112
Constructive Thinking – ILT Congruence ........................................................ 116
Theoretical and practical implications .............................................................. 120
Limitations ........................................................................................................ 123
CHAPTER 4 IFT Congruency: Exploring the Relationships between Leaders’
Information Processing Style, Constructive Thinking, and Implicit Followership
Theories ................................................................................................................... 125
Person-environment fit ......................................................................................... 128
Theory Development ............................................................................................ 128
Hypothesis development: Information Processing Style and IFTs ................... 130
Hypothesis development: Constructive Thinking and IFTs ............................. 132
Method .................................................................................................................. 137
Participants........................................................................................................ 137
Measures ........................................................................................................... 138
Procedure .......................................................................................................... 139
Results .................................................................................................................. 140
Common method bias ....................................................................................... 140
Exploratory Factor Analysis – IFT Measure .................................................... 141
ix
Data screening & assumption checking ............................................................ 145
Descriptive Statistics, Scale Reliabilities and Scale Inter-correlations ............ 145
Hypothesis Testing ........................................................................................... 147
Hierarchical ....................................................................................................... 149
Regression ......................................................................................................... 149
Discussion ............................................................................................................ 156
Information Processing Style – IFT Congruence ............................................. 156
Constructive Thinking – IFT Congruence ........................................................ 160
Further theoretical and practical implications .................................................. 167
Limitations ........................................................................................................ 169
CHAPTER 5 Leaders’ Information Processing Style and Constructive Thinking
as Follower LMX Antecedents: The Moderating Influence of Followers’
Implicit Ideal Leadership Theory ......................................................................... 171
Leader-Member Exchange (LMX) Theory .......................................................... 172
Linking Cognitive Experiential Self Theory and LMX ....................................... 176
Follower LMX Dependent Variable ..................................................................... 181
Hypothesis development: Leader Information Processing and Follower LMX
.......................................................................................................................... 182
Hypothesis development: Leader Constructive Thinking and Follower LMX 184
Method .................................................................................................................. 192
Participants........................................................................................................ 192
Measures - Leader ............................................................................................. 193
Measures - Follower ......................................................................................... 194
Procedure & Analytical Approach .................................................................... 196
Results .................................................................................................................. 197
Hypothesis Testing ........................................................................................... 198
Discussion ............................................................................................................ 212
Leaders’ information processing and followers’ LMX .................................... 214
Leaders’ constructive thinking and followers’ LMX ....................................... 216
Other comments and practical implications...................................................... 222
Limitations ........................................................................................................ 225
x
CHAPTER 6 Followers’ Information Processing Style and Constructive
Thinking as Leader LMX Antecedents: The Moderating Influence of Leaders’
Implicit Ideal Followership Theory ...................................................................... 227
What Resources Can Leaders Receive From Followers? .................................... 228
Hypothesis development ...................................................................................... 229
Followers’ Information Processing and Leaders’ LMX ................................... 229
Followers’ Constructive Thinking and Leaders’ LMX .................................... 233
Method .................................................................................................................. 238
Participants........................................................................................................ 238
Measures - Follower ......................................................................................... 239
Measures - Leader ............................................................................................. 240
Procedure & Analytical Approach .................................................................... 241
Results .................................................................................................................. 242
Hypothesis Testing ........................................................................................... 243
Discussion ............................................................................................................ 253
Followers’ Information Processing and Leaders’ LMX ................................... 254
Followers’ Constructive Thinking and Leaders’ LMX .................................... 258
Concluding Remarks and Practical Implications .............................................. 263
Limitations ........................................................................................................ 265
CHAPTER 7 Review of Studies and Conclusions ............................................... 266
Studies 1 and 2 – General Findings and Conclusions .......................................... 268
Studies 3 and 4 – General Findings and Conclusions .......................................... 270
Other Theoretical Implications ............................................................................. 275
General Research Strengths and Limitations ....................................................... 277
Practical Implications ........................................................................................... 280
Concluding Remarks ............................................................................................ 284
References ............................................................................................................ 286
APPENDICES ........................................................................................................ 360
APPENDIX A Rational Experiential Multimodal Inventory (Norris & Epstein,
2011) ..................................................................................................................... 361
APPENDIX B ILT Questionnaire (Offerman, Kennedy & Wirtz, 1994) ............ 363
APPENDIX C IFT Questionnaire (Sy, 2010) ...................................................... 365
xi
APPENDIX D Leader-Member Exchange (LMX-MDM; Liden & Maslyn, 1998)
.............................................................................................................................. 366
APPENDIX E Exploratory Analyses of Relationships between Follower
Information Processing Style and Constructive Thinking with Implicit
Followership Theories .......................................................................................... 367
APPENDIX F Exploratory Analyses of Relationships between Leader Naïve-
Optimism Sub-scales and Follower LMX ............................................................ 369
xii
List of Figures
Figure 1.1. Destructive and Constructive Pathways of Automatic Thinking (adapted
from Epstein, 1998). ................................................................................................... 20
Figure 2.1. Research model integrating Cognitive Experiential Self Theory, Implicit
Leadership Theory, Implicit Followership Theory and Leader-Member Exchange
(LMX). ....................................................................................................................... 76
Figures 5.1a. and 5.1b. Graphical Representation of Proposed Moderation Models.
.................................................................................................................................. 181
Figure 5.2. Moderating effect of follower sensitivity IILT (1±SD) on the relationship
between leader imagination (1±SD) and follower LMX. ......................................... 204
Figure 5.3. Moderating effect of follower sensitivity IILT (1±SD) on the relationship
between leader emotional coping (1±SD) and follower LMX. ................................ 205
Figure 5.4. Moderating effect of follower dedication IILT (1±SD) on the
relationship between leader behavioural coping (1±SD) and follower LMX. ......... 207
Figure 5.5. Moderating effect of follower sensitivity IILT (1±SD) on the
relationship between leader behavioural coping (1±SD) and follower LMX. ......... 208
Figure 5.6. Moderating effect of follower sensitivity IILT (1±SD) on the
relationship between leader categorical thinking (1±SD) and follower LMX. ........ 210
Figure 5.7. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader naïve-optimism (1±SD) and follower LMX. ............... 211
Figure 5.8. Moderating effect of follower dedication IILT (1±SD) on the
relationship between leader naïve-optimism (1±SD) and follower LMX. ............... 212
Figure 6.1. Moderating effect of leader committed IIFT (1±SD) on the relationship
between follower imagination (1±SD) and leader LMX. ......................................... 249
Figure 6.2. Moderating effect of leader conformity IIFT (1±SD) on the relationship
between follower emotional coping (1±SD) and leader LMX. ................................ 250
Figure 6.3. Moderating effect of leader insubordination IIFT (1±SD) on the
relationship between follower categorical thinking (1±SD) and leader LMX. ........ 253
Figure 7.1. General research model tested in studies 1 to 4. ................................... 267
Figure 8.1. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader over-optimism (1±SD) and follower LMX. ................ 370
Figure 8.2. Moderating effect of follower dedication IILT (1±SD) on the
relationship between leader over-optimism (1±SD) and follower LMX. ................ 371
xiii
Figure 8.3. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader stereotypical thinking (1±SD) and follower LMX. ..... 373
Figure 8.4. Moderating effect of follower dedication IILT (1±SD) on the
relationship between leader stereotypical thinking (1±SD) and follower LMX. ..... 374
Figure 8.5. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader pollyann-ish thinking (1±SD) and follower LMX. ..... 375
Figure 8.6. Moderating effect of follower dedication IILT (1±SD) on the
relationship between leader pollyann-ish thinking (1±SD) and follower LMX. ..... 377
xiv
List of Tables
Table 1.1. Summary of Associative vs. Rule-Based Processing ................................. 8
Table 1.2. Basic Needs and Associated Beliefs Posited in Cognitive-Experiential
Self-Theory (adapted from Epstein, 2003)................................................................. 14
Table 1.3. Constructive Thinking Inventory Scales and Sub-Scales with associated
Definitions - Adapted from Epstein (2001) ............................................................... 18
Table 3.1. Direct Oblimin Rotated Pattern Matrix of the 22-Item Adapted ILT
Measure (n = 311) ...................................................................................................... 98
Table 3.2. Means, Scale Reliabilities, and Inter-correlations of Scales on 22 Item
Adapted ILT Measure ................................................................................................ 99
Table 3.3. Descriptive statistics on REIm factors for overall group and as a
factor of gender ........................................................................................................ 101
Table 3.4. Correlations between REIm (n = 311), CTI (n = 273) and ILT
Dimensions ............................................................................................................... 103
Table 3.5. Unstandardised (B) and Standardised (β) Regression Coefficients, and
Squared Semi-Partial Correlations (sr2) For Each Predictor Predicting Sensitivity
Trait Preference ........................................................................................................ 106
Table 4.1. Direct Oblimin Rotated Pattern Matrix of the 14-Item Adapted IFT
Measure (n = 133) .................................................................................................... 143
Table 4.2. Means, Scale Reliabilities, and Inter-correlations of Scales on 18 Item
Adapted IFT Measure .............................................................................................. 144
Table 4.3. Descriptive statistics on REIm dimensions for overall group and as a
factor of gender ........................................................................................................ 146
Table 4.4. Correlations between REIm (n = 133), CTI (n = 118) and each IFT
Dimension ................................................................................................................ 148
Table 5.1. Implicit Ideal Leadership Theory Dimensions and Items from Exploratory
Factor Analysis conducted in Chapter 3 .................................................................. 182
Table 5.2. Descriptive statistics, inter-correlations and scale reliabilities for all
manager (n = 26) and follower (n = 106%) main variables used in the study. ......... 201
Table 5.3. Hypotheses 1 and 1a - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ rational processing and followers’ intelligence as
moderator ................................................................................................................. 202
xv
Table 5.4. Hypotheses 2 and 2a - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ experiential processing and follower sensitivity as
moderator ................................................................................................................. 202
Table 5.5. Exploratory Hierarchical linear modelling for interaction between
leaders’ imagination and followers’ sensitivity for predicting followers' LMX ...... 203
Table 5.6. Hypotheses 3 and 3a - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ emotional coping and followers’ sensitivity as
moderator ................................................................................................................. 205
Table 5.7. Hypotheses 4 and 4a - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ behavioural coping and followers’ dedication as
moderator ................................................................................................................. 206
Table 5.8. Hypothesis 4b - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ behavioural coping and followers’ sensitivity as
moderator ................................................................................................................. 208
Table 5.9. Hypotheses 5 and 5a - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ categorical thinking and followers’ sensitivity as
moderator ................................................................................................................. 209
Table 5.10. Hypotheses 6 and 6a - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ naïve-optimism and followers’ attractiveness as
moderator ................................................................................................................. 210
Table 5.11. Hypothesis 6b - Hierarchical linear modelling results predicting
followers’ LMX with leaders’ naïve-optimism and followers’ dedication as
moderator ................................................................................................................. 212
Table 6.1. IIFT Factors and Items from EFA conducted in Chapter 4 ............... 229
Table 6.2. Descriptive statistics, inter-correlations and scale reliabilities for all
manager (n = 18) and follower (n = 77) main variables. ......................................... 244
Table 6.3. Hypotheses 1 and 1a - Hierarchical linear modelling results predicting
leaders’ LMX with followers’ rational processing and leaders’ conformity IIFT as
moderator ................................................................................................................. 246
Table 6.4. Hypotheses 2 and 2a - Hierarchical linear modelling results predicting
leaders’ LMX with followers’ experiential processing and leaders’ insubordination
IIFT as moderator ..................................................................................................... 246
xvi
Table 6.5. Hypothesis 2b - Hierarchical linear modelling results predicting leaders’
LMX with followers’ experiential processing and leaders’ committed IIFT as
moderator ................................................................................................................. 247
Table 6.6. Hypothesis 2b - Hierarchical linear modelling results predicting leaders’
LMX with followers’ imagination and leaders’ insubordination IIFT as moderator
.................................................................................................................................. 248
Table 6.7. Hypothesis 2b - Hierarchical linear modelling results predicting leaders’
LMX with followers’ imagination and leaders’ committed IIFT as moderator....... 248
Table 6.8. Hypotheses 3 and 3a - Hierarchical linear modelling results predicting
leaders’ LMX with followers’ emotional coping and leaders’ conformity IIFT as
moderator ................................................................................................................. 250
Table 6.9. Hypotheses 4 and 4a - Hierarchical linear modelling results predicting
leader LMX with Follower Behavioural Coping and Leader Enthusiasm IIFT as
Moderator ................................................................................................................. 251
Table 6.10. Hypothesis 4b - Hierarchical linear modelling results predicting leader
LMX with Follower Behavioural Coping and Leader Committed IIFT as Moderator
.................................................................................................................................. 251
Table 6.11. Hypotheses 5 and 5a - Hierarchical linear modelling results predicting
leaders’ LMX with followers’ categorical thinking and leaders’ insubordination IIFT
as moderator ............................................................................................................. 252
Table 8.1. Exploratory Correlations between REIm and Five ILT Factors ............ 367
Table 8.2. Exploratory Correlations between CTI and Five ILT Factors ............... 368
Table 8.3. Exploratory hierarchical linear modelling results predicting follower
LMX with Leader over-optimism and Follower Attractiveness as Moderator ........ 370
Table 8.4. Exploratory hierarchical linear modelling results predicting follower
LMX with Leader over-optimism and Follower Dedication as Moderator ............. 371
Table 8.5. Exploratory hierarchical linear modelling results predicting follower
LMX with Leader stereotypical thinking and Follower Attractiveness as Moderator
.................................................................................................................................. 372
Table 8.6. Exploratory hierarchical linear modelling results predicting follower
LMX with Leader stereotypical thinking and Follower Dedication as Moderator .. 374
xvii
Table 8.7. Exploratory hierarchical linear modelling results predicting follower
LMX with Leader pollyann-ish thinking and Follower Attractiveness as Moderator
.................................................................................................................................. 375
Table 8.8. Exploratory hierarchical linear modelling results predicting follower
LMX with Leader pollyann-ish thinking and Follower Dedication as Moderator .. 377
1
CHAPTER 1
Introduction and Thesis Overview
Leadership is a feature inherent in all facets of society including businesses,
politics and sports teams (Brown, 1991; van Vugt, Hogan, & Kaiser, 2008) and has
important consequences for individuals, teams, organisations, countries, economies,
and the world community (Dionne, Yammarino, Atwater, & Spangler, 2004;
Dulebohn, Bommer, Liden, Brouer, & Ferris, 2012; Ikenberry, 1996; Lowe, Kroeck,
& Sivasubramaniam, 1996). For example, leadership can affect an individual’s
satisfaction, an organisation’s financial performance, or even if a country goes to
war. Leadership, however, has traditionally been confused with the leader
themselves rather than as a process which is co-created by both leaders and followers
through their relationship (Lord & Brown, 2001; Shamir, 2007). Accordingly, the
traditional focus of leadership research has been on the leader while the follower is
seen a passive receptacle to the effects of the “heroic” leader (Goffee & Jones, 2001;
Lord & Maher, 1993). Followers, however, are not passive individuals (Lord &
Brown, 2001; Oc & Bashshur, 2013). Like the leader, they play active roles in team,
group and organisational successes and failures (Bennis, 2008; Chaleff, 1995; 2008;
2009; Kelley, 1988; 2008). Consequently, there have been calls for research to take a
more holistic or integrated approach to understanding leadership that incorporates
both the leader and the follower (e.g. Shamir, 2007; Uhl-Bien, Riggio, Lowe, &
Carsten, 2014).
By adopting an integrated/holistic model of leadership, which draws and
expands upon existing relational perspectives of leadership, it was recognised that
two research gaps in the leadership/followership literature could be addressed
2
concurrently. The first gap addressed was the lack of theory and research on the
antecedents of leaders and followers’ implicit leadership and followership theories.
These essentially are the beliefs held in regards to leaders and followers respectively,
such as what they look, sound, and act like (see Lord & Maher, 1990; 1993; Sy,
2010). Second, this research sought to address the relative lack of research on the
antecedents of leaders and followers’ working relationship perceptions (Mahsud,
Yukl, & Prussia, 2010) as conceptualised in leader-member exchange theory
(Dansereau, Graen, & Haga, 1975; Graen, Dansereau, Minami, & Cashman, 1973;
Graen & Uhl-Bien, 1995; Hollander, 1971; 1992; Shamir, 2007). Specifically, do
leaders and followers’ cognitive preferences or tendencies in the form of information
processing styles and constructive thinking relate to their dyadic partners perceptions
of the leader-follower relationship.
Practically, taking a holistic outlook could provide new ways to understand
and approach organisational leadership and its development, in addition to informing
team composition and employee selection/recruitment. In particular, leadership
development programs could benefit by recognising and incorporating the active role
of the follower. Specifically, the predominant approach in most current mainstream
leadership programs is to focus on the leader with the goal of increasing their
knowledge, skills and abilities (Day, 2001; Velsor & McCauley, 2010). Such
programs have traditionally been marketed as leadership development programs but
could be more accurately considered as leader development programs (Day, 2001).
Consequently, mainstream leadership development approaches tend to fail to capture
the followers’ active and dynamic impact on the leadership process. A holistic
approach though, which incorporates the personal preferences and beliefs of the
individuals involved (i.e. the leader and follower), could provide new insights and
3
avenues to improving leadership outcomes. For example, instead of seeking to
mould a leader to a particular model like current mainstream approaches tend to do,
a holistic approach would look to develop both the leader and follower to be more
aware of each other’s tendencies and beliefs and how these influence the leadership
process. By consequence, this approach would not only seek to improve leaders’
knowledge, skills and abilities but also followers with the goal to bring about more
effective leadership and followership.
Positioning the Research’s Theoretical Approach
Conceptualisations that focus on the leader-follower relationship appear best
aligned addressing the research requirement to take a more holistic or integrated
approach to leadership. This is because leadership is typically defined in this sub-
area of leadership research as a dynamic process, which is mutually influenced by
both leaders and followers (e.g. see Dansereau et al., 1975; Graen et al., 1973; Graen
& Uhl-Bien, 1995; Hollander, 1971; 1992; Shamir, 2007). In particular, leader-
member exchange (Dansereau et al., 1975; Graen et al., 1973; Graen & Uhl-Bien,
1995) is a social exchange perspective of leadership as it assumes a give-and-take of
valued emotional and tangible resources between a leader and follower (Blau, 1964;
Graen & Scandura, 1987; Liden, Sparrowe, & Wayne, 1997; Sparrowe & Liden,
1997). Importantly, it explicitly recognises that followers are a heterogeneous group
who can perceive, interpret and react in very different ways to the same things and
that leaders can act in different ways with each follower (Dansereau et al., 1975;
Graen et al., 1973; Northouse, 2007). This understanding appears to be well suited to
conceptualising followers as active in the leadership process (see Shamir, 2007).
4
Consequently, the view of leadership throughout this dissertation is relationship
based.
There are, however, issues with some elements of LMX theory and how it is
typically operationalised in research. Specifically, we cannot assume that both
leaders and followers will perceive the nature of the relationship in the same way as
research demonstrates low correlations between leader and follower ratings of the
relationship (see Gerstner & Day, 1997; Sin, Nahrgang, & Morgeson, 2009; van
Gils, van Quaquebeke, & van Knippenberg, 2010). Yet, research has typically only
measured the relationship from the perspective of the follower on the assumption
that it will be experienced by the leader in a similar way (Wilson, Sin, & Conlon,
2010; van Gils et al., 2010). Consequently, we need to reconceptualise this
assumption of LMX to better understand what may be contributing to these
differences in ratings. The internal worlds of leaders and followers were theorised to
be important for this.
Like their respective roles in the observable world, leaders and followers
have just as active and dynamic internal worlds (e.g. see Oglensky, 1995; Stech,
2007). These internal worlds exist within all humans and influence our perceptions,
evaluations, feelings and behaviours (Epstein, 1994; 1998; 2003; 2014; Norris &
Epstein, 2011; Pacini & Epstein, 1999a; 1999b). Indeed, psychodynamic theories of
leadership (e.g. see Bornstein, 2003; Stech, 2007) and followership (e.g. see
Oglensky, 1995; Zaleznik, 1965) are focused on understanding the consistent
patterns in the way leaders and followers think, feel and act towards the world and
others and how these patterns impact on relationships at work (Kets de Vries, 2004;
Kroeger, Thuesen, & Rutledge, 2002; Shamir, 1991; Stech, 2007). There is,
5
therefore, precedence to believe that the internal world of leaders and followers will
influence the nature of the leader-follower relationship.
The ideas within the psychodynamic literature (e.g. see Bornstein, 2003;
Oglensky, 1995; Stech, 2007) appear to be quite useful for understanding leader-
follower relationship dynamics (Stech, 2007). However, they are limited in terms of
their scientific defensibility (Epstein, 2003) and their useability when engaging in
leadership training and development (Stech, 2007). Consequently, there existed a
need in the development of the current research to identify a model that would
conceptualise the internal worlds of leaders and followers in a more scientifically
defensible way while still being compatible with the ideas in the psychodynamic
literature. Dual processing theories seemed to provide an avenue for this.
Conceptualising the Internal World of
Leaders and Followers: Dual-Processing Models
There has been a long recognition in psychology of a “duality” in the ways
human process information (Moskowitz, Skurnik, & Galinsky, 1999). Indeed
William James over 120 years ago was amongst the first to propose that humans’
process information via two different systems, which he called associative thought
and true reasoning (Sloman, 1996). Since this time, there have been a significant
number of dual-system and dual-processing models described in the literature (e.g.
Epstein, 1994; 2003; Kahneman & Frederick 2002; Sloman, 1996; Smith &
DeCoster, 2000; Strack & Deutsch 2004). Earlier models tended to be used to
explain a variety of processes including persuasion (e.g. see elaboration likelihood
model; Petty & Cacioppo, 1986), social cognition, judgement and behaviour (e.g. see
6
Bargh, 2006; Epstein, 1994; Strack & Deutsch, 2004), attitude-behaviour relations
(e.g. Fazio, 1990) and stereotyping/prejudice (e.g. Devine. 1989).
Each of these models generally propose that cognitive processes have two
major information processing systems, each of which have different capacities and
ways of processing information (Evans, 2008; Evans & Over, 1996; Gawronski &
Creighton, 2013; Kahneman, 2011; Kahneman & Frederick 2002). The first system
is typically seen as having a high capacity and processes information implicitly or
automatically at the unconscious level. In contrast, the second system is lower in its
capacity and is engaged in more explicit or controlled processes. These systems have
been labelled differently (e.g. Epstein, 2003; Nisbett, Peng, Choi, & Norenzayan,
2001; Schneider & Schiffrin, 1977). For example, Hammond (1996) labels these two
system as Intuitive and Analytic whilst Strack and Deutsch (2004) refer to them as
Impulsive and Reflective. Kahneman (2011) refers to them simply as System 1 and
System 2.
Detailed reviews and critiques of specific dual-processing and dual-systems
models have been presented elsewhere (e.g. see Chaiken & Trope 1999; Frankish &
Evans, 2009; Gawronski & Creighton, 2013; Smith & DeCoster, 2000).
Consequently, a more general review of the key dual-processing theories is presented
here, which focuses on the generalised dual-processing theories as opposed to the
more formalised dual-processing theories (see Gawronski & Creighton, 2013 for
more detail about this distinction). Given the complex nature of leader-follower
relationships, generalised models seemed more relevant as they explain a large
variety of phenomenon (Gawronski & Creighton, 2013). Dual-process models
reviewed include Smith and DeCoster’s (2000) Associative vs. Rule based
processing, Kahneman’s (2003) System 1 and System 2 processing, Strack and
7
Deutsch’s (2004) reflective and impulsive model, and Epstein’s (Epstein, 1987;
1994; 2003) Cognitive-Experiential Self-Theory. An in-depth focus though is
provided on the latter as it was ultimately integrated into the four studies presented in
this dissertation due to its fundamental and explicit links with psychodynamic
theories, its integration with affect (of which is a key area of study within the
leadership literature), and the availability of reliable and valid measurement tools
specifically aligned with the CEST theory (see Epstein, 2001; Norris & Epstein,
2011).
Smith and DeCoster’s (2000) Associative vs. Rule Based Processing
Combining elements of Sloman’s (1996) associative vs. rule based
processing with theory on fast vs. slow learning memory systems (see McClelland,
McNaughton, & O’Reilly, 1995), Smith and DeCoster (2000) developed a dual-
process model to explain how people make judgments and decisions. They argued
that humans’ have two distinct processing modes: associative processing, which is
aligned with the properties of a slow-learning memory system; and rule-based
processing, which is aligned with a fast-learning memory system.
According to Smith and DeCoster (2000) associative processing consists of a
pattern-completion or similarity-based retrieval mechanism. Specifically, prevalent
cues from a current object or event are drawn upon to retrieve representations from
memory that contained similar cues. These representations in memory exist through
simple associations, which have been slowly learned over time via multiple
experiences (Gawronski & Creighton, 2013). Such processing is automatic and
outside of conscious awareness (Smith & DeCoster, 2000). Through this processing,
8
stored information from previous experiences can be used to fill in missing
information regarding the current object or event both quickly and automatically.
Rule-based processing, in contrast, uses “symbolically represented and
intentionally accessed knowledge as rules to guide processing” (Smith & DeCoster,
2000, p.111). These rules are quickly learnt, requiring only one or a few experiences.
Unlike associative processing, which occurs in parallel, rule-based processing is
sequential as only one rule can be applied to processing at a time. As a consequence,
rule-based processing is effortful and will be undertaken only where capacity and
motivation exist (Smith & DeCoster, 2000). However, individuals will be conscious
of the processes and outcomes via rule-based processing. A summary of the
theoretical properties of these two modes of processing are shown in Table 1.1.
Table 1.1. Summary of Associative vs. Rule-Based Processing
Associative Processing Rule-Based Processing
Draws on associations Draws on symbolically represented rules
Associations are based upon similarity and
contiguity
Rules are based upon language and logic
Associations take a long-time to develop and
requires significant amounts of experience
Symbolic knowledge is quickly learned,
requiring only a single or few experiences
Processing occurs automatically Processing are controlled, if both capacity
and motivation are present.
Processing is quick and in parallel Processing is slow and sequential
Preconscious; aware only of outcome of
processing
Conscious; aware of steps of processing as
well at its outcomes
Adapted from Smith and DeCoster (2000)
Kahneman’s (2003) System 1 and System 2 processing
In an attempt to integrate earlier findings around heuristics and bias,
Kahneman (2003) conceptualised two systems labelled System 1 and System 2 in
recognition of their neutrality (see Stanovich & West, 2000). System 1 is “fast,
automatic, effortless, associative, and often emotionally charged” (Kahneman, 2003,
p.1451). However, information in this system is slow to change. In contrast, System
2 is significantly slower, controlled, effortful rule-governed, but flexible and quick to
9
change. Kahneman (2003) distinguishes a perceptual system as well, which shares
similar processes to that of System 1. Where it differs though is in terms of content
that is processed; specifically that System 1 and 2 are “not restricted to the
processing of current stimulation” (Kahneman, 2002, p.450-451). Within
Kahneman’s (2002; 2003) model, intuitive impressions of objects are the output of
System 1. In contrast, explicit and intentional judgments are the product of System 2.
An important consequence of this distinction is that System 2 monitors all
judgments, even if they are the result of the impressions and processes occurring in
System 1. Consequently, System 2 has important implications for mental responses
(e.g. allowing System 1 response to occur; moderating the system 1 response;
correcting the System 1 response; or negating the response (Gawronski & Creighton,
2013). Accessibility (which incorporates stimulus salience, selective attention, and
cognitive priming) of mental content is important for the response generation of
System 1, whereas in contrast, system 2 applies logical rules to guide itself to an
outcome (Higgins, 1996; Kahneman, 2002). What is most accessible at any one
moment is determined primarily by the properties of the objective currently being
attended to.
Strack and Deutsch’s (2004) Reflective-Impulsive Model (RIM)
Stack and Deutsch (2004) proposed the Reflective-Impulsive Model (RIM)
of social behaviour. In this model, social behaviour is assumed to be jointly
influenced by two systems referred to as the Reflective System (RS) and the
Impulsive System (IS). Both systems operate simultaneously and interact with each
other, although the IS assumes the primary role due to the RS being limited in
10
capacity (Gawronski & Creighton, 2013). The IS system is seen as a simple,
associative learning system, which contains all ones implicit knowledge (Stack &
Deutsch, 2004). Links between knowledge, or elements, in the IS system are formed
and activated according to similarity and contiguity principles in addition to one
motivations (Gawronski & Creighton, 2013). Behavioural outcomes in this system
are due to a process of spreading activation (via either input from the environment or
through rumination) of elements via their linkages (Stack & Deutsch, 2004). In
contrast, the relationships between the elements in the RS differ. Specifically, the IS
connects elements via propositional representations, through the use of truth of false
flags (Gawronski & Creighton, 2013; Stack & Deutsch, 2004). This facilitates
thinking about future possibilities and comprehends when something is not the case,
both of which are something that IS system cannot undertake (Gawronski &
Creighton, 2013; Stack & Deutsch, 2004). Behaviour in the RS is a consequence of
the values and probabilities associated with various outcomes (Stack &
Deutsch,2004). These outcomes are evaluated and a decision for behavioural action
is made.
RIM states that behaviour is influenced by both the RS and IS via a common
pathway whereby behavioural schemata are activated (Stack & Deutsch, 2004).
These are activated via either a direct process (i.e. spreading activation through the
IS system) or by an indirect process (i.e. via behavioural intentions through the RS).
However, given these divergent pathways, it is possible that the behavioural
schemata that are activated will not be compatible. According to RIM, it is the
strongest activated scheme that will be activated. In this regards, RIM is a parallel-
competitive dual-processing model (see Evans 2007; 2008).
11
Epstein’s Cognitive Experiential Self-theory
Similar to the other dual-processing models described above, CEST proposes
that humans process information through two independent information-processing
systems: 1) a conscious rational system; and 2) a preconscious experiential system
(Epstein, 1994; 2003; 2014). However, where CEST differs is that the dual-
processing model is placed in the wider context of a global theory of personality
(Epstein, 2003; Pacini & Epstein, 1999a). Consequently, the nature of dual
processing is not seen in isolation from other individual factors and a more holistic
understanding of individuals can be provided. In addition, compared to other dual
processing theories, it has been explicitly influenced by psychodynamic concepts,
but with the goal of overcoming their scientific indefensibility (Epstein, 1994;
Frankish & Evans, 2009). Indeed, Frankish and Evans (2009) referred to this
integration of the Freudian ideas with the cognitive unconscious “a landmark
development” (p.19).
As a result of this integration, emotions also are explicitly recognised as vital
in human processing, a feature that is not typical in other dual processing theories
(Evans, 2008). Emotions have been increasingly recognised as important in the
workplace (e.g. Bono, Foldes, Vinson, & Muros, 2007; Brief & Weiss, 2002).
Consequently, their explicit recognition in CEST would mean that, as a theory, it
would be aligned with this emerging direction in the leadership field. Finally, from
an operationalisation perspective, a number of reliable and valid measurement tools
aligned with the CEST theory have been developed (see Epstein, 2001; Norris &
Epstein, 2011). Given these fundamental strengths, it appeared that CEST may be
well suited as a theoretical framework for understanding the internal worlds of
leaders and followers for the current research.
12
The Rational System. According to CEST, each the rational and experiential
systems operate by a set of different rules. The rational system is analytical, logical,
deliberate, process oriented, is slow to process, but updates quickly when new
information is made available, encodes in abstract symbols, words, and numbers, and
demands high levels of cognitive resources (Epstein, 2003). Unlike the experiential
system, the rational system operates mostly as the conscious level (Epstein, Pacini,
Denes-Raj, & Heier, 1996). However, as the processing in the rational system is
effortful and demands high levels of cognitive resources, it is not well suited for
processing large amounts of information in an effective manner (Sadler-Smith &
Shefy, 2004). Indeed, Epstein (2003) states, “Even mundane activities such as
crossing the street would be excessively burdensome if you had to rely exclusively
on conscious reasoning” (p. 161). In such activities or situations, humans mostly rely
on the experiential system.
The Experiential System. The experiential system is holistic, rapid,
outcome oriented, automatic, changes slowly despite evidence to the contrary,
encodes in concrete images, metaphors and narratives, and demands little in terms of
cognitive resources (Epstein, 2003; Epstein, et al., 1996). It learns from experience,
especially those that are emotionally arousing, as opposed to the rational system,
which learns by logical inference (Epstein, 2003; Epstein, et al., 1996). Importantly,
unlike the rational system, the experiential system is closely linked with affect.
Affect drives the learning and schema development in the experiential system and is
what underpins an individual’s subsequent motivations and behaviours (e.g. to
maximise pleasure and avoid or minimise pain; Epstein, 2003). As noted above, the
13
experiential system though is a preconscious system as this processing mostly occurs
outside of an individual’s awareness (Epstein et al., 1996).
The Interactive Nature of the Rational and Experiential Systems. While
the rational and experiential systems are considered to be independent, they operate
in parallel and interact with each other (Epstein, 1998; 2014; Epstein et al., 1996). It
is the amalgamation of the processes between the two systems that influences an
individual’s behaviour. However, CEST goes further to argue that most human
behaviour is influenced by the experiential system (Epstein, 1994; 2003). Certainly,
the experiential system, unlike other theories of the unconscious, is seen as a largely
adaptive system rather than a source only for non-adaptive schemas (Epstein, 2003).
Specifically, CEST stipulates that the experiential system automatically encodes
experiences, organises them, and directs behaviour, all of which occurs continuously
outside of the individual’s awareness (Epstein, 1998; 2003). Most individuals would
be aware of their own rational system as it operates at the conscious level. However,
due to the predominately-unconscious nature of the experiential system, many
individuals are unaware of its influence on cognitive processing and behaviours.
Indeed, CEST assumes that the experiential system will come to influence the
processing that subsequently occurs at the conscious level (Epstein, 1998).
Therefore, the rational system can operate without knowing the influence of the
experiential system on its processing. Recognising that most of our behaviours are
influenced by the experiential system, therefore, places the cognitive unconscious at
an important level of helping to understand human behaviour (Epstein, 1994)
including the interactions that occur between leaders and followers.
14
Basic Needs and Beliefs of the Experiential System. Learning over time
influences the general properties of schemas within the experiential system (Epstein,
1994; 1998; 2003). The experiences throughout life, especially those that occur in
early childhood, serve to lay down core beliefs about the self, the world and one’s
place within it. Specifically, all individuals will automatically construct an overall
implicit theory of reality that includes a self-theory, a world-theory and connecting
propositions (Epstein, 1994; 2003). This implicit theory of reality results in a vast
conceptual structure of interrelated schemas with general, abstract schemas at the
broadest level that then funnels down to situation specific schemas at the other end
(Epstein, 2003).
At the top of this hierarchy, humans hold four basic beliefs: 1) the world as
benign vs. malevolent; 2) the world as meaningful vs. non-meaningful; 3) people as
helpful and trustworthy vs. dangerous and untrustworthy; and 4) the self as worthy
vs. unworthy. Individuals will vary on these beliefs due to differences in life
experiences and thus influence how they tend to respond to and act in the world.
These basic beliefs correspond to four basic needs as shown in Table 1.2 (Epstein,
2003). These needs are similar to other personality theories (e.g. psychodynamic,
object-relations, phenomenological perspectives, and self-enhancement) but in the
context of CEST, each of these needs is equally important (Epstein, 1994; 2003).
Table 1.2.
Basic Needs and Associated Beliefs Posited in Cognitive-Experiential Self-Theory
(adapted from Epstein, 2003)
Basic Need Associated Basic Belief
Maximise pleasure, minimise pain Benign vs. malevolent world
Stability World as meaningful vs. non-meaningful
Relatedness People as helpful & trustworthy vs.
dangerous and untrustworthy
Self-enhancement Self as worthy vs. unworthy.
15
As the basic beliefs are atop of an individual’s implicit theory of reality, it
results in them being very stable and difficult to invalidate (Epstein, 1998). Any
changes at this level would cause immense repercussions throughout an individual’s
implicit theory of reality. In contrast, specific situational schemas are easier to
change due to highly specific content. Changes at this level, therefore, do not
invalidate other parts of the system. Such schemas come to influence how we attend,
interpret, and interact with the world – “the schemas in the experiential system are
the ‘scaffolding’ of a person’s personality” (Epstein, 2001, p.1). Therefore, they
serve to influence the ways in which all individuals attend to the world and
assimilate information in a given moment. This, therefore, would include those
occurring in both leaders and followers during their interactions.
Rational vs. Experiential Information Processing Styles. The degree to
which either the rational or the experiential system is deployed depends on
situational demands, emotional involvement, and experience (Epstein, 2003; Epstein
et al., 1996). For current purposes, the more important aspect to recognise is that
individuals vary in the degree to which they use either the rational or the experiential
processing system (i.e. they have different preferences; Epstein 2003; Norris &
Epstein, 2011). These differences influence an individual’s perceptions, judgements
and subsequent decisions (Epstein et al., 1996). This would include those occurring
for both leaders and followers. Measures focused on differences in intuition have
typically captured at least part of individual differences in the processing occurring
in the experiential system. However, emotionality and imagination are also part of
experiential processing (Norris & Epstein, 2011). Individuals who are high in
emotionality favour the use of their emotions and experience them with greater
16
frequency and intensity. Individuals who possess high imagination scores are
appreciative of, and engage in, imagination, aesthetic productions, and imagery.
Each component of information processing has advantages and disadvantages
associated with it (Epstein, 2003) as demonstrated by links with a range of
theoretically relevant variables (e.g. see Epstein et al., 1996; Norris & Epstein, 2011;
Pacini & Epstein, 1999a). These are discussed in more detail in the pertinent sections
of this dissertation. For current purposes, it is important to recognise that these
differences are theorised to influence how leaders and followers process information.
Constructive Thinking. CEST also contends that the rational and
experiential systems have different natures of intelligence that correspond to the
processing rules of each (Epstein, 1992; 2001; 2003; 2014). The nature of
intelligence in the rational system is logical and analytical and can be assessed by
current intelligence measures (e.g. Wechsler Adult Intelligence Scale; Wechsler,
2008). The intelligence of the experiential system though is focused on the
adaptiveness of individual’s spontaneous thoughts referred to as constructive
thinking (Epstein, 1994; 2001). Specifically, constructive thinking refers to the
degree to which a person’s automatic thinking facilitates solving problems in
everyday life at a minimum cost in stress (Epstein, 1998; 2001). As operationalized
in the Constructive Thinking Inventory (CTI; Epstein, 2001), the constructive
components of the experiential system include global constructive thinking,
emotional coping, and behavioural coping. The destructive components are personal
superstitious thinking, categorical thinking, esoteric thinking, and naïve-optimism
(Epstein, 2001). With the exception of personal superstitious thinking, each of these
17
components is derived from other sub-scales. These are listed along with their
definitions in Table 1.3.
There is a chain of events underpinning constructive thinking as shown in
Figure 1.1 (see Epstein, 1998). As can be observed, an initial interpretation of an
event or situation occurs which subsequently influences the individual’s initial
emotional reaction. These initial interpretations and emotions cognitively frame and
influence secondary interpretations and emotional reactions to the event. The final
step in this are the resulting behaviours aligned to these preceding thoughts and
emotions. Importantly, this chain of events and the resultant behaviours can be either
destructive or constructive in terms of its pathway and all individuals have
tendencies to adopt one of these (Epstein, 1994; 1998). For example, the tendency to
perceive problems as challenges to be overcome rather than as threats and/or to view
issues positively, but not to an unrealistic degree, could be considered as moving
along the constructive pathway (Epstein, 2001). This chain occurs across all events,
including those involving followers’ interactions with leaders.
A variety of relationships between the different constructive thinking factors
and a variety of outcome variables exist including positive relationships to success in
work, love, social relationships, and negative relationships to emotional and physical
18
Table 1.3.
Constructive Thinking Inventory Scales and Sub-Scales with associated Definitions - Adapted from Epstein (2001)
Constructive Thinking Inventory
Scales and Sub-Scales
Definition
Global Constructive Thinking Overall composite score of other main scales except Esoteric Thinking. Good constructive thinkers tend
to score high on the emotional and behavioural main scales and low on the other scales. It is assumed
that higher scores indicate flexible thinkers who can adjust their behaviour appropriately depending on
the situation.
Emotional Coping Degree to which an individual’s automatic thoughts allow them to deal with frustration, failure and
disappointment without undue distress.
Self-Acceptance Degree to which an individual has positive self-esteem and an overall favourable attitude to oneself.
Absence of Negative
Overgeneralisation
Degree to which an individual avoids overgeneralizing unfavourable experiences.
Non-sensitivity Degree to which an individual is able to tolerate disappointment, rejection, criticism and disapproval; the
degree to which they are thick-skinned.
Absence of Dwelling Degree to which an individual does not obsess over negative events.
Behavioural Coping Degree to which an individual thinks in terms of engaging in effective action.
Positive Thinking Degree to which an individual sees ‘the silver lining’ in things; the degree to which they see the positive
in events.
Action Orientation Degree to which an individual thinks in terms of ways to facilitate effective action.
Conscientiousness Degree to which an individual thinks in terms of hard work, planning, and doing one’s best.
19
Table 1.3. Continued from previous page
Constructive Thinking Inventory
Scales and Sub-Scales
Definition
Personal Superstitious Thinking Degree to which an individual has developed private superstitions. For example, if something good
occurs, something bad will also occur to balance it out.
Categorical Thinking Degree to which an individual sees the world in categorical terms; that the world is black and white
without any grey. Degree also to which an individual does not trust and is intolerant of non-similar
others.
Polarised Thinking Degree to which an individual will make extreme judgements, ignoring the middle ground.
Distrust of Others Degree to which an individual regards others as untrustworthy and a potential threat.
Intolerance Degree to which an individual is intolerant, judgmental, and unforgiving.
Esoteric Thinking Degree to which an individual holds beliefs in the unusual and scientifically questionable phenomena
e.g. conventional superstitions, ghosts, and astrology.
Belief in the Unusual Degree to which an individual holds beliefs in the usual e.g. ghosts, mind reading, and clairvoyance.
Formal Superstitious Thinking Degree to which an individual holds beliefs in conventional superstitions e.g. astrology and good/bad
omens.
Naïve-Optimism Degree to which an individual thinks in simplistic, stereotyped ways and are unrealistically optimistic.
Over-Optimism Degree to which an individual thinks in naively positive ways and overgeneralize following favourable
outcomes.
Stereotypical Thinking Degree to which an individual thinks in stereotyped ways e.g. everyone should love their parents.
Pollyanna-ish Thinking Degree to which an individual thinks in simplistically favourable ways e.g. everyone is basically good at heart.
20
Figure 1.1. Destructive and Constructive Pathways of Automatic Thinking (adapted
from Epstein, 1998).
wellbeing, self-discipline problems, alcohol and drug problems (Epstein, 1992;
2001; Epstein & Katz, 1992; Epstein & Meier, 1989). Importantly, constructive
thinking is also associated with a variety of leadership and workplace outcomes. For
example, emotional and behaviour coping are positively related to self-reported
ratings of transformational leadership (Cerni, Curtis, & Colmar, 2008). High
constructive thinking in nurses has been found to be related to lower occupational
stress and higher job satisfaction and performance (Judge & Bono, 2001; Stacciarini
& Tróccoli, 2004). Green (as cited in Epstein, 2001) also found that good
constructive thinkers demonstrate less job-related stress and greater happiness
21
despite voluntarily taking on more workloads. Furthermore, Nadeau (as cited in
Epstein, 2001) found that behavioural coping was related to supervisory ratings of
work quantity while emotional coping was negatively associated with absenteeism
and externally produced stress impacts on work performance.
It is clear from the above that information processing style and constructive
thinking have significant influences on individuals’ behaviours across a wide range
of contexts and situations. Consequently, the stability of these personality constructs
should have a strong influence on the general behaviours that are manifested in
leader-follower relationships by both leaders and followers. This served as part of
the rationale for looking at how these elements may influence each other’s
perspective of the relationship on the assumption that these would affect a wide
range of leader and follower behaviours which would be observed and evaluated by
their dyadic partner (i.e. follower and leader respectively).
Implicit Leadership and Followership Theories
However, while leaders and followers may vary in their behaviours, which
the current study assumes to be heavily influenced by differences in their
information processing style and constructive thinking, theory and research suggests
that individuals will not interpret these behaviours in the same way (e.g. Eden &
Leviathan, 1975; 2005; Epitropaki & Martin, 2004; 2005; Lord, Foti, & de Vader,
1984: Lord & Maher, 1990; 1993; Lord, de Vader, & Alliger, 1986; Phillips & Lord,
1982; Poole, Gioia & Gray, 1989; Sy, 2010). Therefore, we need to understand what
may be contributing to these differences in interpretation. Implicit leadership and
followership theories appeared to offer an avenue for this.According to implicit
leadership and followership theories, individuals hold idiosyncratic beliefs about the
22
nature of leaders and followers (i.e. what they look like, sound like, and how they
should behave; see Lord & Maher, 1990; 1993; Eden & Leviathan, 1975; 2005; Sy,
2010). These beliefs serve to guide the observers’ interpretations of leader and
follower actions, what a perceiver attends to, what they encode, and the information
they retrieve when recalling leader or follower related information (Eden &
Leviathan, 1975; 2005; Lord & Alliger, 1985; Lord & Emrich, 2001; Lord & Maher,
1990; 1993; Sy, 2010).
Implicit leadership and followership theory are based on role theory (Lord &
Maher, 1990; 1993; Sy, 2010) which takes the perspective that the actions of
individuals in most social activities are born from the knowledge of socially defined
categories held by all to some degree (Ritzer, 2000). When a person holds a role
status in a given context (such as a mother, a police officer, a leader or follower), it
serves as a script to guide their behaviour. Such deployment of social categories in
everyday functioning facilitates predictability in the social world (Ritzer, 2000;
Weick, 1995). Leaders and followers, therefore, should draw upon their beliefs about
how the dyadic partner should act in their interactions which then should influence
how leaders and followers perceive the nature of the relationship (e.g. see van Gils et
al., 2010). Consequently, implicit leadership and followership theories were seen as
potential reasons that underpin differences in leaders’ and followers’ perceptions of
their relationships.
There is, though, limited research on what predicts differences in these
beliefs (c.f. Keller, 1999; Kruse & Sy, 2011; Schyns & Meindl, 2005). There is,
however, a reason to suspect that information processing and constructive thinking
should influence leaders and followers preferences for different traits and behaviours
in followers and leaders respectively. Specifically, connectionist models emphasise
23
that basic knowledge is stored in nodes each, of which, are connected to other nodes
(Hanges, Lord, & Dickson, 2000; Lord & Brown, 2001). Input from the environment
or other nodes activates or inhibits other nodes based on the weighting and type of
connections (i.e. inhibitory or activation). It is the pattern of node activation that
represents complex knowledge. This description is somewhat similar to the
description of the self-concept in the experiential system in that the latter also
suggests that the self-concept is seen as a vast conceptual structure of interrelated
schemas with general, abstract schemas at the broadest level which then funnels
down to situation specific schemas at the other end (see Epstein, 2003). Given that
constructive thinking and information processing influence cognitive-emotional
processing, it is then theoretically possible that these personality factors will be
related to leaders’ and followers’ preferences for particular traits and behaviours in
both followers and leaders respectively. This reasoning served as the basis to
consider the potential links between leaders’ and followers’ information processing
style and constructive thinking and their implicit leadership and followership
theories.
Integrating Leadership and Followership Theories
The preceding review strongly suggested a potential for integrating CEST
(Epstein, 2003; 2014), implicit leadership (Eden & Leviathan, 1975; 2005 Lord &
Maher, 1993) and followership theories (Sy, 2010) into a cohesive theoretical
framework. By assimilating these theories into such a framework, the research is,
therefore, addressing the call for a more holistic approach to understanding
leadership (e.g. Shamir, 2007) and that leadership outcomes are the result of leaders
and followers enacting their roles through their working relationship (see Lord &
24
Brown, 2001). Specifically, the four research studies presented in this dissertation
drew upon CEST (Epstein, 2003; 2014): (1) to help explain followers’ and leaders’
variations in their implicit leadership and followership theories respectively; and (2)
to help understand how implicit theories and CEST can be integrated together to help
explain the leaders’ and followers’ perceptions of their relationship (see Gerstner &
Day, 1997; Sin et al., 2009).
Organisation of This Dissertation
To help achieve these two goals, a literature review was undertaken which
resulted in the identification of several gaps (which have been briefly outlined in this
introductory chapter) and current directions in the leadership field (see Chapter 2).
An integrated research model was developed based on this review (see Chapter 2
also). This model was subsequently tested through four separate, but inter-related,
pieces of research (Chapters 3 to 6). This model integrated CEST (Epstein, 2003;
2014); implicit leadership and followership theories (Eden & Leviathan, 1975; 2005;
Lord & Maher, 1990; 1993; Sy, 2010) with Leader-Member Exchange (Dansereau et
al., 1975; Graen & Uhl-Bien, 1995; Northouse, 2007) to: a) to help explain followers
and leaders variations in their implicit leadership and followership theories
respectively (studies 1 and 2 respectively); and b) to help understand how followers’
implicit leadership theories and leaders’ implicit followership theories can be
integrated with information processing style and constructive thinking to help
explain the leader’s and follower’s perceptions of their relationship (studies 3 and 4).
25
CHAPTER 2
Literature Review and Research Model
In the previous chapter, an overview of cognitive experiential self-theory
(CEST) was presented (Epstein, 2003). A rationale was also presented for
considering the leader-follower relationship as the core focus for studying leadership
and that the nature of leader-follower relationships can be considered partly as a
product of the interacting patterns of thinking in both leaders and followers.
However, as leaders and followers are in fact social roles we must likewise consider
how these elements play out in the relationship. The current chapter reviews
leadership and followership definitions, theories and respective research that are
relevant to this rationale.
Several key issues were observed in this literature review. First, it cannot be
assumed leaders and followers experience their relationship similarly. Second, there
is limited research on the antecedents of implicit leadership and followership
theories despite the fact that these influence how individuals process information
about leaders and followers. Finally, more research is required to better understand
the antecedents of followers’ and leaders’ perceptions of the quality of their working
relationship (Mahsud, Yukl, & Prussia, 2010). It is argued though that these issues
can be overcome by integrating CEST, implicit leadership and followership theories,
and leader-member exchange (Dansereau et al., 1975; Graen & Uhl-Bien, 1995) into
an overarching general model to better understand leader and follower perceptions of
their relationship. Such integration aligns with more recent conceptualisations of
leadership that recognise that its outcomes are the result of the leader-follower
relationship which is jointly influenced by leader and follower beliefs and
behaviours (Lord & Brown, 2001). To this end, the chapter concludes with the
26
presentation of a research model that will guide the studies presented in chapters 3,
4, 5 and 6.
Leadership Definitions
Over 40 years ago Stogdill (1974) noted, “there are almost as many
definitions of leadership as there are persons who have attempted to define the
concept” (p. 259). This lack of a clear definition is perhaps the core reason why
understanding the nature of leadership remains elusive. However, three general types
of definitions in the leadership area have emerged: 1) leader-centric; 2) follower-
centric; and 3) relational definitions. Leader-centric definitions typically define
leadership as where the leader influences a follower to achieve some desired
outcome (Northouse, 2007; Uhl-Bien et al., 2014; Yukl, 1998). For example, Katz
and Kahn (1978) see the leadership process as going beyond simple positive and
negative reinforcement to encompass an “influence increment” (p. 528).
Alternatively, Northouse (2007) and Bass (1985) state that it is what the leader does
to influence their followers to focus their attention on the achievement of some goal
or end state. Adopting a leader-centric definition implies that leadership is something
that the leader does to the follower to bring about desired attitudes and behaviours in
them. Consequently, leaders are the causal agents of organisational performance
(Lord & Maher, 1993) while followers are relegated to the role of “an empty vessel
waiting to be led, or even transformed, by the leader” (Goffee & Jones, 2001, p.
148). In other words, followers are the recipients, or at best, moderators, of the
leaders’ influence (Shamir, 2007).
27
In recognition that leader-centric definitions miss the active role of the
follower (Chaleff, 2008; Kelley, 2008) another type of leadership definition which
has been adopted is the process of being perceived by others as a leader (e.g. Lord et
al., 1984; Lord & Maher, 1990; 1993). This moves the focus to the perceivers of
leaders. In particular, the emphasis is on the cognitive processes occurring when
thinking and interacting with the leader and how this influences the leadership
process (Eden & Leviatan, 1975; 2005; Lord & Brown, 2001).
Leadership has also been defined as occurring in the social and relational
interactions between people (Fairhurst & Uhl-Bien, 2012; Graen & Uhl-Bien, 1991;
1995; Hollander, 1971; 1992; 2012; Lord & Brown, 2001; Uhl-Bien & Graen, 1993).
Research in this tradition explicitly understands leadership as a multi-faceted and
dynamic process which relies not just on the influence of the leader on the follower,
but also the follower’s influence on the leader (Baker, Anthony, & Stites-Doe, 2015;
Lord & Brown, 2001; Oc & Bashshur, 2013). Definitions in this field, therefore, tend
to emphasise the active role that followers play in the leadership process and that
leadership itself is inherent to the social exchanges occurring between leaders and
followers.
The different definitions influence the types and focus of research undertaken
(Uhl-Bien et al., 2014). For example, the leader-centric definition lends itself more to
research questions focused on leaders’ behaviours and characteristics and how these
influence some desired outcome (see Gibb, 1947; Jenkins, 1947; Stogdill, 1948). The
relationship based definition, though, lends itself to questions targeted towards how
relationships develop, are maintained and how these relationships affect desired
outcomes (e.g. see Graen & Uhl-Bien, 1991; 1995; Hollander, 1971; 1992; Uhl-Bien
28
& Graen, 1993). The following leadership theory classification system is generally
aligned with these definitions.
Leadership Theory Classification1
Traditionally, leadership theories have been classified according to how
various leader characteristics (e.g. traits, behaviours, or styles) make them effective
or ineffective across different situations. However, Graen and Uhl-Bien (1995) argue
it is more beneficial to classify leadership theories into three general domains: (1)
those focused on the leader; (2) those focused on the follower; and (3) those focused
on the relationship2. Such a classification system helps to focus attention beyond the
traditional leader-centric theories to the other important leadership domains and how
they may work together. Bringing attention to these other domains is important for
current purposes, as a key aim of the author was to understand how different theories
across these domains may be integrated. However as will be observed, such a
classification also fails to recognise the followership process. This can be understood
as the opposite side of the coin to leadership.
Leader-Centric Theories of Leadership
This section reviews theories falling under the leader-centric domain of
leadership. As will be observed, followers at best are seen as moderators to the
effectiveness of the leader in these theories.
1 In general, the presentation of the theories reflects the historical timeline of their appearance in the
scientific literature with earlier theories presented first. 2 For simplicity, the classification system does not cover other domains such as the group, team, and
organisation and readers are directed to other sources for reviews of these types of leadership (e.g.
Northouse, 2007; Zaccaro, Rittman, & Marks, 2001).
29
Trait and “Great Man” theories. Trait theories of leadership argue that
effective leaders possess, or have inherited, specific core traits, or an integrated
pattern of traits, which distinguish them from others and which are applicable to
effective leadership across all contexts (Northouse, 2007; Segal, 1985; Stogdill,
1948; Zaccaro, Kemp, & Bader, 2004). Early studies sought to identify these
distinguishing features. Characteristics which were observed to predict effective
leaders included physicality (height, weight, appearance), speech fluency,
intelligence, higher levels of task/area knowledge, need for achievement and power,
emotional control, confidence, initiative, masculinity, dominance, extroversion,
confidence and social skills (see reviews by Gibb, 1947; Jenkins, 1947, Mann, 1959;
Stogdill, 1948).
Despite a wide range of traits identified, reviews of this research body
concluded there was a lack of evidence to support the idea that leader traits alone
could adequately explain (in)effective leadership (see Jenkins, 1947; Mann, 1959;
Stogdill, 1948). For example, Mann (1959) concluded that traits identified as the
most consistent predictors of effective leadership (intelligence, dominance,
personality adjustment) had weak relationships at best (Kenny & Zaccaro, 1983).
Similarly, Jenkins (1947) concluded that no single trait, or group of traits, had been
identified as consistently distinguishing a leader from a non-leader. Based on these
reviews, Stogdill (1948) and Mann (1959) stated that the traits and competencies of
an effective leader depend, to a large extent, on situational demands. They also
identified that a key weakness of the trait approach was that it ignored the
interactions between leaders and followers (Stogdill, 1974). Most important though
was that the trait approach failed to distinguish between the leader as a person and
leadership as a process (Calder, 1977). The result of these reviews was a departure
30
from a strict trait-based approach to looking at how situational and contextual factors
could influence leadership (see Leadership Style Theories) (Bass, 1990; Kenny &
Zaccaro, 1983; Mitchell, 1982; Yukl, 1998; Zaccaro, 2007).
During the 1980s though, arguments emerged that the departure from traits
was premature and that they are indeed important in leadership (Kenny & Zaccaro,
1983; Zaccaro, 2007). One reason is that, until recently, there was a lack of a
coherent personality framework (e.g. Big-Five Model of Personality1; see Goldberg,
1990), which may have contributed to the high variability observed in the early trait
research (Barrick & Mount, 1991; Judge, Bono, Illies, & Gerhardt, 2002). Another
reason for looking at leader traits is that they are important for leadership when taken
from the perspective of the follower (Lord et al., 1986). This reason is particularly
relevant for current purposes as it actively recognises followers in the leadership
process. More detail on the importance of leader traits, when considered from the
perspective of the follower, is presented later (see Implicit Leadership Theories sub-
title).
Leadership style theories. During the 1950s and 1960s, the trend in
leadership research moved away from the traits of leaders to their behaviours (Derue,
Nahrgang, Wellman, & Humphrey, 2011). The focus was to identify how leaders
behave and compare these to leadership outcomes (Yukl, 1998). The underlying
assumption was that a universal set of behaviours could explain effective leadership
across all situations and contexts (Fleishman, 1957; Northouse, 2007).
1 One of the potential issues with the five-factor model of personality is that it may be covering the
more conscious elements of personality rather than the unconscious elements (Epstein, 2003). It
therefore does not cover important personality characteristics pertaining to the experiential system.
31
Researchers at Ohio State University and Michigan University were among
the first to study leadership in this way (Fleishman, 1953; Fleishman & Peters, 1962;
Halpin & Winer, 1957; Hemphill & Coons, 1957). At Ohio State University,
researchers found that leaders’ behaviours could be best classified on two
independent factors: consideration and initiating structure (Fleishman & Peters,
1962). The former is defined in terms of a leader’s orientation toward a group’s
welfare such as addressing their feelings, showing appreciation, and being friendly
and approachable. The latter dimension is defined in terms of the extent to which
they delineate group roles, activities, set goals, and schedule work (Fleishman, 1957;
Fleishman & Peters, 1962; House, 1971; Judge, Piccolo, & IIies, 2004; Tracy, 1987).
Two similar factors - employee orientation and production orientation - were also
observed by researchers at Michigan University (Kahn, 1956; Katz & Kahn, 1990).
Employee orientation aligns with consideration while production orientation with
initiating structure. Unlike consideration and initiating structure, though, employee
orientation and production orientation were considered polar opposites (Northouse,
2007). However, after further research, they were later re-conceptualised as
independent.
Research has, in general, found relationships between the two core leadership
dimensions identified in the Ohio State University and Michigan University studies
and various effectiveness criteria (Korman, 1966). Initiating structure has
demonstrated positive relationships with various group cohesiveness variables (e.g.
Christner & Hemphill, 1955), group performance (e.g. Filley & House, 1969) and
satisfaction with leaders (e.g. Parker, 1963). Consideration has demonstrated
negative associations with grievances and employee turnover (Fleishman & Harris,
1962) and a positive relationship with individual satisfaction (Badin, 1974)
32
However, inconsistencies and contradictions with these relationships within
and across studies exist (Fleishman, Harris, & Burtt, 1955; House, 1971; House,
Filley, & Kerr, 1971; Larson, Hunt, & Osborn, 1974; Yukl, 1998). For example,
Filley and House (1969) found that low and unskilled workers tended to be less
satisfied with supervisors who were high on initiating structure behaviours. In more
skilled workforces, though, initiating structure has been positively associated with
satisfaction (House, 1971; House et al., 1971). Results similar to these led Korman
(1966) to question whether these leadership dimensions “have any predictive
significance at all” (p. 360). This conclusion echoed by Yukl (1998), also argued that
the only reliable finding was that leaders who are high on consideration tended to
have more satisfied followers.
Leader-Context Match Theories.
Recognising that the trait and behavioural factors were too simplistic in
explaining the complex phenomenon of leadership, research began to look at the
match between a leaders’ style and the context1 in which they are leading (Seyranian,
2009). Consequently, these theories are referred to here as leader-context match
theories. Four theories are reviewed here: 1) situational leadership theory
(Blanchard, 1985; Hersey & Blanchard, 1969); 2) path-goal theory (House, 1971;
1996; House & Mitchell, 1975); 3) Fiedler’s contingency theory of leadership
effectiveness (Fiedler, 1964; 1967) and; 4) cognitive resource theory of leadership
(Fiedler, 1986; Fiedler & Chemers, 1984).
1 The term context is used here in to refer to both the organisational context and the follower.
33
Situational leadership theory. Situational leadership theory places the
emphasis on the need for a leader to adapt their style to the developmental level of
the follower (Blanchard, 1985; Blanchard, Zigarmi, & Zigarmi, 1985; Hersey &
Blanchard, 1969; 1977; Northouse, 2007). Corresponding to the traditional
behavioural leadership categories of initiating structure and consideration, they argue
that leaders need to adapt their use of directive (or task behaviours) and supportive
(or relationship) behaviours based upon followers’ maturity levels (Graeff, 1983;
Northouse, 2007). Leaders can be high or low in these dimensions resulting in four
different leadership styles: directing style (high directive, low supportive); coaching
style (high directive, high supportive); supporting style (low directive, high
supportive); and delegating style (low directive, low supportive).
The choice of which style to deploy requires the leader to assess their
followers’ maturity level on two dimensions: psychological maturity and job
maturity. The former refers to the followers’ commitment, motivation and
willingness to accept responsibility while the latter refers to their experience,
knowledge and understanding of the task (Chemers, 1997). Followers can be high or
low on both dimensions resulting in four developmental categories in which they can
be placed: 1) willing and capable (high psychological and job maturity); 2) unwilling
but capable (low psychological maturity but high job maturity); 3) willing but not
capable (high psychological maturity but low job maturity) and; 4) unwilling and not
capable (low psychological and job maturity; Hersey & Blanchard, 1977). Based on
the developmental level of the follower, the model then prescribes a particular
leadership style for that type of follower. However, while this has intuitive appeal,
research generally fails to support the premise that such matching is required or
34
appropriate except for new employees for whom a directive style is found to increase
satisfaction (Fernandez & Vecchio, 1997; Vecchio, 1987).
Path-goal theory of leadership. Path-goal theory provides a model for
understanding the processes by which leaders motivate changes in followers’
psychological states to achieve some specified goal (Chemers, 1997; House, 1971;
1996; House & Mitchell, 1975). It was one of the first leadership theories to
explicitly emphasise leader-follower interactions (Chemers, 1997). Path-goal theory
states that followers assess the intrinsic and extrinsic value of achieving some
specified work goal in addition to the probability that current behaviours will result
in the achievement of this outcome (House, 1971). It is assumed that followers are
more motivated when there are higher perceived intrinsic and extrinsic values
attached to, and higher levels of expectancy in achieving that goal. Consequently,
effective leadership requires the leader to positively influence the followers’
perceived value attached to the outcomes and the probability of achieving them. This
includes increasing the payoff to achieving a work goal, communicating clearly the
pathway to the goal, and removing any obstacles to achieving it (House & Dessler,
1974).
Path-goal theory also describes what leadership style a leader should adopt
based on the needs of the follower and the situational demands (Northouse, 2007). It
specifies four leadership styles (1) directive; (2) achievement-oriented; (3)
participative; and (4) supportive (House, 1971; 1996; Martin, 2009). The choice of
leadership style depends upon a wide range of environmental (e.g. task structure,
authority system, work-group) and follower factors (e.g. personality and ability;
35
House, 1971; House & Mitchell, 1975). The number of possible contingency factors,
therefore, facilitates extensive potential predictions for which style to choose.
While much research on path-goal theory of leadership has focused on
environmental factors (e.g. see Dessler & Valenzi, 1977; Downey, Sheridan, &
Slocum, 1975; 1976; Keller, 1989; Schriesheim & Schriesheim, 1980), in particular
the degree of ambiguity in followers’ jobs (e.g. Downey et al., 1975; Schriesheim &
DeNisi, 1981), there has been relative neglect on the impact of follower
characteristics (Chemers, 1997). However, follower characteristics and needs
influence how they react to the different leadership styles. For example, followers
with high growth needs reported less positive experiences when the leader engaged
in structuring and directive behaviours even when the work-task was unstructured
(Griffin, 1980; 1981). In contrast, followers low in growth needs remained stable
across all conditions. It seems logical to conclude that any theoretical formulation of
leadership, therefore, needs to account for the needs and personalities of followers to
fully understand the impact of leader behaviours (see also Chemers, 1997).
Fiedler’s contingency theory of leadership effectiveness. The contingency
theory of leadership effectiveness (Fiedler, 1964; 1967) argues that leadership
success is dependent on the interaction between the leader’s behavioural style (i.e.
relationship or task style) and the characteristics of the work environment (Fiedler &
Chemers, 1984; Peters, Hartke, & Pohlmann, 1985). It is assumed that certain leader
styles, which are relatively fixed and dependent on leaders’ personality, are favoured
to suit particular situational demands (Peters et al., 1985).
36
Situations are defined in terms of three key factors: (1) leader-member
relations; (2) task-structure; and (3) position power (Northouse, 2007). Leader-
member relationship refers to the nature (positive vs. negative) of the leader-follower
relationship. Task structure (structured vs. unstructured) refers to the degree to which
the task is clearly stated and understood by those involved. Position power refers to
the strength or power (strong vs. weak) a leader has through their position to punish
(e.g. dismiss) or reward (e.g. bonuses, promotion) their direct reports. The
combination of these factors results in eight different situational types, known as
octants 1-8 (Northouse, 2007; Seyranian, 2009). These situations are further
classified as favourable, intermediate, or unfavourable. Each situation is assumed to
be best suited to leaders with a particular leadership style. For example, a
relationship focused leader would be most effective in situations where there are
good leader-member relations, but there is low task structure and they hold weak
positional power. Consequently, there are leadership styles that are mismatched to
the situation (Fiedler & Chemers, 1984).
The propositions of the contingency theory of leadership effectiveness have
been generally supported (see meta-analysis by Peters et al., 1985). However, there
have been criticisms of its theory and methodology. At a theory level, there exists an
inability to explain the underlying processes that facilitate the prediction of
performance (Chemers, 1997; Fiedler & Garcia, 1987). Theoretically and
methodologically, there are also problems using only the leader when assessing the
quality of the leader-follower relationships. As Hollander (1971) notes,
“interpersonal processes involve reciprocity of perception, not just the leader’s own
perceptions [and requires]… awareness of the interplay of situational factors in a
system of relationships including personality, structural, and tasks variables” (p.2).
37
In support, meta-analytical research demonstrates that leaders and followers do not
usually see the same relationship in the same way (Gerstner & Day, 1997; Sin et al.,
2009). Consequently, using a singular perspective to assess the quality of the leader-
follower relationship does not appear suitable.
Cognitive resource theory. In reaction to the inability to explain the
underlying processes underpinning performance, Fiedler proposed cognitive resource
theory (Fiedler, 1986; Fiedler & Chemers, 1984; Fiedler & Garcia, 1987). Cognitive
resources theory integrates the propositions of contingency theory with leaders’
cognitive resources to predict performance. Cognitive resources are the “intellectual
abilities, technical competence, and job-relevant knowledge obtained by formal
training or experience in the organization” (Fiedler & Garcia, 1987, p.2) and are
assumed to be the major contributor to the plans, decisions and strategies of group
action (Fiedler & Chemers, 1984; Fiedler & Garcia, 1987). These cognitive
resources are particularly important when tasks are complex and require the
application of logic and analytical analysis. However, stress and anxiety can
undermine a leader’s ability to concentrate and think effectively (Fiedler & Chemers,
1984; Fiedler & Garcia, 1987). Consequently, decisions about effective courses of
action can be undermined. Indeed, under certain conditions, intelligence has even
been negatively related to performance (see Fiedler, 1986). Leaders’ experience and
intelligence though can mitigate the effects of some of this stress on their decisions
and actions (Fiedler, 1986; Fiedler & Garcia, 1987). Specifically, under highly
stressful situations, a leader’s experience should result in better decisions as they are
able to call upon previous experiences of similar situations and learnt techniques to
appropriately guide them. For less experienced leaders, though, this lack of
38
experiential knowledge may result in poorer decisions and outcomes. In general,
though, leader abilities will only support performance when followers are motivated
and supportive of the leader (Fiedler, 1986).
The propositions of cognitive resource theory have been generally supported
by research demonstrating positive relationships between leader intelligence and
experience with performance under conditions of low and high stress respectively
(e.g. Fiedler, 1986; Fiedler & Leister, 1977; Fiedler, Potter, Zais, & Knowlton, 1978;
1979; Potter & Fiedler, 1981). A meta-analysis by Judge, Colbert, and Ilies (2004)
further reinforces these findings. Specifically, under conditions of low stress, leader
intelligence shows a positive relationship with leadership. Under conditions of high
stress though, this relationship disappears. While the key aspect of cognitive
resource theory is the focus on the cognitive elements and experience of the leader
though, it did not explicitly recognise humans as having two independent cognitive
systems as described in CEST (Epstein, 1998; Cerni, Curtis, & Colmar, 2010).
Furthermore, stress and anxiety (i.e. emotions) are seen as the outcome of
unconscious information processing within CEST (Epstein, 1998; 2003). By looking
purely at emotions rather than the underlying cognitions contributing to them,
therefore, misses an important element in human cognition.
New Leadership Theories
A core issue with the leadership theories presented thus far is that none
explicitly recognise the emotional and visionary aspects of leadership influence
(Yukl, 1998). Instead, each is focused more on the transactional aspects of managing
others (Chemers, 1997). In contrast, new leadership theories can be seen as an
39
attempt to incorporate the affective reactions of the follower to the leader as they
articulate an inspiring vision (Uhl-Bien et al., 2014). Chemers (1997) states this in a
different way when he says “many of the components of transformational leadership
are present in other approaches… What is unique… is the extent to which a
transcendent vision is the source of motivation for followers” (p.92-93). Most new
leadership theories are typically described in similar ways (Chemers, 1997;
Northouse, 2007). Thus, for current purposes, only two key new leadership theories
are reviewed and briefly critiqued here: charismatic leadership (Conger & Kanungo,
1987; House, 1977) and transformational leadership (Avolio, 1999; Bass, 1985; Bass
& Avolio, 1990).
Charismatic leadership. House (1977) initially put forward the theory of
charismatic leadership. He argued that the effects of a charismatic leader can be
identified through followers’ extraordinary levels of commitment, identification, and
emulation towards, and with, the leader (Chemers, 1997). House (1977) identified
three categories which encapsulated charismatic leadership: (1) personal
characteristics of the leader; (2) behaviours of the leader; and (3) situational factors.
Person characteristics can be understood as the traits of the leader and include
extremely high levels of self-confidence, a need to dominate others, a moral
conviction in the righteousness of their beliefs, and a high need to influence others.
Behaviours are focused on: (1) the role modelling of their belief and values; (2)
building an image which facilitates the perception of charisma in followers of
competence, power, and concern; (3) stating of goals in ways which are
transcendent, ideological and are of moral concern; (4) setting high expectations of
followers to achieve these lofty goals; (5) show they are confident that the followers
40
can achieve these goals; and (6) does this in ways which bring about cognitive and
emotional states in followers which are aligned to goal related behaviours.
Situational factors include those contextual factors that support but are not
necessarily needed, for the charismatic leadership process to occur (e.g. stressful
situations which provide an opportunity for a leader to engender clear ways to think,
feel and behave in the follower).
Transformational leadership (Full-range leadership theory).
Transformational leadership is similar in nature to charismatic leadership (Bass,
1985; Burns, 1978). Burns (1978) first described the transforming leadership
concept in research on political leaders and defined it as a process by which leaders
and followers transform each other to achieve higher levels of morale and
motivation. A transforming leader does this through their personality, traits, and
ability to make a significant change through example, articulation of an energising
vision, and challenging goals. It is about transforming followers’ perceptions, values
and expectations to achieve better goals than they could have thought initially
possible. He set transforming leadership apart from transactional leadership based on
the premise of transactions that occur between leader and follower in their
relationship. Bass (1985) extended on Burns work to explain the underlying
psychological mechanisms by which transforming and transactional leadership
emerge; replacing the term transforming with transformational. In addition, whereas
Burns (1978) clearly delineated between transformational and transactional
leadership, Bass (1985) argued that leaders are able to exhibit both transformational
and transactional leadership styles at the same time. The result was the development
41
of the Full-Range Leadership Theory (FRLT; Antonakis, Avolio, &
Sivasubramaniam, 2003; Avolio & Bass, 1991; Bass & Avolio, 1997).
FRLT incorporates transformational, transactional, and laissez-faire
leadership categories into an overall leadership model (Bass, 1985; Bass & Avolio,
1993; Yukl, 1998). Transactional leadership refers a style of leadership which is
based on a contractually based exchange process; specifically, where the leader
specifies the requirements and rewards for completion, and monitors and controls
progression to the outcome (Antonakis et al., 2003; Bass & Avolio, 1994). Laissez-
faire leadership, at the opposite end of the spectrum, represents the active choice by a
leader to disengage from making decisions, and deploying the authority of their
position and is considered the least effective form (Antonakis et al., 2003).
Transformational leadership, in contrast, is argued to be the most effective style as it
inspires followers towards positive change through expanding their awareness of
what is possible (Antonakis et al., 2003; Avolio & Bass, 1991; Bass & Avolio,
1997).
Issues with new leadership theories. Charismatic and transformational
leadership theories have been popular in research and practice but have been
critiqued on several grounds (Northouse, 2007). In particular, while both theories
acknowledge followers in the leadership process (e.g. meeting their needs and
inspiring them to become more valuable individuals) they are still not given
prominence in the leadership process. Specifically, no explicit recognition is given to
the characteristics of the follower and the active and reciprocal role they have in the
leadership process (Yukl, 1999). Instead, the focus is still on the independent and
42
heroic leader who transforms the follower (Northouse, 2007). Yet research
demonstrates that follower developmental characteristics (Dvir & Shamir, 2003),
personality (Moss & Ngu, 2006; Shalit, Popper, & Zakay, 2010; Suls, Martin, &
David, 1998), and values (Ehrhart & Klein, 2001; Shin & Zhou, 2003; Wofford,
Whittington, & Goodwin, 2001) can influence the effectiveness of charismatic and
transformational leadership. A related criticism is also the question of where
charisma resides. Specifically, whether charisma is determined by followers (i.e.
bestowed upon the leader; see Lord & Maher, 1993; Willner, 1984) or is inherent to
the characteristics of a leader (see Conger & Kanugo, 1998). Given that both the
leader and follower are interdependently influencing each other’s thoughts, emotions
and behaviours (Hollander, 1992) it would seem likely that the attribution of
charisma is dependent on both parties.
Follower Centric Theories of Leadership
So far, the focus has been on the leader and how they influence the follower
to achieve particular outcomes. However, in the last 25 years or so, a substantial
body of theory and research has emerged demonstrating that effective leadership is
not only dependent on leaders’ behaviours but also how followers construct leaders
(Felfe, 2005; Lord & Maher, 1993). The general lack of focus on followers prior to
this recent period of time is surprising for three reasons: (1) leadership by definition
implies followers; (2) followers’ are not passive, they play active roles in team,
group, and organisational successes and failures (Avolio & Reichard, 2008; Baker,
2007) and; (3) followers’ make up the bulk of an organisation’s workforce and do
most of the heavy lifting (Bennis, 2008). This section, therefore, reviews relevant
43
follower-centric theories of leadership including the attribution theory of leadership
(Calder, 1977), implicit leadership theory (Eden & Leviathan, 1975; Lord & Maher,
1993), and the romance of leadership (Meindl, 1995; 1998; Meindl, Ehrlich, &
Dukerick, 1985) with a particular focus on implicit leadership theory due to its core
integration with the research undertaken in this dissertation.
Attribution theory of leadership. Calder (1977) argued that leadership is a
dispositional attribute. However, it cannot be directly observed, only inferred from
its observable impacts within a particular social context. Consequently, Calder
argued that leadership research needed to reorientate its focus to how people socially
construct leadership – “How people make inferences about and react to leadership is
itself an important behaviour to explain” (Calder, 1977, p.186). At its most basic, the
attribution theory of leadership (Calder, 1977) describes how people work
backwards to infer leadership in a social context. This process begins with the
seeking of evidence of leadership through actual, or indirect (e.g. through
discussions with others of the effects of someone’s behaviours), observations of
behaviour. If evidence exists to suggest leadership, then it is assessed against a
leader prototype. This process can be considered a sense-making exercise as
behaviours become linked to the underlying beliefs of leaders (Calder, 1977).
Points raised by Calder (1977) preceded some of the follower-centric theories
described below (e.g. implicit leadership theory). However, a key distinction
between Calder and these was the position that leadership itself was not important,
only the perception of leadership (Chemers, 1997; Lord & Maher, 1990; 1993). Such
a singular position seems unlikely. Without some objective object (in this case a
44
leader) by which to apply some standard (i.e. leadership prototype), a judgement as
to whether something exists (i.e. leadership) cannot occur. Consequently, it is more
likely that both objective and perceived leadership are contributing factors to the
leadership process.
Implicit leadership theory. Implicit Leadership Theory (ILT) evolves from
Eden and Leviatan’s (1975) initial focus on Implicit Organisational Theory and
refers to an individual’s personal assumptions of the traits and abilities of leaders’
(Eden & Leviathan, 1975; 2005; Lord & Maher, 1993). They are the knowledge
structures or schemas which specify and organise information around the traits and
behaviours of a leader and act to help people make sense of the organisational
environment specifically in relation to leaders (Poole et al., 1989; Weick, 1995).
They are activated when a follower comes into contact with a known leader or a
person exhibits characteristics associated with being a leader (Lord & Maher, 1993)
and have important implications for group and organisational outcomes
(Hansbrough, 2005; Nye, 2005). For example, leaders who match their followers’
ILTs are more likely to be given an acknowledgement for success, are perceived to
be more powerful, and obtain more positive evaluations (Lord et al., 1984; Nye &
Forsyth, 1991).
Leadership categorisation theory sought to explain the cognitive processes
underpinning the recognition and categorisation of leaders (Lord, Foti, & Philips,
1982; Rosch, 1978). Within this theory, leader categories are hierarchically
organised at three levels: (1) the superordinate level; (2) the basic level; and (3) the
subordinate level. The superordinate contains the attributes common to all leaders
45
(“family resemblance”) and those that distinguish them from “non-leaders” (Hoogh,
Den Hartog, & Koopman, 2005; Lord et al., 1984). The basic level contains
perceptual and functional attributes that are more refined to the context in which
leadership is being experienced (e.g. business leaders, political leaders, military
leaders, and religious leaders; Lord et al., 1984). The subordinate level contains
more distinctive types of leaders that emerge at the basic level. For example,
business leaders may be further distinguished by those at lower levels of the
organisation to those in strategic leadership roles (Lord & Maher, 1993).
One of the weaknesses of leadership categorisation theory is the static view
of leadership schemas. Specifically, it assumes that prototypes of leaders are
relatively fixed (Lord, Brown, Harvey, & Hall, 2001; Lord & Emrich, 2001).
However, while ILTs are generally quite stable over time (see Epitropaki & Martin,
2004), they can vary across contexts (e.g. see Epitropaki & Martin, 2005). A
connectionist model of ILTs, therefore, was proposed which takes a more dynamic
view of leader-prototype construction (Hanges et al., 2000; Lord et al., 2001).
Connectionist models emphasise the interconnections between units represent
the basic knowledge blocks of the connectionist network and can be activated by
environmental input or by other units in the network (Hanges et al., 2000). Units in
this network are activated via initial environmental input. This activation then
selectively spreads throughout the entire network. The spread of this activation
depends on the inputs and constraints acting in a particular context and the weighting
of the relationships between units. It is the pattern of this activation that represents
complex knowledge (Lord & Brown, 2001). Lord and colleagues (2001) identify
four different types of contextual constraints that influence units in an ILT network
to become activated: (1) the organisational culture; (2) the leader; (3) the follower;
46
and (4) the current task. Thus, the network is regenerated during each experience of a
leader, in relation to current contextual factors, and can help explain both the
flexibility as well as the stability of ILTs (Lord & Brown, 2001; Hanges et al., 2000;
Shondrick, Dinh, & Lord, 2010). This description of the connectionist model appears
compatible with CEST descriptions of the self-concept (Epstein, 2003; see also
chapter 1) suggesting a theoretical crossover between CEST and ILTs.
Early ILT research was primarily concerned with how ILTs influenced
leadership measurement. When the leader category is made salient, the perceiver
actively processes and manipulates information about that leader through the
information and characteristics contained in their ILT (Bass & Avolio, 1989; Lord et
al., 1984; Rush, Thomas, & Lord, 1977). Drawing upon adaptive resonance theory
(see Grossberg, 2013) Shondrick and Lord (2010) describe how this may happen
using an information processing approach to demonstrate that resonance (i.e. where
an external stimulus such as a leader is matched to a specific cognitive category)
creates difficulty in distinguishing between the ILT and the actual leader’s
behaviours. Indeed, early research demonstrates that people’s ILTs of prototypical
leaders typically represent the same factor structure as evaluations of actual leaders
(Eden & Leviatan, 2005; Schyns & Meindl, 2005) especially when the amount and
access to information about the leaders’ behaviour are small or vague (Eden &
Leviatan, 1975; Uhl-Bien et al., 2014). Consequently, when individuals are asked to
rate a leader, a systematic error may occur due to individuals drawing upon their
ILTs to fill in the missing information gaps about the leader (see Bass & Avolio,
1989; Eden & Leviatan, 2005; Lord et al., 1984; Schyns & Meindl, 2005).
Lord and colleagues (1984) were amongst the first to conduct research to try
to identify the specific contents of ILTs as previous research had typically measured
47
them on an ad-hoc basis. They identified 59 traits and behaviours (e.g. intelligent,
decisive, fair, and persuasive) all of which were associated with leaders in varying
degrees. Offerman, Kennedy and Wirtz (1994), recognising the need to develop a
reliable and valid measurement tool for measuring ILTs, identified 41 traits and
behaviours people associate with leaders. These load onto eight core ILT factors: (1)
Sensitivity (2) Dedication; (3) Charisma; (4) Attractiveness (5) Masculinity (6)
Intelligence (7) Strength and; (8) Tyranny and characterize individual understandings
of leaders, effective leaders, and supervisors albeit to different degrees (e.g.
sensitivity was seen as less important in supervisors than ineffective leaders).
Research supports the generalizability of these ILT factors (Epitropaki, 2000)
although some studies suggest that, in the general working population, six factors
may be a better reflection of ILTs (see Epitropaki & Martin, 2004; 2005).
Only a limited amount of theory and research looking at the individual
characteristics that contribute to variations in these ILT factors has emerged (c.f.
Ayman-Nolley & Ayman, 2005; Felfe, 2005; Keller, 1999). This is despite Hunt,
Boal and Sorenson (1990) over 20 years ago noting that characteristics of the
individual may predict ILTs. Schyns and Meindl (2005) again more recently called
for more research in this area, a call that the current dissertation seeks to address.
Within the limited prediction-focused research body, though, Keller (1999) found
that personality traits and parental traits predicted students’ ILTs. Specifically,
drawing from similarity and positive self-illusion literature, she found that the
personality traits of agreeableness, conscientiousness, extroversion, openness and
neuroticism tended to be related to similar preferences for traits and behaviours in
their leader. There was also evidence that self-reported maternal and paternal traits
were related to the students’ preferences for leader traits. For example, higher levels
48
of parental tyranny were associated with higher levels of tyranny in their ideal
leader. Keller (1999) also found evidence that some individuals preferred a leader
who was different from the self. In particular, individuals who experience high levels
of negative emotions (e.g. guilt, fear, anxiety) preferred an ideal leader to be
sensitive suggesting that there is a compensatory component in more neurotic
followers ILTs (Keller, 1999). Specifically, such these types of followers may find
that a caring and sympathetic leader is ideal as they provide a sense of safety and
thus compatibility. Thus, she concludes that people not only look for symmetry in
leader traits to that of their own but also idealise traits which will complement them.
While these findings by Keller (1999) are important, there are several
theoretical and methodological issues that should be considered. First, CEST states
there are two self-concepts that exist independently in each system (Epstein, 2003).
Research suggests that mainstream personality measures, akin to that used by Keller
(1999), primarily measure only the personality traits within the rational system
(Epstein, 1992; 2001). Considering how the experiential components of personality
may influence an individual’s ILT would, therefore, be a potentially important area
to study, especially considering that the experiential system is seen to influence most
people’s behaviours (Epstein, 1998; 2003). Second, her use of students (reported
mean age of 20) is unlikely to reflect a general working population who would tend
to be older and less educated. Furthermore, they are also likely to have less
experience with business leaders compared to a general working population.
Consequently, Keller’s (1999) results may not be fully generalisable to the general
working population.
49
Romance of leadership. The romance of leadership as described by Meindl
and colleagues (1985) can be considered a special form of ILTs (Meindl, 1995;
1998). This is because it puts a particular focus on how people use the concept of
leadership to “understand, interpret, and otherwise give meaning to organizational
activities and outcomes” (Meindl et al., 1985, p. 78). The use of the word “romance”
is to draw attention to the observation that people in society generally bias, or
assume, leadership is the most important factor in the (dys)functioning and
(under)performance of groups and organisations whilst relegating other elements as
less important (Felfe, 2005; Meindl, 1995; Meindl et al., 1985). This bias draws the
focus away from followers and the situational factors, both of which may be
objectively important causal factors to organisational success and failure. This occurs
because it is often cognitively simpler to explain various organisational events and
outcomes as due to a single factor rather than try to understand their real complexity
(Schyns & Bligh, 2007). Such propositions help to understand, at least partly, why
such intense interest is targeted towards leaders in general at the expense of other
important leadership factors. It further suggests a need for researchers to move away
from this societal perceived singular impact of leaders to other elements to better
understand the dynamic leadership process. Another way that leadership, therefore,
has been considered is through the relationship between leaders and followers.
Relationship-Focused Perspectives of Leadership
Relationship-focused perspectives of leadership typically recognise that
leadership occurs within leader-follower interactions. These approaches, therefore,
understand leadership as a dynamic process mutually influenced by both leaders and
followers. Relationship based perspectives which are reviewed here include
50
psychodynamic approaches to leadership (Bornstein, 2003; Stech, 2007), the
relational view of leadership (Hollander, 1971; 1992) and leader-member exchange
(Dansereau et al., 1975; Graen et al., 1973).
Psychodynamic approaches to leadership. One of the core aims of the
psychodynamic approach, in regards to leadership, is to focus on providing insight
into leaders’ and followers’ own personality types and discussing the implications of
these to their working relationship1 (Stech, 2007). Rather than a single theory, the
psychodynamic approach is a collection of theories and models focused on
understanding the consistent patterns in the way people think, feel, and act towards
the world, and others, and how these impact on relationships at work (Stech, 2007).
These patterns are referred to as a person’s personality and our early experiences in
life have a critical impact on its development (Bornstein, 2003). The psychodynamic
approach to leadership differs from previous leadership theories so far described in
several key ways including: (1) a recognition of the importance of the unconscious in
influencing peoples’ functioning and; (2) a focus on how the personalities of the
leader and follower influence their working relationship (Bornstein, 2003; Stech,
2007). It is because of the focus on the unconscious mind that psychodynamic
approaches are not mainstream within the study of leadership (Jackson & Parry,
2011). Yet the theories and ideas in this field can provide significant insight into the
interactions between leaders and followers.
A key psychodynamic theory underpinning the highly used Myers-Briggs
Type Inventory (MBTI; Myers & McCaulley, 1985) comes from Carl Jung,
1It is because of these features of psychodynamic conceptualisations of leadership that they are
considered here as relationship approaches.
51
Katherine Briggs and Isabelle Briggs-Myers (Jung, 1923; 1991; Myers & Myers,
2010). Together they argued that individuals vary on four key dichotomous cognitive
preference dimensions regarding: sensing-intuition, thinking-feeling, extraversion-
introversion, and judging-perceiving. These preferences influence how people
perceive the world and make decisions. For the sensing-intuition dimension, those
high in sensing gather information in a highly precise way and emphasise facts and
detail. In contrast, those high in intuition prefer to focus on the theoretical and
abstract. In terms of the thinking-feeling dimension, those high on thinking prefer to
make objective decisions based upon a rational and analytical consideration of the
evidence. Those high on feeling though prefer to make decisions in more subjective
ways. For the extraverted-introverted dimension, those high on extraversion typically
are energised by, and process information through, interacting with the physical
world. Introverted types, in contrast, are energised by, and process information, by
seeking quiet internal reflection. In terms of the judging-perceiving dimension, those
high on judging prefer structure and planning and are more decisive and deliberate.
In contrast, those high on perceiving are more flexible and adaptable to things. A
total of 16 personality types can be identified by these four cognitive dimensions.
Followers’ personality type has been found to predict ratings of their leader’s
transactional and transformational leadership (Atwater & Yammarino, 1993;
Hautala, 2005; 2006). For example, individuals high on extraversion and feeling,
tend to perceive higher levels of transformative behaviour in their leaders than those
of introverts and thinking types (Hautala, 2005). Atwater and Yammarino (1993)
found that subordinates’ thinking/feeling predicted rating of their leader’s
transactional and transformational leadership behaviours in a military setting.
Interestingly, Atwater and Yammarino (1993) used an earlier version of the
52
Constructive Thinking Inventory (Epstein & Meier, 1989) and found a negative
relationship between a subordinates’ emotional coping and their rating of their
leader’s transformational leadership. They argued that, unlike poor emotional copers,
good emotional copers are secure in their self and, thus, less likely to be influenced
by others’ opinions of them (i.e. more resilient).
Attempts have also been made to describe how specific leader personality
types should interact with followers of a specific personality type (e.g. Kroeger et al.,
2002). For example, Kroeger and colleagues (2002) state that extraverted leaders
should be aware that silence is not indicative of agreement or consent in introverted
followers. Consequently, they need to provide them space for reflection and allow
them to digest ideas and decisions. In contrast, if the follower is also an extravert, the
leader needs to be aware that the quantity of discussion is not the same as the quality
of outcomes. Furthermore, they need to provide extensive time for the discussion of
ideas. To the author’s awareness, though, no empirical assessment of these
recommendations of how leaders should adapt to follower personality types have
been undertaken.
Northouse (2007) argues that the ideas within the psychodynamic literature
could be useful to better understand the leadership process. Indeed, while
management scholars have focused less on the role of followers, those in the field of
psychoanalysis and anthropology have specifically identified the need to focus on
the psychological link between leaders and followers and the effects this link has on
their lives at the individual level (Baker, 2007). Sociological scholars have also
identified leadership as an intertwined relationship between leaders and followers
and that the former need to live according to the norms of those they lead (Adorno,
Frenkel-Brunswik, Levinson, & Sanford, 1950; Homans, 1950). Thus,
53
psychodynamic theories of leadership have implications for where to focus attention
in the leadership process. Specifically, the emphasis is taken away from the leader
and instead is on the interactions of each other’s personality (Kets de Vries, 2004;
Winkler, 2010), a focus that is not explicitly present in the other leadership theories
so far presented. Consequently, within the psychodynamic approach, the
effectiveness or quality of the leader-follower working relationship is a result of the
combination of leaders’ and followers’ personality (Stech, 2007).
However, psychodynamic approaches to leadership are not without critical
limitations. First, many psychodynamic theories were not developed through
observation of normal populations but in the context of clinical observation and the
treatment of people with significant psychological dysfunctions (see Berne, 1961;
Stech, 2007). Generalisability of these theories, therefore, is a potential issue.
Second, the theories and methods used in the psychodynamic approach are limited in
their scientific defensibility for understanding the unconscious (Epstein, 2003). In
recent decades, though, there has been an increasing recognition that many of the
ideas within the early psychodynamic approaches are indeed compatible with
cognitive psychology models of perception, memory and information processing
(Bornstein, 2003; Bucci, 1997; Horowitz, 1988; Stein, 1997). Importantly, CEST
with its approach to understanding the cognitive unconscious appears to be quite
well suited to developing a better understanding of psychodynamic concepts in the
domain of leadership in a much more scientifically defensible way (Epstein, 2003).
Third, from a practical perspective, the concepts and methods within the
psychodynamic approach limit themselves in terms of a how we approach leader
training and development (Stech, 2007). At best they argue for leaders and followers
to become aware of their own and other’s interpretative and behavioural patterns in
54
order to become more tolerant towards the self and others’ reactions (Stech, 2007).
While this insight is necessary, it does not lend itself to changing these interpretative
and behavioural patterns. CEST though is of more relevance from a practical
perspective given that information processing styles and constructive thinking
elements can be changed in leaders (Cerni et al., 2010). Consequently, these
elements should be amenable to change in leaders and followers and, thus, go
beyond simply bringing awareness to how two personality types may operate within
leader and follower interactions to actively changing these to improve the
relationship.
Finally, while psychodynamic approaches recognise the need to look at the
interaction between leader and follower personalities and their impact on their
working relationship, they do not explicitly recognise that leaders and followers hold
different beliefs about the roles of follower and leader. Yet ILTs (see above) and
implicit followership theories (discussed below; Sy, 2010) also influence how
leaders and followers perceive and evaluate their dyadic partner. Consequently, there
exists a need to integrate these different elements in a more holistic way to better
understand the leader-follower relationship.
Relational view of leadership. Hollander originally critiqued leadership
research (primarily trait and situational leadership theories), which focused on the
leader holding a specific position of authority rather than as a process involving both
the leader and the follower (Hollander, 1971; 1992). In his relational view of
leadership (see Hollander 1971; 1992; 2012), leadership is not bound to the leader as
a person, but rather as a process which “must be gauged in light of the attributes and
55
perceptions of the led and the structure and setting within which the leader and
followers interact for the achievement of specifiable organizational ends”
(Hollander, 1971, p.1). Leadership, therefore, is understood to operate within a
system of constraining and enabling relationships in regards to both task demands
and follower characteristics (Hollander & Offermann, 1990). Similar to Leader-
Member Exchange (Graen & Uhl-Bien, 1991; 1995; Uhl-Bien & Graen, 1993)
described below, exchanges or transactions between leaders and followers occur
(Hollander & Julian, 1969; Hollander & Offermann, 1990). Specifically, the
resources provided to the follower by the leader serve to increase the levels of
influence and status the leader is given by the follower resulting in the leader holding
the greater legitimate authority to direct the follower. Importantly, though, the leader
was still seen by Hollander to hold the more important and influential role in this
process.
Leader-member-exchange (LMX). The key focus of Leader-Member
Exchange Theory (LMX) is on the dyadic relationships between leaders and
followers (Dansereau et al., 1975; Graen et al., 1973; Graen & Uhl-Bien, 1995;
Northouse, 2007). Originally called Vertical Dyad Linkage Theory (Dansereau et al.,
1975; Graen et al., 1973, LMX has gone through a number of evolutions since its
initial inception to now become the most prominent relational or dyadic leadership
theory (Bauer & Erdogan, 2016; Day & Miscenko, 2016).
LMX was originally proposed due to problematic assumptions in previous
models of leadership which included: (1) the view that followers are essentially seen
as a homogeneous set of individuals who perceive, interpret and react in very similar
56
ways; and (2) that the leaders act similarly with each of their followers (Dansereau et
al., 1975). It emphasises the process of role making between a leader and follower
and takes the position that leaders will develop different types of exchange
relationships with each of their followers (Graen & Uhl-Bien, 1991; 1995). These
relationships will range from low quality, or purely contractual based (“the out-
group”), to high quality which are characterised by mutual trust, respect, liking and
reciprocal influence (“the in-group”; Liden & Maslyn, 1998). Since this initial
conception, LMX has gone through several changes to present a more prescriptive
approach to effective leadership (Graen & Uhl-Bien, 1995). There have also been
recent trends at extending the dyadic view to both the group (e.g. Boies & Howell,
2006; Gajendran & Joshi, 2012) and network levels (e.g. Goodwin, Bowler, &
Whittington, 2008; Venkataramani, Green, & Schleicher, 2010) on the recognition
that leadership is not only dyadic in nature but also occurring among multiple
members within an organisational network (Graen & Uhl-Bien, 1995). However,
these are not of immediate interest for the current review purposes.
There is now substantial evidence to demonstrate that the quality of the
leader-follower relationship is a key factor in leadership effectiveness (Dulebohn et
al., 2012; Erdogan & Liden, 2002; Graen & Uhl-Bien, 1995; Ilies, Nahrgan, &
Morgeson, 2007; Liden, Wayne, & Stilwell, 1993). When high quality relationships
exist there are associated improvements in performance evaluations, organisational
commitment, innovation and creativity, employee turnover, and job stress for both
followers and leaders (Dulebohn et al., 2012; Dunegan, Uhl-Bien, & Duchon, 2002;
Erdogan & Liden, 2002; Graen & Uhl-Bien, 1995; Judge & Ferris, 1992; Liden et
al., 1993; Tierney, 2016). In addition, followers work harder (Basu & Green, 1997;
Duchon, Green, & Taber, 1986), engage in more effective communication (Fairhurst,
57
1993; Fairhurst & Chandler, 1989), and show increased organisational citizenship
behaviours (Nystrom, 1990; Seers & Graen, 1984). Furthermore, many of these
outcomes are observed across different cultures and nations (Anseel & Lievens,
2007; Eden, 1993; Erdogan, Liden, & Kraimer, 2006; Schyns, Paul, Mohr, & Blank,
2005; Wakabayashi & Graen, 1984) demonstrating that relationship quality is a
highly generalisable construct (van Gils et al., 2010). Given the wide-ranging
important organisational outcomes, it becomes important to understand LMX’s
antecedents.
While the consequences of LMX have been extensively researched, its
antecedents have been somewhat less studied (Mahsud et al., 2010). However, recent
meta-analytical research (Dulebohn et al., 2012) identifies several general antecedent
categories: (1) follower characteristics; (2) leader characteristics and (3)
interpersonal variables. Under the follower characteristics category, a leader’s
perception of competence and ability of the follower is positively related to the
follower’s perception of LMX quality. Follower personality traits of agreeableness,
conscientiousness, and extraversion, locus of control, and positive affect were also
positively related to perceptions of follower LMX. Such traits and behaviours are
likely to be viewed by leaders as positive and productive as they help them move
towards their goals in their role. For example, high levels of conscientiousness are
one of the most reliable predictors of job performance (Barrick & Mount, 1991).
Agreeableness is also associated with increased pro-social and helping behaviours
and cooperation (Graziano, Habashi, Sheese, & Tobin, 2007) in addition to
reciprocity (Perugini, Gallucci, Presaghi, & Ercolani, 2003). High conscientiousness
and agreeableness in followers are therefore likely to result in more respect and trust
58
for the follower which facilitates more effective working relationships or higher
quality working relationships (i.e. high LMX).
Other studies in the school of information processing have found that a
variety of cognitive factors are related to LMX (Day & Miscenki, 2016). Engle and
Lord (1997) found that if leaders and followers perceived that their attitudes were
similar to their dyadic partner, the higher their LMX rating. Similarly, other research
demonstrates that similarity in leaders and followers’ identities predict LMX ratings
(Jackson & Johnson, 2012). In this study, when both leaders and followers perceived
themselves as unique individuals, who are motivated to fulfil their partners role
expectations and better their welfare, and saw themselves as part of a collective (i.e.
held compatible self-definitions), they were more likely to report higher LMX levels.
Emotional intelligence similarity between leaders and followers has also been
associated with higher follower LMX rating (Sears & Holmvall, 2010). However,
not all research has found that congruence between leaders and followers will
ultimately result in increased LMX (e.g. Oren, Tziner, Sharoni, Amor, & Alon,
2012). Given the integral links with affect in CEST (Epstein, 2003), it is also
important to note that affect influences LMX ratings (Sears & Holmvall, 2010;
Xiaqi, Kun, Sufang, 2012; Xu, Liu, & Guo, 2014). Xu and colleagues (2014), for
example found that emotional masking by followers negatively impacted on their
LMX rating, which subsequently had deleterious effects on their wellbeing, job
satisfaction and turnover intent. Xiaqi and colleagues (2012) found that leaders with
high emotional intelligence tended have followers who rated their relationship as
more positive.
While extensive research has adopted LMX as an approach to understanding
and studying leadership, there are problems in its initial theoretical basis in that it
59
assumes that leaders and followers experience the relationship similarly (Graen &
Uhl-Bien, 1995). Graen and Uhl-Bien (1995) state that the “expected agreement
between leader and member reports is positive and strong and used as an index of
quality of data” (p. 237). Consequently, research has usually only measured
perceptions of the relationship from a single source, the follower, on the assumption
that both the leader and follower will experience the relationship in similar ways
(Graen & Uhl-Bien, 1995; Wilson et al., 2010; van Gils et al., 2010). Indeed, Sin and
colleagues (2009) found that only about 10% of the research papers published
between 1996 and 2008 on LMX assessed the construct from both the leaders’ and
followers’ perspective. Similarly, more recently Day and Miscenki (2016) state that
“there are fewer studies in the literature that have examined leader (compared with
follower) traits in relation to LMX, which is somewhat surprising” (p. 18). Yet meta-
analytical research (Gerstner & Day, 1997; Sin et al., 2009) demonstrates that the
correlation between leaders’ and followers’ rating of their relationship is actually
quite low (r’s between .29-.37). If it is correct to assume that leaders and followers
experience the relationship to a similar extent, then a much stronger correlation
should be expected. Consequently, future research on LMX needs to reconsider this
aspect of the theory. Specifically, it would need to theorise what is contributing to
these differences between leaders and followers (e.g. see Wilson et al., 2010; van
Gils et al., 2010), which the current research attempts to address.
The preceding review has focused on leadership from leader, follower and
relationship perspectives. Followers though have important influences on the
leadership process (Uhl-Bien et al., 2014). Indeed, Lord and Brown (2001) stated
that leadership is an interdependent process that relies on leaders leading and
followers following within a context. It appears, though, that most of the leadership
60
theories presented so far have failed to consider the impact of followers and
following on leadership. Consequently, a review followership literature is necessary.
Followership
Compared to the leader-domain, research on followers and followership is
heavily under-researched (Uhl-Bien et al., 2014; Bjugstad, Thach, Thompson, &
Morris, 2006). A search conducted February 2014 on ProQuest containing the
keyword “leadership” returned 2,195,501 articles. In contrast, the keyword
“followership” returned only 718. This is a ratio of approximately 3000 to 1 in
favour of leadership research. This lack of focus on followers is likely, in part,
underpinned by the cultural stereotypes which revolve around leaders and followers
(see Alcorn, 1992; Lundin & Lancaster, 1990; Meindl, 1995; Meindl et al., 1985).
Specifically, whereas leaders are typically ascribed with more positive connotations
in society (e.g. Meindl et al., 1985), followers are typically depicted in a more
negative fashion (e.g., sheep, passive, obedient; see Berg, 1998).
Various writers argue though that the success of organisations depends not
only on good leaders but also on active and courageous followers (e.g. Chaleff,
1995; Kelley, 1988; 2008; Lundin & Lancaster, 1990). Indeed, the move to more
decentralised, flatter organisational structures over the last few decades has meant
followers are becoming increasingly expected to take more independent and active
roles which do not depend on regular direction from those above (Lord, 2008;
Parker, Williams, & Turner, 2006). In particular, followers of highly skilled
professions are expected to be self-directed, self-motivated and act with minimal
direction (Howell & Méndez, 2008). Such proactivity in followers helps contribute
61
to both positive individual and organisational outcomes (e.g. Seibert, Kraimer, &
Crant, 2001; van Scotter, Motowidlo, & Cross, 2000). However, such proactivity
does not align with the stereotype of the passive follower.
Encouragingly, there is an increasing recognition that the follower is an
active contributor to organisational success and failure (e.g. Awamleh & Gardner,
1999; Ayman, 1993; Gardner & Avolio, 1998; Nye, 2002). Indeed, over 50% of the
718 followership articles were published after the year 2000. Furthermore, the
number of books which hold the word “followership” in their title have increased in
recent years (Jackson & Parry, 2011). A review of this body of followership theory
and research is presented below. A core thread throughout is that the effectiveness of
a leader is dependent on their followers’ consent to play their role (Bjugstad et al.,
2006). If leaders do not have the confidence and support of their followers, they are
unlikely to function effectively.
Followership Theories
According to Uhl-Bien and colleagues (2014), followership theories can be
broadly classified under role and relationship based perspectives.
Role-Based Theories of Followership
Role-based theories are focused on how followers undertake their role in
relation to the leader and how this impacts on the leadership process and
organisational outcomes. According to role theory, the actions of individuals in most
social activities are born from the knowledge of socially defined categories held by
62
all to some degree (Ritzer, 2000). When a person holds a role status in a given
context (such as a leader or follower) it serves as a script to guide the individual’s
behaviour (Biddle, 1979). The deployment of social categories in everyday
functioning facilitates predictability in the social world (Ritzer, 2000; Weick, 1995).
When individuals actively identify with their roles they come to think, feel and act in
ways which are aligned with them (e.g. Chiricos, Barrick, Bales, & Bontrager, 2007;
Jussim, Nelson, Manis, & Soffin, 1995). Relevant theories falling under role-based
perspectives, reviewed below, include the dynamics of subordinacy (Zaleznik,
1965), effective followership (Kelley, 1988; 1992; 2008), courageous followership
(Chaleff, 1995; 2003; 2009), obedience and proactive behaviours (e.g. Crant, 2000;
Reicher, Haslam, & Smith, 2012), and implicit followership theories (Sy, 2010). A
particular focus is on implicit followership theories due to their importance in the
research undertaken in this dissertation.
Dynamics of subordinacy. Zaleznik (1965) proposed a role based theory of
followership called the Dynamics of Subordinacy which draws on psychodynamic
concepts of projection and transference. Zaleznik argued that it is through the
relationship with authority figures that the followers’ internal motives, desires and
wishes are played out. There are two key dimensions on which these motives, desires
and wishes emerge: (1) dominance-submission; and (2) activity-passivity. For
dominance-submission, followers can range from those who want to dominate or
control authority figures to people who seek to be controlled by those in authority
positions. For the activity-passivity dimension, followers can range from those who
seek to initiate and intrude into their working environment, to those who prefer for
those in positions of authority to initiate actions and provide direction. From these
63
two dimensions, followers can be classified into four types: (1) impulsive (dominant
- active), (2) compulsive (dominant - passive), (3) masochistic (submissive - active),
and (4) withdrawn (submissive - passive). By being aware of such follower types
and their underpinning internal conflicts, Zaleznik sought to help leaders achieve
greater sensitivity and judgement in their relationships with followers (1965).
As the basis of the dynamics of subordinacy theory is psychodynamic in
nature (Zaleznik, 1965), it has the same limitations as those presented previously for
psychodynamic approaches to leadership. However, as noted, CEST (Epstein, 2003)
is able to integrate psychodynamic theories in a more scientifically defensible
manner. Consequently, the ideas presented within this and subsequently related
psychodynamic theories (e.g. see Oglensky, 1995) appear to be quite useful when
drawn upon to help understand how certain aspects of the followers’ unconscious
world influence the leader-follower relationship. For example, Oglensky (1995)
argues that followers’ relationships with leaders are complex emotional attachments
that reflect followers’ internal needs and dispositions. This is a dynamic perspective
of the follower as it places them as active in a hierarchically based social
relationship, which interacts with their internal cognitive-emotional world.
Consequently, in this view, the followers’ inner world influences the relationship
with the leader, a view that had not been traditionally recognised within mainstream
leadership theories. However, we must also consider the internal worlds of the leader
in these relationships as the same will exist for them.
Effective followership. Kelley (1988) published a widely cited paper in
which he argued that organisational successes (and failures) are not only dependent
64
on leaders, and how well they lead, but also on how effective followers are in
following. He identified two underlying behavioural bi-polar dimensions of follower
qualities: (1) dependent-independent and; (2) passive-active (Kelly, 1988; 1992;
2008). The dependent-independent dimension refers to the degree followers are
independent, critical thinkers. The passive-active dimension refers to the degree
followers are active and take the initiative in decision making. Based on whether a
person falls high or low on these two dimensions, five different follower styles are
identified: (1) alienated followers (high independence – low active); (2) star
followers (high independence – high active); (3) passive followers (low
independence – low active); (4) conformist followers (low independence – high
active) and (5) pragmatist followers who fall midway on both dimensions. Star
followers are seen as most effective as they are positive, independent thinkers who
can achieve without direction and demonstrate the courage to voice antithetical
positions in a respectful and positive way to their leaders (Kelly, 1988; 1992).
However, while star followers may be theorised by Kelly as the most effective,
subjectively it would appear that leaders do not evaluate these follower types in such
a positive manner (e.g. see Crant, 2000; Reicher et al., 2012)
Courageous followership. In a similar vein to Kelley (1998), Chaleff (1995;
2003; 2009) argues that effective followership is necessary to bring about effective
leadership. Chaleff (1995; 2008) identifies two dimensions on which followers can
be placed: (1) low-high support and (b) low-high challenge of the leader. Based upon
these, he identifies four different styles of follower: (1) the implementer (high
support – low challenge); (2) the partner (high support – high challenge); (3) the
individualist (low support – high challenge); and (4) the resource (low support – low
65
challenge; Chaleff, 2008). For Chaleff, partners are the most effective type of
follower as they have both the courage to forcefully stand up for, and support, their
leader in their aim to achieve group and organisational goals, but also to challenge
and confront them when their actions may be morally questionable or result in poor
operational outcomes. In Chaleff’s view, effective followers, therefore, are active
contributors to achieving a positive working environment and are, at least partly,
accountable for the achievement of (and indeed the failure to achieve) group and
organisational goals (Chaleff, 1995; 2003; 2009).
It is important to note here that Chaleff (2009) sees effective leadership and
followership as fundamentally intertwined. It is not only the followers who need to
be courageous, but leaders also need to be courageous in the face of the candour of
their followers. If they do not demonstrate such courage to potentially critical
feedback, then they are likely to instil a culture of silence that undermines both
effective leadership and followership. Indeed, he states “leaders rarely use their
power wisely or effectively over long periods [of time] unless they are supported by
followers who have the stature to help them do so” (Chaleff, 2003, p.1). The
recognition that both leaders and followers need to be courageous is an important
notion, as it requires a consideration of the reasons as to why they may or may not be
“courageous”. It is proposed that their internal world may hold the key to
understanding this.
Obedience and proactive follower behaviours. Specific follower
behaviours have also been studied. These include obedience and proactive
behaviours. As already noted, followers are typically portrayed in more negative
66
terms such as sheep, passive and obedient (Baker, 2007; Berg, 1998). When
individuals actively identify with their roles they come to think, feel and act in ways
which are aligned with them (e.g. Chiricos et al., 2007; Hoption, Christie, & Barling,
2012; Jussim et al., 1995). In conjunction with the more positive and active beliefs
about leaders (Berg, 1998; Meindl et al., 1985) and the belief that they have an
inherent right to hold a position of authority (Hollander & Offerman, 1990), the
result is a tendency for followers to relegate themselves to simply following the
directions of the leader while the leader takes overriding responsibility (e.g. Blass,
2009; Milgram, 1965; Reicher et al., 2012; Vanderslice, 1988). While the image of
the obedient follower is prevalent, it is clear that not all followers are willing to be so
blindly obedient (e.g. see Carsten, Uhl-Bien, West, Patera, & McGregor, 2010;
Milgram, 1965) nor do some leaders expect this of followers (e.g. McGregor, 1960;
Sy, 2010). Instead, some followers take a dynamic and active approach to their role
which is also expected by some leaders.
Proactive behaviour refers to the challenging of the status quo through
actively adapting to the current situational demands in order to bring about improved
outcomes (Belschak, Hartog, & Fay, 2010; Crant, 2000; Grant & Ashford, 2008).
Research on proactive behaviour has primarily been concerned with four key areas:
(1) voice; (2) issue-selling, (3) taking charge; and (4) helping (Grant, Parker, &
Collins, 2009). Such behaviours are inherently connected with how employees
behave in the workplace to effect change and have generally been associated with a
wide range of positive individual and organisational outcomes including increased
performance (Rank, Carsten, Unger, & Spector, 2007), quicker career advancement
(Seibert, Crant, & Kraimer, 1999), increased organisational commitment (Hartog &
Belschak, 2007) and job satisfaction (Wanberg & Kammeyer-Mueller, 2000),
67
organisational strategy (Dutton & Ashford, 1993), firm success (Baer & Frese, 2003;
Frese & Fay, 2001), and team effectiveness and satisfaction (Kirkman & Rosen,
1999).
Due to the fact that leaders and followers do not usually see their relationship
in the same ways (Gerstner & Day, 1997; Sin et al., 2009), perhaps the most relevant
research in the proactivity literature to the current dissertation is how leaders’
interpret follower proactive behaviours. Despite the generally positive outcomes of
follower proactive behaviour, not all leaders interpret proactive behaviours in the
same way (Grant et al., 2009) and there is a general lack of understanding why this is
the case (see Grant & Ashford, 2008). Some have argued that supervisors may see
follower proactive behaviours as a threat (e.g. Grant et al., 2009) or an ingratiation
attempt (Bolino, 1999; 2003) rather than as a constructive attempt at improving the
situation. Argyris and Schön (1978) argue that many supervisors are driven to
protect themselves from various negative experiences due to potential
embarrassment, personal threat and/or feelings of vulnerability or incompetence.
Undeniably, there is strong evidence demonstrating that people often feel threatened
when confronted with negative criticism or feedback (Carver, Antoni, & Scheier,
1985; Meyer & Starke, 1982; Sachs, 1982; Swann & Read, 1981). This can even
motivate individuals to avoid or dismiss the criticism and even attack the
messenger’s credibility (Ilgen, Fisher, & Taylor, 1979; Morrison & Milliken, 2000).
Proactive behaviours in followers, therefore, appear to have the potential to engender
negative reactions in some leaders (Frese & Fay, 2001) due to their inherent nature
of questioning the status quo (e.g. questioning leaders’ directions, decisions and
actions).
68
Leaders’ negative reactions to followers’ proactive behaviour may also be
related to the way in which leaders’ process information. For example, emotional
coping contains items targeting how “thick-skinned” someone is to disappointment,
rejection criticism and disapproval (Epstein, 2001), all of which are behaviours a
proactive follower would likely demonstrate to some degree. Consequently, it
appears there are potential links between the reactions in leaders to proactive
follower behaviour and constructive thinking as conceptualised in CEST.
Implicit followership theory. Despite being a relatively new concept in the
literature, implicit followership theory (IFT) holds significant potential for
influencing the way we conceptualise leadership and followership (Avolio,
Walumbwa, & Weber, 2009; Epitropaki, Sy, Martin, Team-Quon, & Topakas,
2013). IFT corresponds to ILT but focuses its attention on the knowledge structures
or schemas, which specify and organise information around the traits and behaviours
of a follower and thus act to help people make sense of the organisational
environment specifically in relation to followers (Sy, 2010). Once someone is
attributed with followership status, perceptions and judgements are filtered through
the individual’s IFT, which in turn, have a subsequent influence on the behaviours in
relation to the follower (Lord & Maher, 1993; Poole et al., 1989; Sy, 2010). Like
ILTs, both cognitive categorisation and connectionist models have been drawn upon
to explain the cognitive processes underpinning the recognition and categorisation of
followers (Epitropaki et al., 2013).
The roots of IFTs can be observed in McGregor’s (1960) observation that
managers can hold two different understandings about the nature of followers, which
69
he referred to as Theory X and Theory Y managers. Theory X managers’ have rather
negative beliefs of followers, seeing them as lazy, unmotivated, and selfish (e.g.,
only work for payment). In contrast, Theory Y managers’ beliefs are more positive.
They tend to see employees as self-motivated and will, when provided the
opportunity, seek to do well in their tasks. Actions of the manager become aligned
with how the manager sees the nature of followers. For example, monetary rewards
contingent on work output and systems of control are more applicable for Theory X
managers’. In contrast, Theory Y managers will tend to engage in more democratic
and trusting behaviours with their followers.
As IFT is a relatively new area of research, there have only been a small
number of studies conducted so far. Research looking at the content of IFTs
identified 18 traits and behaviours which managers associate with followers (Sy,
2010). These loaded on to six main factors which they labelled: (1) Industry; (2)
Enthusiasm; (3) Good-citizen; (4) Incompetence; (5) Insubordination; (6)
Conformity. These six factors also loaded onto a second-order two-factor structure:
Followership Prototype (corresponding to the first 3 main factors presented above)
and Followership Anti-prototype (corresponding to the last 3 main factors). These
two second-order factors appear to somewhat align with Theory X and Theory Y
managers described above (see McGregor, 1960).
Research looking at the consequences of IFTs has found that leader IFTs are
positively related to both leaders’ and followers’ perceptions of relationship quality,
trust, liking and job satisfaction (Sy, 2010). Recent research has also found that
when leaders’ hold positive IFTs, it creates a naturally occurring Pygmalion effect in
followers (Whiteley, Sy, & Johnson, 2012). Specifically, positive leader IFTs result
in higher performance expectations and relationship quality, which subsequently
70
increase performance in followers. These results suggest that leader IFTs, at least in
part, shape the pattern of leaders’ interactions with followers and form the basis by
which the leader perceives and judges the followers behaviour in a dyadic
relationship (Sy, 2010). Due to these findings, Whiteley and colleagues (2012)
recently recommended that future research should look at the antecedents of IFTs.
Given the relative newness of IFTs though, only a limited amount of research
on IFT antecedents has so far emerged (cf. Derler & Weibler, 2014; Kruse & Sy,
2011). Kruse and Sy (2011) manipulated positive and negative affect in students,
employed adults and current leaders and found that positive affect was positively
related to individuals reporting more prototypical IFTs while negative affect was
positively related to anti-prototypical IFTs. The researchers suggest that, in addition
to cognitive components, affect also contributes to congruent effects on people’s
IFTs as would be predicted by connectionist, dynamic views of implicit theories.
Like previous work though on ILTs (e.g. Keller, 1999), these researchers did not
recognise that humans have two information processing systems. Consequently, one
of the aims of the current research was to assess whether these systems would be
related to leader IFTs.
Relationship Based Perspectives of Followership
In addition to the role-based perspectives discussed above, Shamir (2007) has
outlined a relational approach to followership called the co-production of leadership.
71
Co-production of leadership. Drawing upon LMX theory as its basis,
Shamir (2007) argues leadership is a social relationship and that the creation, nature,
and outcomes of leadership are dependent on, or co-produced by, both leaders and
followers. In this view, therefore, followers are not passive individuals but are active
in affecting the nature of the relationship and its outcomes. The co-production of
leadership readdresses the bias on leaders in previous leadership theory and research.
Specifically, there exists a need not only to understand how leaders’ characteristics
and behaviours influence the leadership relationship, but also how the followers’
characteristics and behaviours influence it (e.g. Carsten & Uhl-Bien, 2012; Carsten
et al., 2010; Shamir, 2007). Furthermore, there is an explicit recognition that the
dyadic partner of leaders and followers influence their characteristics and behaviours
in any given relationship (Shamir, 2007). This comprehensive and balanced
approach to leadership, therefore, strongly suggests that leadership research needs to
turn its focus to understanding how both leaders and followers jointly contribute to
the leadership process.
Limitations in Leadership/Followership Research:
Proposal of an Integrated Research Model
As the literature review demonstrates, leadership has traditionally been
approached from a leader-centric perspective. Even the focus in today’s current
mainstream leadership theories remains on the “heroic leader” while the followers’
impact on the leadership process is still relatively ignored. These issues are echoed
by practitioners who recognise there are problems in failing to account for the
follower in leadership training programmes (e.g. Hosking, 2002; Schyns, Kiefer,
72
Kerschreiter, & Tymon, 2011; Schyns, Tymon, Kiefer, & Kerschreiter, 2012). It is
encouraging though to observe an increase in followership research to help
understand the role that followers and following have on the leadership process –
where there has been a “reversing of the lens” (Shamir, 2007, p. X). Certainly, more
recent conceptualisations recognise the interdependent nature of leadership and
followership; seeing leadership outcomes as resulting from the leader-follower
relationship, of which, is jointly influenced by leader and follower beliefs and
behaviours (Lord & Brown, 2001). This reconceptualization towards a more holistic
view of leadership is required to move the field of leadership forward.
This need for integration in leadership theory formed the basis for the
development of a general research model. This model addressed three issues
observed in the literature review while also acknowledging the more recent dynamic
conceptualisations of leadership (e.g. Lord & Brown, 2001; Shamir, 2007). The first,
which formed the basis for approaching the remaining two issues, is that relationship
quality is usually only measured from the perspective of follower without
consideration of leaders’ perspectives (Sin et al., 2009; Wilson et al., 2010). The
evidence is clear though that there are problems with the assumption in LMX theory
that either a leader’s or follower’s perspective of their relationship is indicative that
their dyadic partner holds a similar view (Gerstner & Day, 1997; Graen & Uhl-Bien,
1995; Sin et al., 2009). It is proposed, therefore, that ILTs (Lord & Maher 1993) and
IFTs (Sy, 2010) may help to explain this divergence in leader and follower views of
their relationship.
ILT and IFT are based in role theory as they specify and organise information
around, the traits and behaviours of both leaders and followers and serve to guide
observers’ interpretations of their respective actions, what perceivers attend to, what
73
they encode, and the information they retrieve when recalling leader or follower
related information (Eden & Leviathan, 1975; 2005; Epitropaki & Martin, 2004;
2005; Lord & Maher, 1990; 1993; Lord et al., 1984; Phillips & Lord, 1982; Poole et
al., 1989; Sy, 2010). ILTs and IFTs, therefore, bias perceivers’ attention towards,
and act as a basis for judgement for, specific leaders and followers actions
respectively (Lord & Maher, 1993; Sy, 2010). Interactions between leaders and
followers should make salient, at both a conscious and unconscious level, their ILT
and IFT. Indeed, the behaviours of people change in the face of authority (Chaleff,
2009; Milgrim, 1965). Consequently, ILTs and IFTs should identify what is
specifically prevalent in the minds of leaders and followers when they interact, and
how they observe, judge, and behave in that relationship. These differences should
help too, at least in part, explain leader and follower LMX variations.
Second, this dissertation seeks to add to the limited amount of research
undertaken on ILT and IFT antecedents (c.f. Keller, 1999; Kruse & Sy, 2011; Schyns
& Meindl, 2005). In particular, what individual factors predict variations in
followers’ ILTs and leaders’ IFTs. Specifically, previous ILT and IFT research have
not considered the experiential components of personality as potential antecedents.
However, there are theoretical reasons to expect that they will be related. Kruse and
Sy (2011) found that manipulating affect has the potential to influence IFTs.
Specifically, increased positive affect increased people’s expression of prototypical
followership traits while negative affect increased people’s expression of anti-
prototypical traits. However, given the more transient nature of affect, these changes
are likely to be somewhat short-lived. Consequently, once the affect has worn off,
people are likely to revert to their original IFTs.
74
From the perspective of CEST, affect is an outcome of underlying cognitive
processes occurring in the experiential system (Epstein, 1998; 2003; 2014). Changes
in these underlying cognitive processes, therefore, should provide more stable and
long-lasting changes in IFTs. This is because it changes the underlying cognitive
processes that contribute to leaders’ affective experiences. Theoretically, the same
should also apply to followers’ ILTs. If these cognitive processes were to relate to
leaders’ and followers’ beliefs, then it would provide a framework for better
understanding the interplay of factors influencing the nature of the leader-follower
relationship. In turn, greater understandings of these contributing factors may lead to
opportunities for these beliefs to be changed in both leaders and followers via
training (see Schyns & Meindl, 2005). This also suggests that leadership outcomes
could not only be improved by targeting the leader, but also follower characteristics
and beliefs.
The third issue is that LMX antecedents have been less researched relative to
its consequences (Mahsud et al., 2010). There are reasons to believe that both
leaders’ and followers’ information processing styles (Norris & Epstein, 2011) and
degree of constructive thinking (Epstein, 2001) should be linked to their dyadic
partners’ LMX rating. Specifically, an individual’s information processing style and
constructive thinking have significant influences on individuals’ behaviours across a
wide range of contexts and situations (see Epstein, 2003; 2014; Epstein et al., 1996;
Norris & Epstein, 2011; Pacini & Epstein, 1999a). In fact, CEST states that much of
our behaviour appears to be influenced by schemas in the experiential system
(Epstein, 1994; 1998; 2001) including those occurring in the leader-follower
relationship. Consequently, the general stability of these personality constructs
should influence the general behaviours that are manifested in leader-follower
75
relationships (e.g. see. Cerni, Curtis, & Colmar, 2012; Curtis & Lee, 2013). A key
aim of the current research, therefore, was to investigate if both leader and follower
information processing style and constructive thinking, as described within CEST
(Epstein, 2003; 2014), would predict their dyadic partner’s LMX rating.
Proposal of a Research Model and Outline of Chapters 3 to 6
Throughout the above is the recognition that the three gaps observed in the
literature can be overcome by integrating and expanding upon leadership theory in a
holistic and integrative way which is aligned with modern conceptualisations of
leadership (e.g. Lord & Brown, 2001). Specifically, LMX, ILT, and IFT theories and
CEST appear to show a high degree of overlap. Integrating these may, therefore,
help explain variations in followers’ ILTs, leaders’ IFTs, in addition to differences in
leaders’ and followers’ LMX. A graphical integration of these theories is shown in
Figure 2.1.
76
Figure 2.1. Research model integrating Cognitive Experiential Self Theory, Implicit
Leadership Theory, Implicit Followership Theory and Leader-Member Exchange
(LMX).
This model served as a guide for conceptualising the research undertaken
across the next four chapters. It was based on the identification of four independent,
but fundamentally inter-related, research questions aligned with the limitations
discussed above. Study 1 (top left-hand relationship) sought to investigate whether
followers’ information processing style and constructive thinking influence their
ILT. Study 2 (top right-hand relationship in Figure 2.1) sought to understand how
leaders’ information processing style and constructive thinking influences their IFT.
Study 3 sought to investigate whether leaders’ information processing style and
constructive thinking influence their followers’ LMX. In addition, given that
follower ILTs serve to guide their interpretations of leaders’ actions (Lord & Maher,
77
1993), what they attend to, what they encode, and the information they retrieve when
recalling leader related information (Lord et al., 1984: Lord & Maher, 1990; Phillips
& Lord, 1982), study 3 also sought to investigate whether the followers’ ILT would
moderate these relationships. Study 4 sought to investigate whether followers’
information processing style and constructive thinking influence their leaders’ LMX.
It also sought to investigate whether the leaders’ ILT would moderate these
relationships.
78
CHAPTER 3
ILT Congruency: Exploring the Relationships between Followers’ Information
Processing Style, Constructive Thinking, and Implicit Leadership Theories
As noted in the previous chapter, there has been an increasing level of
attention towards a follower-centric model of leadership in recent years, which
acknowledges that effective leadership is not solely dependent on leaders’ traits and
behaviours but also on how followers socially construct leaders (Felfe, 2005).
Consequently, the role of followers’ perceptions and expectations of the leadership
process is emphasised (Foti & Lord, 1987; Kenney, Schwartz-Kenney, & Blascovich
1996; Larson, 1982; Lord & Alliger, 1985; Lord & Maher, 1993; Lord et al., 1984).
Implicit Leadership Theories (ILTs) are an individual’s personal assumptions of the
traits and abilities of leaders (Eden & Leviathan, 1975; Lord & Maher, 1993; Lord et
al., 1984). They are stable dynamic cognitive schemas, or knowledge structures,
which specify and organise information around, the traits and behaviours of a leader
(Epitropaki & Martin, 2004; 2005; Poole et al., 1989). Thus, they serve to guide
people’s interpretations of leaders’ actions (Lord & Maher, 1993), what a perceiver
attends to, what they encode, and the information they retrieve when recalling leader
related information (Lord et al., 1984: Lord & Maher, 1990; Phillips & Lord, 1982).
Previous ILT research has primarily focused on their content (e.g. Deal &
Stevenson, 1998; Lord et al., 1984), their impacts on the measurement of leadership
(e.g. Bass & Avolio, 1989; Eden & Leviatan, 1975; 2005; Nye & Forsyth, 1991;
Rush et al., 1977), and organisational outcomes (Hansbrough, 2005; Nye, 2005). An
important study conducted by Offerman and colleagues (1994), sought to uncover if
there was a core set ILT dimensions held by people. They found evidence for eight
ILT dimensions, which they labelled sensitivity, dedication, charisma,
79
attractiveness, masculinity, intelligence, strength and tyranny. Inter-individual
variations are observed across these (e.g. Epitropaki & Martin, 2004; Koommoo-
Welch, 2008; Offerman et al., 1994). Furthermore, differences are observed across
gender, with males rating ideal leaders as less sensitive and more tyrannical than
females (Deal & Stevenson, 1998; Epitropaki & Martin, 2004).
Only a small body research though has been published on what individual
factors predict variations in ILTs (c.f. Keller, 1999; Schyns & Meindl, 2005). The
research (Keller, 1999; Epitropaki & Martin, 2004; 2005) which has been conducted
has yet to consider the rational and experiential processing systems as posited in
Cognitive-Experiential Self Theory (CEST; Epstein, 2003) as different components
of personality. As Epstein (2003) states: “If there are two different information-
processing systems, it can only be a source of confusion to conduct research as if
there were only one, which is the customary practice (p.180). In addition, the use of
organisational samples has been limited (c.f. Epitropaki & Martin, 2004; 2005)
despite the potential importance ILTs could have on effective organisational
functioning (Schyns & Meindl, 2005). The current study, therefore, seeks to add to
this area by drawing upon CEST (Epstein, 1998; 2003) and person-supervisor fit
literature to explain variations in ILTs in a general working population. It is
generally posited that followers develop preferences for traits in leaders that are
congruent with their preferred information processing style (Epstein et al., 1996;
Norris & Epstein, 2011) and the degree to which they process information
constructively (Epstein, 1998; 2001).
80
Person-Supervisor Fit: The Importance of Similarity
Person-environment fit (P-E fit) is defined as generally occurring when there
is a compatibility, or congruence, between an individual and the work environment
(Kristof-Brown, Zimmerman, & Johnson, 2005). From this broad definition, several
conceptualisations of P-E fit have emerged in the literature including: (1) person-
vocation fit; (2) person-job fit; (3) person-organisation fit; (4) person-group fit; and
(5) person-supervisor fit (Kristof, 1996; Kristof-Brown et al., 2005). Whilst the
majority of research has focused on the initial four conceptualisations, the most
relevant for the current research is person-supervisor fit (P-S fit). This has been
defined as the compatibility between an employee and their supervisor (Kristof-
Brown et al., 2005). There have also been multiple ways of conceptualising and
measuring fit (see Caplan, 1987; Judge & Ferris, 1992; Kristof-Brown et al., 2005).
The most relevant is in terms of needs-supply which broadly refers to where leaders
meet or address followers’ needs, values and preferences (Caplan, 1983; 1987;
Kristof, 1996).
The similarity between leaders and followers has been demonstrated as a key
factor in the quality of leader-follower relationships (Bauer & Green, 1996),
performance (Wexley & Pulakos, 1982), and job satisfaction (Turban & Jones,
1988). Schaubroeck and Lam (2002) note that similarities in leader-follower
attitudes, dispositions, values, goals and intentions are typically more prevalent in fit
assessments than more visually obvious demographics such as gender, age or race.
When personality traits are shared between leaders and followers, they are more
likely to successfully work together due to a set of common reference points in
perceiving, understanding and behaving when confronted with the same
environmental inputs (Schoon, 2008). Indeed, in a key study drawing upon similarity
81
theory to explain variations in follower ILTs, Keller (1999) found students preferred
leaders who demonstrated similar personality traits – where individuals tend to
prefer leaders similar to the self. However, she also found that neurotic individuals
who experience high levels of negative emotions (e.g. guilt, fear, anxiety) preferred
an ideal leader to be sensitive – a more complementary trait.
The above suggests that individual processing style preferences influence
what followers looks for in leaders; where they would prefer a leader whose traits
and behaviours are congruent with their preferred information processing style. The
independence of the two systems and the differences in processing rules also
suggests that what is necessarily congruent with one system may not be for the other.
Furthermore, followers more prone to experiencing negative affect may seek
compensatory traits in leaders. Given that emotions are the domain of the
experiential system (Epstein, 2003; 2014), it, therefore, becomes necessary to look at
its role in ILT preferences.
Theory Development
It is assumed that individuals seek to think, behave and communicate in ways
that are congruent with their information processing style (see Berne, 1961).
However, environmental demands can force people to think, behave or communicate
outside of their preferred information processing style resulting in the experience of
negative affect (e.g., see McCann & Higgins, 1988). As described within the CEST
framework, this may manifest in negative vibes (i.e. vague feelings that may exist at
the border awareness) such as agitation, irritation, tension, disquietude, queasiness,
edginess, and apprehension (see Epstein, 2003) or more overtly in the form of clearly
defined emotions (e.g. fear, anxiety, worry). Each of these is a non-satisfying need
82
state. In contrast, when demands are congruent to individuals’ processing styles, they
are likely to experience more positive vibes and emotions such as wellbeing,
calmness and positive anticipation (Epstein, 2003) – a satisfying need state (Epstein,
1998; 2014).
In the context of the current study, leaders are theorised as a potent
environmental demand source for influencing what (i.e. content) and how (i.e.
process) followers process information. This is due to leaders’ inherent, positional
power that provides them more control over the nature and frequency of
communication that occurs in leader-follower interactions (e.g. see Tost, Gino, &
Larrick, 2013; Watson, 1982). Furthermore, leaders’ dispositions have indirect
impacts on the nature of people’s experiences at work (Grojean, Resick, Dickson, &
Smith, 2004; Pirola-Merlo, Härtel, Mann, & Hirst, 2002). For example, leaders have
been found to positively and negatively influence team climate (Pirola-Merl et al.,
Hirst, 2002). According to CEST (Epstein, 2003; 2014), followers would process
incoming information during on-going direct and indirect experiences of leaders via
both the rational and experiential systems. Over time, via these direct and indirect
leader-work events, followers will come to associate specific leader traits and
behaviours which enable them to adopt (i.e. congruent with) their preferred
information processing style and those traits and behaviours which do not allow
them to adopt (i.e. are not congruent with) their style. Specifically, the positive or
negative vibes and emotions experienced create the learning opportunities for
followers to associate leaders’ traits and behaviours with their experience of
emotions. Drawing on these lines of reasoning and findings, several hypotheses
regarding the relationships between followers’ information processing style and their
ILT preferences can be made.
83
Hypothesis Development: Information Processing Style and ILTs
Followers who demonstrate a preference for a rational processing style may
come to favour leader traits associated with intelligence (e.g. intellectual, clever,
educated). An intelligent leader would tend to engage in familiar behaviours (e.g.
articulates reasoning behind decisions; is logical), which aids in communication and
understanding, and facilitate a work climate aligned to the rational follower’s ways
of processing information (e.g. allow follower time to consider their positions).
Indeed, previous research has demonstrated that rational leaders tend to be rational
persuaders (Curtis & Lee, 2013). Previous research has also demonstrated that
individuals tend to be more receptive to messages aligned with their processing style
(see Rosenthal & Epstein, 2000 as cited in Epstein, 2003). It is, therefore,
hypothesised that:
Hypothesis 1: A positive relationship between followers’ rational information
processing and their preference for leader intelligence will exist.
There are several ILT dimensions that appear more congruent with
experiential system processing, albeit for different reasons: charisma, dedication,
strength, tyranny, sensitivity and attractiveness. Charisma includes traits such as
inspiring, dynamic, and energetic while dedication includes traits such as dedicated
and goal-oriented (Offermann et al., 1994). The act of inspiring is best achieved
through the arousal of positive and strong emotions (e.g. optimism and positive
thinking) in others in order to bring about desire and commitment to achieve a goal
(see; Chemers, 1997; McColl-Kennedy & Anderson, 2002; Uhl-Bien et al., 2014;
Yukl, 1998). This is especially evident when the outcome requires a highly creative
84
approach. Indeed, charismatic leadership theory (Conger & Kanungo, 1987; House,
1977) and transformational leadership theory (Avolio, 1999; Bass, 1985; Bass &
Avolio, 1990) recognise the importance of emotional appeals to followers when
trying to influence others (Shamir, House, & Arthur, 1993). Furthermore, leaders are
in a position to facilitate or minimise the barriers to, the ‘cognitive space’ for
employees to be creative (e.g. to take risks, explore new cognitive pathways;
Oldham & Cummings, 1996). Charisma and dedication related traits appear to be
highly congruent with the experiential system due to its intimate connections to
emotion and creativity. Therefore, it is hypothesised that:
Hypotheses 2 and 3: Positive relationships between followers’ experiential
information processing and their preference for leader charisma and dedication will
exist.
The strength dimension of ILTs includes the items strong and bold
(Offerman et al., 1994). Synonyms of these items include confidence, courage and
conviction; of having a force of character to push things through (Bold, n.d.; Strong,
n.d.). In charismatic leadership, being bold, courageous, and strong are important
traits for leaders (see Conger & Kanungo, 1987; 1998). Strength demonstrates high
positive correlations with the charisma and dedication dimensions (see Epitropaki &
Martin, 2004; Offerman et al., 1994). Such leaders, therefore, may tend to use
emotional and experientially (i.e., based on experience and actual events) based
communication (e.g. see Bass, 1985; House, 1977). High experiential processors
have shown to be more receptive to messages such as personal appeals and vivid
individual cases (Rosenthal & Epstein, 2000 as cited in Epstein, 2003). Followers
85
with high experiential processing style, therefore, may prefer a strong and bold
leader. It is, therefore, hypothesised that:
Hypotheses 4: A positive relationship between followers’ experiential information
processing and their preference for leader strength will exist.
The tyranny dimension of ILTs contains, what could be reasonably described
as, the less socially desirable traits of leaders such as power-hungry, demanding,
obnoxious and pushy (Offerman et al., 1994). Leaders who adopt such traits can
actually increase their influence over followers but via intimidation rather than
respect (Cheng, Tracy, Foulsham, Kingston, & Henrich, 2013). These traits,
therefore, have increased potential to elicit more negative vibes and emotions (e.g.
fear, anxiety, worry) in followers resulting in lower levels of satisfaction and poorer
leader-follower relationships (see Schyns & Schilling, 2013). In support, Fulk and
Wendler (1982) found that leaders who were achievement-oriented in combination
with the use of arbitrary and punitive behaviours increased anxiety in subordinates.
Followers who experience more frequent and intense emotions, therefore, may not
prefer such leaders as they would be likely to trigger more frequent and intense
negative vibes and emotions compared to low experiential processors. It is, therefore,
hypothesised that:
Hypothesis 5: A negative relationship between followers’ experiential information
processing and their preference for leader tyranny will exist.
86
Sensitivity includes what could be considered as socially positive traits such
as sympathetic, helpful, understanding, warm and sincere (Offerman et al., 1994).
Such traits are intimately linked to the ability to empathise and understand the needs
of others (Mahsud et al., 2010). It is also about knowing what others are feeling
(Shamay-Tsoory, Aharon-Peretz, & Perry, 2009). Indeed, charismatic and
transformational leadership theories state that leaders should adapt their style to the
needs (task and emotional) of their followers (Avolio, 1999; Bass, 1985; Bass &
Avolio, 1990; Conger & Kanungo, 1987; House, 1977). Thus, these theories require
leaders to empathise with their followers to some degree. The development of
positive and effective relationships, which includes empathy, is associated with the
experiential system (Pacini & Epstein, 1999a). Specifically, high experiential
processors tend to be more empathic and have higher quality social relationships
(Norris & Epstein, 2011; Pacini & Epstein, 1999a). Followers with high experiential
processing styles may, therefore, prefer sensitive leaders as they tend to engage in
warmer and more supportive interactions as well as creating more socially positive
(e.g. interpersonal trust) work environments. For example, leaders with low
sensitivity would be more likely to create opportunities for conflict (e.g. see Xin &
Pelled, 2003) resulting in negative vibes and emotions experienced for high
experiential processors. Therefore, it is hypothesised that:
Hypothesis 6: A positive relationship between followers’ experiential information
processing and their preference for leader sensitivity will exist.
High experiential processers tend to be attracted to beauty and art (Norris &
Epstein, 2011). Emotional experience is intimate to the aesthetic experience (Leder,
87
Belke, Oeberst, & Augustin, 2004). Followers who exhibit high experiential
processing styles may, therefore, prefer, or be more receptive to, attractive leaders
(e.g. physical attractiveness, stylish clothes and ‘classy’ behaviours). High
experiential processors also tend to rely more on heuristic rules when making
decisions (e.g. see Denes-Raj & Epstein, 1994; Denes-Raj, Epstein, & Cole, 1995;
Kirkpatrick & Epstein, 1992; Pacini & Epstein, 1999a, 1999b). The intuitive
tendencies of high experiential processors, therefore, may see them more likely to
draw upon general leader stereotypes, which include an attractiveness element (see
Offerman et al., 1994). Consequently, they may be more receptive to attractive
leaders. It is, therefore, hypothesised:
Hypothesis 7: A positive relationship between followers’ experiential processing
and their preference for leader attractiveness will exist.
Hypothesis Development: Constructive Thinking and ILTs
As the components of constructive thinking also influence the ways in which
individuals attend to the world and assimilate information in a given moment (see
Epstein, 2003; 2014) these may also relate to followers’ ILT preferences in a similar
way to that described for information processing. Specifically, followers may
develop preferences for leader traits that are congruent with their tendency to process
information either constructively or destructively. For example, a follower who tends
to processes criticism more constructively (i.e., criticism is less likely to impact on
their self-worth) would be less likely to experience negative vibes and/or emotions.
Consequently, they would be less cognitively motivated to develop a preference for
leader traits and behaviours, which facilitate more sensitive or empathic interactions.
88
Given the large number of factors contributing to an individual’s constructive
thinking (see Epstein, 2001), a specific theoretical basis for each factor is described
below.
Individuals with low emotional coping tend to have automatic thoughts to
disappointing outcomes, which result in feelings of distress and anxiety (Epstein,
2001). Individuals with low emotional coping may, therefore, develop preferences
for traits in leaders, which serve to mitigate or avoid the destructive pathway
tendencies that result in their experience of negative affect. It may, therefore, be that
sensitive leaders, who display empathic behaviours, appeal to poor emotional copers
as the former would help remove or minimise followers’ experience of distress and
anxiety when they are under threat (e.g. high workloads, issues and problems, inter-
employee conflict; see Fogarty, Curbow, Wingard, McDonnell, & Somerfield, 1999;
Fulk & Wendler, 1982; Roter, 2000). In contrast, high emotional copers who process
information more constructively (e.g., self-accepting; not dwelling on issues) would
place less personal value on sensitive leaders due to better emotional coping skills.
Indeed, given that teams and organisations exist at the most fundamental level to
perform in some way (e.g. financially, unit sales) some strong emotional copers may
even prefer less sensitive leaders as they see them to be ‘getting the job done’ (i.e.
task focused) without needing to focus on the perceived irrelevance of emotional
states to achieving team/organisational performance (e.g. see Ehrhart & Klein, 2001
for descriptions from followers of task vs. relationship based leaders). Therefore, it
can be hypothesised that:
Hypothesis 8: A negative relationship between followers’ emotional coping and
their preference for leader sensitivity will exist.
89
Whilst sensitive leaders may help and support followers with low emotional
coping to better deal with more general workplace threats to self-esteem, tyrannical
leaders may be seen as a key threat source. Tyrannical leaders are described as
conceited, power hungry, demanding, pushy and loud (Offerman et al., 1994).
Followers with low emotional coping may be particularly sensitive to such leader
due an increased probability of conflict, verbal abuse and non-supportive behaviours
(see Schyns & Schilling, 2013). Some followers may also dwell on their negative
interactions and experiences with tyrannical leaders. Such dwelling is a sub-
dimension of emotional coping (see Table 1.3 and Epstein, 2001). In contrast, good
emotional copers would be more likely to have constructive cognitions when faced
with tyrannical leaders; where they are more thick-skinned to their negative
behaviours. Some followers with good emotional coping may even perceive
tyrannical leaders positively as they see such behaviours support task completion
(see Ehrhart & Klein, 2001). Indeed, toxic leaders, who have similar traits to
tyrannical leaders, at least on the face may appear adaptive in terms of high
performance when achieving organisational goals (see Whicker, 1996). For example,
in the TV show Hell’s Kitchen, Gordon Ramsay shouts at the contestants and is
highly demanding of them in order to intimidate and motivate them to perform (van
Kleef, Homan, Beersma, & van Knippenberg, 2010). In reality, though, the actions
of toxic leaders can result in a net cost to the organisation in terms of the “carcasses
of those who work for them” (Reed, 2004, p.68). These organisational outcomes may
not always be immediately apparent to such followers given that these negative
repercussions typically are long-term and less visible (e.g. employee turnover;
commitment) than the short-term positive outcomes (e.g. a report being completed).
Therefore, it is hypothesised that:
90
Hypothesis 9: A positive relationship between followers’ emotional coping and their
preference for leader tyranny will exist.
Good behavioural copers tend to be optimistic, positive thinkers and
deliberate in terms of hard work, planning and doing one’s best and of engaging in
effective action (Epstein, 2001). Good behavioural copers also tend to be more
extraverted (Epstein, 2001). The ILT dimensions which appear to be congruent with
behavioural coping include charisma and dedication. Charismatic leaders are more
likely to act in ways that inspire and motivate followers to do their best (e.g. see
House, 1977). Highly dedicated leaders, with traits such as hard-working and goal-
orientation would serve to support a follower as they would help them to remain on
track and show commitment to the achievement of specified goals (e.g. see case
study by Toegel, Liu, Coughlan, & Perrinjaquet, 2011). Therefore, it is hypothesised
that:
Hypotheses 10 & 11: Positive relationships between followers’ behavioural coping
and their preference for leader charisma and dedication will exist.
Naïve-optimism is a form of destructive thinking where individuals think in
simplistic, stereotyped ways (Epstein, 2001). Naïve optimists are unrealistically
optimistic and think the best is going to turn out despite evidence to the contrary. For
example, they tend to endorse the belief that people should look happy no matter
what they feel or that everyone can get ahead if they work hard. It is positive
thinking at its most extreme but with the associated cost of failing to plan for the
future (Epstein, 1998; 2001). As naïve optimists tend to endorse stereotyped ways of
91
thinking, they may conceptualise an ideal leader as culturally stereotypical – as one
that is high on charisma and dedication (e.g. see Offermann et al., 1994). Therefore,
naively optimistic followers may be particularly receptive to charismatic and
dedicated leaders. In addition, such followers are also more likely to associate
attractive leaders with other positive elements (i.e. halo effect; e.g. see Verhulst,
Lodge, & Lavine, 2010). It was, therefore, hypothesised that:
Hypotheses 12, 13 & 14: Positive relationships between followers’ naïve-optimism
and their preference for leader charisma, dedication and attractiveness will exist.
Method
Participants
A total of 352 currently employed employers who were not a
manager/supervisor and were over the age of 18 volunteered to complete an online
two-stage survey in return for confidential written developmental feedback on their
constructive thinking. Of these individuals, 242 (68.8%) were female and 110
(31.3%) were male. The mean age for the overall group was 40.09 (SD = 12.36;
range 18-78). Individuals had been in their current organisation for an average period
of 70.07 months (SD = 79.49; median = 43.00). The mean age for females was 40.28
(SD = 12.66; range 18-78). The mean age for males was 39.67 (SD = 11.69; range
18-65). There were no significant differences between males and females on age,
t(350) = .425, p = .671, or organisational tenure, (male M = 65.73, SD = 6.32; female
M = 72.03, SD = 5.46), t(349) = .754, p = .452.
92
Measures
Rational/Experiential Multimodel Inventory (REIm; Norris & Epstein,
2011). The REIm (Norris & Epstein, 2011; see Appendix A) is based on the CEST
theoretical framework (Epstein, 1994; 2003) and is designed to measure the degree
to which an individual prefers to use a rational or experiential processing style. It
contains 42 items in total; 30 in the main Experiential scale and 12 in the Rational
scale. Responses are made on a 5-point Likert scale indicating how much a statement
is true of them (1 = Definitely False; 5 = Definitely True). The rational scale
measures the degree an individual uses their rational system to process information
(e.g. “I have a logical mind.”). Overall experiential processing is targeted via three
sub-scales which contain 10 items each: (1) intuitive; (2) emotionality and; (3)
imagination. The intuitive facet is designed to assess the ability and engagement of
an individual’s use of intuitive judgements (e.g. “I often go by my instincts when
deciding on a course of action”.) The emotionality sub-scale assesses the intensity,
frequency, duration, and favourable attitude towards strong affect (e.g. “My emotions
don’t make much difference in my life”; reversed scored). The imagination sub-scales
assesses the engagement in, and appreciation of, imagination, aesthetic productions,
and imagery (e.g. “Sometimes I like to just sit back and watch things happen.”)
Higher scores indicate a higher preference for processing on that particular
dimension.
Constructive Thinking Inventory (CTI; Epstein, 2001). The CTI is a
standardised test designed to measure the constructive and destructive automatic
thoughts in the experiential system (Epstein, 2001). It contains 108 self-report items
which include a global scale of constructive versus destructive automatic thinking
93
and six main dimensions. These dimensions were previously presented in Table 1.
People respond on a 5-point Likert scale how much a statement is true of them (1 =
Definitely False; 5 = Definitely True). Raw scores are entered into an electronic
scoring programme (Epstein & PAR Staff, 2008) and converted to T-Scores based
upon gender norms. Research demonstrates that the main scales demonstrate
good/excellent reliability (Epstein, 2001).
Implicit Leadership Theory Questionnaire (Offerman et al., 1994) –
adapted. The Implicit Leadership Theory Questionnaire (see Appendix B) is
designed to measure an individual’s ILT. It was developed using a rigorous
validation process in both a student and working population sample. In the original
version, individuals are asked to rate on a 7-point Likert scale (1 = Not at all
Characteristic; 7 = Extremely Characteristic) the degree to which 41 traits are
characteristic of “a leader in a business setting”. Epitropaki and Martin (2004) found
support for the evidence validity of this instrument in a variety of different samples.
The target of focus (e.g. leader, supervisor, manager or ideal leader) influences the
overall rating (e.g. Epitropaki & Martin, 2004; Offerman et al., 1994). Therefore,
specifying the target for research is important. The measure was adapted to target
“an ideal leader in a business setting” as it was theorised that a leader who matches a
follower’s ideal leader as opposed to their general leader would be a more
theoretically relevant target for congruence. In addition, a 10-point Likert Scale was
utilised as it was theorised that some of the items may exhibit a floor effect (i.e.
tyranny and masculinity) due to social desirability responding. This alternative scale
has been utilised in previous research (see Koommoo-Welch, 2008). Higher scores
are assumed to indicate a preference for a particular leader dimension.
94
Procedure
A website was developed to support and promote the research. This website
provided information about the research in addition to a link to the survey.
Participants were recruited via several methods. First, the website address along with
some brief information was posted across various social media outlets inviting
individuals to participate. Reposts of these invitations were encouraged to increase
research visibility. Second, invitations were sent to those on industry body and
university alumni e-mail distribution lists. Third, at the end of the survey,
participants were encouraged to invite others to participate (snowball sampling).
Fourth, printed media was posted at local outlets (e.g. libraries, shopping centres),
and small printed leaflets prompting the research were also left at approximately 30
locations where verbal permission was provided (e.g. cafes, office receptions). The
survey remained open for a period of about 6 weeks.
Common method bias has the potential to inflate or deflate any relationship
found between variables because of the use of a single method for gathering the data
(see Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). In order to control for some of
the effects of common-method bias, the survey was split into two stages. However,
these issues may still remain and, therefore, were investigated via statistical means in
the results. In stage one, participants completed the REIm (Norris & Epstein, 2011),
the adapted Implicit Leadership Theory Questionnaire (Offerman et al., 1994) in
addition to some demographics. An ID code was generated for each individual who
completed stage one. After completing the first stage, participants were sent an email
with their ID code two days later, which invited them to complete the second survey
stage. The average time between the completion of stage one and stage two was 5.28
days (SD = 3.88, range 2 - 30). In stage two, participants completed the CTI
95
(Epstein, 2001) only. All participants had the option to receive confidential written
feedback based upon their responses to the CTI (Epstein, 2001). This was offered as
it was assumed this would a) increase the accuracy of responses (i.e. individuals
want accurate feedback) and; b) improve the probability of participants completing
both stages.
Results
Removal of defensive and inattentive participants
Two scales on the CTI (Epstein, 2001) exist to assess the validity of
responses: (1) defensiveness and (2) validity. The defensiveness scale is used to
assess the degree to which a person presents an overly positive image of oneself. The
validity scale is used to assess careless responses and failure in comprehension.
Epstein (2001) recommends that individuals scoring equal or less than -1.5 SD’s
validity scale or equal to or more than 1.5 SD’s on the defensiveness scale should be
removed. While this is in relation to the CTI measure, it was decided that individuals
meeting these criteria would be removed (all responses) from the final data set as
they may have demonstrated similar response patterns on other instruments. This
resulted in 24 individuals being removed resulting in 328 individuals for analysis.
Common method bias
If a substantial amount of common method variance is present, either (a) a
single factor will emerge from the component analysis, or (b) one general factor will
account for the majority of the covariance among the variables (e.g., Andersson &
96
Bateman, 1997; Aulakh & Gencturk, 2000; Krishnan, Martin, & Noorderhaven,
2006; Podsakoff, et al., 2003; Podsakoff & Organ, 1986). Principal components
analysis with direct oblimin rotation with SPSS v21.0 identified 51 components with
Eigenvalues greater than 1, which accounted for 77.39% of the total variance. The
first factor did not account for the majority of the variance (10.24%). Harman’s
single-factor analysis which forces all items to load on a single component also
indicated that a single factor (10.24%) did not account for the majority of the
variance (Podsakoff et al., 2003). It was therefore concluded that no general factor
arising from a common method is apparent.
Exploratory Factor Analysis – ILT Measure
As the current measure used an adapted version of the ILT measure and no
prior assumptions or hypotheses were made regarding what factors would make up
“ideal” leaders (see Finch & West, 1997), an exploratory factor analysis (EFA)
rather than a confirmatory factor analysis (CFA), was undertaken to assess the
underlying factor structure. A total of 17 people had not completed the adapted ILT
measure. Therefore, the EFA uses a sample of 311 individuals. This equates to 7.59
subjects per item, which is higher than the general minimum rule of ‘five subjects-to-
variables’ ratio (see Bentler & Chou, 1987; Bryant & Yarnold, 1995). In addition, a
total of more than 300 participants is typically a good size for factor analysis (see
Comrey & Lee, 1992). Factorability was undertaken through inspection of the
correlations matrix among the 41 ILT items, the Kaiser-Meyer-Olkin measure of
sampling adequacy, and Bartlett’s test of sphericity. All items correlated at least .3
with at least one other item, suggesting reasonable factorability. In addition, the
97
Kaiser-Meyer-Olkin measure of sampling adequacy was .861, above the
recommended value of .6, and Bartlett’s test of sphericity was highly significant, χ2
(820) = 6186.49, p < .001. The diagonals of the anti-image correlation matrix were
also all over .5, supporting the inclusion of each item in the factor analysis. Given
the sample size and overall indicators, factor analysis was conducted with all 41
items.
Principal components analysis was used given that the primary purpose was
to identify and compute composite scores for each ILT factor (see Abdi & Williams,
2010). The Oblimin rotation was used as it was expected that the final factors would
be correlated (Ferguson & Cox, 1993). During an iterative process involving several
steps, a total of 18 items were eliminated because they did not contribute to a simple
factor structure and failed to meet a minimum criteria of having a primary factor
loading of .4 or above, and no cross-loading of .3 or above. While the Male item
loaded on a component similar to the original tyranny factor identified by Offerman
and colleagues (1994), a theoretical argument for its inclusion did not seem
appropriate and, therefore, was removed. All items, except for Enthusiastic, loaded
on the original factors identified by Offerman and colleagues (1994). The
Enthusiastic item loaded on the dedication component but was retained, as it was
conceptually similar to the other items. A principal-components factor analysis of the
22 items indicated five theoretically interpretable factors, which explained 61.89% of
the variance. These five factors were sensitivity, tyranny, intelligence, attractiveness
and dedication. All items had primary loadings over .5 and no items had cross-
loadings above .3. The factor loading matrix for this final solution is presented in
Table 3.1. Descriptive statistics, Cronbach’s alpha and Pearson’s correlation
coefficients for all the scales are presented in Table 3.2. All scales showed
98
acceptable to good reliability. As the Masculinity, Strength and Charisma ILT
dimensions did not emerge in the EFA, hypotheses 2, 4, 10 and 12 were not tested.
Means and standard deviations on each of the new ILT dimensions for males and
females are also shown in Table 3.2.
Table 3.1. Direct Oblimin Rotated Pattern Matrix of the 22-Item Adapted ILT Measure (n = 311)
Items Sensitivity Tyranny Intelligence Attractiveness Dedication Sensitive .763 -.005 .047 -.018 .040 Compassionate .763 -.145 -.080 .031 -.034 Sympathetic .735 .011 -.100 -.112 .023 Warm .730 .075 .149 -.060 -.023 Sincere .527 -.086 .095 .101 -.273 Selfish -.019 .831 -.034 .045 -.026 Manipulative -.118 .814 .007 .124 -.035 Power-Hungry .034 .777 -.112 .035 -.033 Conceited -.118 .667 .040 -.160 .122 Loud .119 .544 .111 -.153 -.032 Intellectual -.094 -.080 .847 -.170 .126 Knowledgeable -.032 -.072 .743 .009 -.121 Clever .147 .172 .742 .021 -.027 Intelligent .040 -.020 .739 .152 -.121 Well-dressed -.019 -.030 .021 -.902 -.119 Well-groomed -.032 -.038 -.066 -.897 -.162 Classy .220 .133 .074 -.707 .082 Motivated .063 .047 -.098 -.017 -.791 Dedicated .080 -.041 .016 -.072 -.744 Goal-Oriented -.103 .063 .009 -.160 -.728 Enthusiastic .208 .024 .064 .082 -.660 Hard-working -.103 -.066 .182 -.018 -.636 Variance Explained 23.95% 14.90% 9.74% 6.96% 6.34% Note: Factor loadings above .300 are bolded to aid readability
99
Table 3.2.
Means, Scale Reliabilities, and Inter-correlations of Scales on 22 Item Adapted ILT Measure
Scale Means (SD) ILT Factors
Overall
(n = 311)
Male
(n = 91)
Female
(n = 220)
1 2 3 4 5
1. Sensitivity 7.83 (1.19) 7.88 (1.18) 7.81 (1.19) (.79) -.12* .23** .27** .43**
2. Tyranny 2.34 (1.23) 2.42 (1.27) 2.30 (1.22) (.78) .01 .26** -.02
3. Intelligence 7.95 (1.27) 8.12 (1.03) 7.97 (1.12) (.79) .20** .43**
4. Attractiveness 6.13 (2.15) 5.83 (2.33) 6.25 (2.06) (.85) .33**
5. Dedication 8.65 (.93) 8.44 (.99) 8.74 (.87) (.80)
Note: Cronbach’s alpha reliabilities are shown on the diagonal in brackets. All inter-correlations use the overall group.
*Indicates p < .05; **Indicates p < .01
100
Data screening & assumption checking
Due to the potential for outliers to influence inferential statistical techniques
(Osborne & Overbay, 2004), prior to inferential analyses, data for all variables were
checked. Any data point that was more than 1.5 times the Interquartile range (IQR)
above or below the first or above the third quartile was classified as an outlier (see
Tukey, 1977). In order to maintain statistical power, outliers were recoded
(truncated) to the nearest highest/lowest non-outlier score as this method retains the
relative order of scores but reduces distribution issues (see Osborne & Overbay,
2004). A total of 124 outliers across all scales were recoded. No more than 17
outliers on any single scale were recoded (M = 3.9 items per factor recoded). While
some minor normality issues remained on some of the scales, the inferential statistics
used are generally quite robust against this (Tabachnick & Fidell, 2007). The tyranny
dimension, however, did violate the normality assumption strongly with a strong
positive skew. Therefore, analyses including this dimension used non-parametric
procedures (Allen & Bennett, 2010). In order to assess linearity and
homoscedasticity, bivariate scatter plots were obtained for the remaining variables.
These indicated the assumptions of both linearity and homoscedasticity had been
met.
Descriptive Statistics
Descriptive statistics for the REIm at both overall group and gender are
shown in Table 3.3. An independent samples t-test was used to compare mean
rational and experiential processing style between males (n = 100) and females (n =
228). Levene’s test was significant for the rational factor (p = .014) indicating equal
101
variance assumption had been violated and, thus, an adapted t-test was used. The
Levene’s test for all other factors did not reach significance. All dimensions except
imagination showed significant differences between males and females in directions,
which were aligned with the research by Norris and Epstein (2011). Males showed
higher rational processing than females, t(163.159) = 3.029, p = .003. In contrast,
females showed higher experiential processing, t(326) = 4.639, p < .001, intuitive
processing, t(326) = 2.142, p = .033, and emotionality, t(326) = 5.962, p < .001, than
males. No significant difference was observed between males and females on the
imagination sub-scale, t(326) = 1.694, although the difference did approach
significance, p = .091.
Table 3.3.
Descriptive statistics on REIm factors for overall group and as a
factor of gender
REIm Scale Means (SD)
Overall
(n = 328)
Male
(n = 100)
Female
(n = 228)
1. Rational 3.88 (.55) 4.03 (.61) 3.82 (.51)
2. Experiential 3.48 (.38) 3.34 (.35) 3.55 (.38)
3. Intuition 3.51 (.50) 3.42 (.48) 3.54 (.50)
4. Emotionality 3.32 (.56) 3.06 (.51) 3.44 (.54)
5. Imagination 3.63 (.55) 3.55 (.52) 3.66 (.56)
Except for tyranny, an independent samples t-test was used to compare the
means between males (n = 91) and females (n = 220) on each ILT factor (see Table
3.2). No significant differences were observed on sensitivity, t(309) = .484, p = .629,
intelligence, t(309) = 1.14, p = .255, or attractiveness, t(309) = 1.580, p = .115.
However, females showed a significantly higher score than males on dedication,
t(309) = 2.717, p = .007. There was no significant difference on tyranny between
102
males (Mean Rank = 162.07) and females (Mean Rank = 153.49) as indicated by a
Mann-Whitney U-test, U = 9458.00, z = .767, p = .433.
Hypothesis Testing
While structural equation modelling (SEM) is a powerful and flexible
statistical technique due to its ability to deal with a system of regression equations as
opposed to a single or multiple regression technique, it is recommended that the
sample size be more than 25 times the number of parameters to be estimate (see
Nachtigall, Kroehne, Funke, & Steyer, 2003). It was estimated that the number of
parameters that would have to be specified in a SEM model for the current research
would be 19. Consequently, a sample of 475 would be recommended. As the current
sample size does not meet this sample size requirement, a decision was made to use
correlations and Hierarchical Multiple Regression for hypothesis testing.
Bivariate Pearson’s product-movement correlation coefficients were
calculated between the rational and experiential scales with the four normally
distributed ILT dimensions: sensitivity, intelligence, attractiveness and dedication.
Due to the non-normal data, Spearman’s rho was used to assess the hypothesised
relationships with tyranny. Table 3.4 presents correlations for demographics, REIm
and CTI factors, against ILT factors.
103
Table 3.4.
Correlations between REIm (n = 311), CTI (n = 273) and ILT Dimensions
ILT Dimensions
Sensitivity Tyranny˄ Intelligence Attractiveness Dedication
Demographic
Gender& -.03 -.04 -.08 .09 .16**
Age .13* -.18** -.16** -.01 .05
REIm
Rational -.02 .04 .35** -.05 .02
Experiential .33** -.01 .01 .13* .29**
CTI
Global Constructive
Thinking .05 .04 -.25** -.01 .02
Emotional Coping -.07 .02 -.30** -.02 -.05
Behavioural Coping .19** .08 -.06 .06 .13*
Personal
Superstitious
Thinking
-.08 .03 .10 .06 -.06
Categorical Thinking -.10 .09 .09 .23** -.02
Esoteric Thinking .08 .11 -.03 .21** .06
Naïve-Optimism .10 .26** -.02 .23** .13*
Notes: *Indicates p < .05; **Indicates p < .01; ˄Relationships indicated by Spearman’s rho. &Dummy variable coded for Gender; Male = 0 and Female = 1.
REIm and ILT Relationships
As predicted by hypothesis 1, a significant positive relationship was observed
between rational processing and intelligence. No other significant correlations for
rational processing emerged with the remaining four ILT dimensions. Hypothesis 3
was supported as shown by the significant positive relationship between experiential
processing and dedication. Hypothesis 5 was not supported as indicated by the non-
significant correlations between experiential processing and tyranny. Hypothesis 6
was supported as demonstrated by the significant positive relationship between
experiential processing and sensitivity. Hypothesis 7 was supported as demonstrated
by the significant positive relationship between the experiential system and
attractiveness.
104
CTI and ILT Relationships
There were 38 individuals who did not complete stage two. Therefore,
analyses made using the CTI are based on a sample of 273. In order to test the
hypotheses concerning the CTI and ILT, bivariate Pearson’s product-movement
correlation coefficients and Spearman’s rho were calculated and are also shown in
Table 3.4. Surprisingly hypothesis 8, which predicted a significant negative
relationship between emotional coping and sensitivity, was not supported.
Hypothesis 9 was not supported as indicated by the non-significant positive
correlation between emotional coping and tyranny. Hypothesis 11 was supported as
shown by the significant positive relationship between behavioural coping and
dedication. Hypothesis 13 was supported by the significant positive relationship
between naïve-optimism and dedication. Hypothesis 14 is supported as shown by the
significant positive relationship between naïve-optimism and attractiveness.
Non-hypothesised relationships
A number of other significant relationships between the CTI main scales and
ILT dimensions, which were not specifically hypothesised, were also observed. First,
there was a significant positive correlation between behavioural coping and
sensitivity. Second, a significant positive correlation emerged between naïve-
optimism and Tyranny. Third, a surprising but theoretically coherent, negative
relationship was observed between global constructive thinking and intelligence.
Fourth, a significant positive relationship between categorical thinking and
attractiveness. Fifth, a significant negative relationship was observed between
emotional coping and intelligence. Finally, a significant positive relationships were
observed between esoteric thinking and attractiveness.
105
Hierarchical Regression Analyses
To further examine the relationships between constructive thinking,
information processing with ILT dimensions, hierarchical multiple regression
analyses were conducted (Thompson, 1978). Bivariate scatter plots were obtained to
assess and satisfy linearity and homoscedasticity for all regressions. The variables
were all demonstrated to be linearly related. Residual scatter plots also demonstrated
that the assumption of homoscedasticity had been satisfied. The correlations of the
variables were assessed for multicollinearity but none was detected as all
correlations were below 0.9 (Tabachnick & Fidell, 2007). All analyses are based
upon n = 273. A dummy variable was used to assess the effects of gender in the
model1.
Thinking Styles and ILT Trait of Sensitivity
With Sensitivity as the dependent variable, a three-step hierarchical multiple
regression was conducted. Age and gender were entered in the first step. Experiential
processing was added in the second step. In step two, Behavioural coping was
entered. This is because one of the sub-scales for Behavioural coping is
Conscientiousness, which theoretically could be considered as congruent with
preferences of sensitivity in leaders. For example, one of the items on this scale is
“When I realise that I have made a mistake, I usually take immediate action to
correct it”. Such actions indicate a degree of sensitivity to things occurring in their
environment. The significant positive correlation between behavioural coping and
sensitivity provides evidence for this. Regression statistics are shown in Table 3.5.
1Male = 0 and Female = 1.
106
Table 3.5.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Sensitivity Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .144 .021 .021^
Gender -.096 -.037 -.622 .001
Age .013 .140 2.320* .020
Step 2 .356 .126 .106**
Gender -.309 -.120 -2.038* .013
Age .013 .142 2.493* .020
Experiential Processing .997 .335 5.706** .106
Step 3 .399 .159 .033**
Gender -.341 -.132 -2.284* .016
Age .011 .113 1.989* .012
Experiential Processing 1.014 .341 5.901** .109
Behavioural Coping .026 .184 3.226** .033
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
The overall regression model at step 1 only approached significance, R2 =
.21, F(2,270) = 2.857, p = .059. However, Age alone significantly contributed 2.0%
to the regression model at this step. The introduction of Experiential processing in
step 2 explained an additional 10.6% in Sensitivity preference variation. This
increase in R2 was significant, F(3,269) = 12.979, p < .001. Age at this step remained
a significant predictor. Gender though also emerged as a significant predictor. The
introduction of Behavioural Coping at step 3 further contributed a significant 3.3%
in variance explained in Sensitivity trait preference, F(4,268) = 12.676, p < .001.
Gender, Age, Experiential Processing, and Behavioural coping all were significant
predictors of Sensitivity preference. The most important predictor in this model was
Experiential Processing, which independently explained 10.9%. This was followed
by Behavioural coping which independently accounted for 3.3%. Overall, the four
independent variables accounted for 15.9% of the variance in Sensitivity preference.
107
Thinking Styles and ILT Trait of Tyranny
With Tyranny as the dependent variable, a two-step hierarchical multiple
regression was conducted. Age and gender were entered in the first step. In step two,
Naïve Optimism was entered. Theoretically, naïve optimists may associate tyrannical
behaviours as being adaptive for achieving team and organisational goals (see
discussion for more detail). The significant positive correlation between Naïve-
Optimism and Tyranny observed above provides evidence for this. Regression
statistics are shown in Table 3.5.
Table 3.5.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Tyranny Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .150 .022 .022*
Gender -.070 -.026 -.428 .001
Age -.014 .006 -.147* .022
Step 2 .265 .070 .048**
Gender -.087 -.032 -.543 .001
Age -.013 -.133 -2.251* .018
Naïve Optimism .026 .219 3.724** .048
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
The overall regression model at step 1 was significant, R2 = .21, F(2,270) =
3.096, p = .047, although age alone (2.2%) significantly to the regression model at
this step. The introduction of Experiential processing in step 2 explained an
additional 4.8% in Sensitivity preference variation. This increase in R2 was
significant, F(3,269) = 6.786, p < .001. With all three independent variables entered,
Age and Naïve Optimism were the only significant predictors of Tyranny preference.
The most important predictor in this model was Naïve Optimism, which
independently explained 4.8%. Overall, the three independent variables accounted
for 7.0% of the variance in Tyranny preference.
108
Thinking Styles and ILT Trait of Intelligence
With Intelligence as the dependent variable, a three step hierarchical multiple
regression was conducted. Age and gender were entered in the first step. Rational
processing was added in the step 2. In step 3, Emotional Coping was entered.
Theoretically, intelligent leaders may be seen by poor emotional copers, as being
able to compensate for their increased tendency to interpret issues and problems via
destructive cognitive pathways. Rational thinking, which itself is associated with
intelligence (Epstein, 2001) is positively associated with leader effectiveness (Cerni
et al., 2010), planning and problem solving (Epstein, 2001). Given that intelligence
was found to have negative correlations with neuroticism (Ackerman & Heggestad,
1997; Furnham, Forde, & Cotter, 1998; Zeidner & Matthews, 2000), it is plausible to
suggest that intelligent leaders may be appreciated by followers with poorer
emotional coping as they help them to identify opportunities to minimise or mitigate
current and future workplace issues. Therefore, there is a theoretical rational to enter
emotional coping at this step. Even though Global Constructive Thinking emerged as
having a significant negative relationship, this was not entered as it would violate the
assumption of singularity (i.e. Global Constructive Thinking is, in part, made up
from items on Emotional Coping; Tabachnick & Fidell, 2007). Regression statistics
are shown in Table 3.7.
109
Table 3.7
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Intelligence Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .151 .023 .023*
Gender -.157 -.056 -.924 .003
Age -.014 -.139 -2.312* .019
Step 2 .389 .151 .129**
Gender -.003 -.001 -.016 .000
Age -.009 -.086 -1.511 .007
Rational Processing .824 .367 6.386** .129
Step 3 .495 .245 .093**
Gender -.018 -.006 -.119 .000
Age .004 .041 .703 .001
Rational Processing .896 .399 7.303** .150
Emotional Coping -.043 -.330 -5.749** .093
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
The overall regression model at step 1 was significant, F(2,270) = 3.143, p =
.045. However, Age alone significantly contributed 1.9% to the regression model at
this step. The introduction of Rational Processing in step 2 explained an additional
12.9% in Intelligence preference variation. This increase in R2 was significant,
F(3,269) = 15.996, p < .001. Age at this step though became non-significant (p =
.132). The introduction of Emotional Coping at step 3 further contributed a
significant 9.3% in variance explained in Intelligence trait preference, F(4,268) =
21.690, p < .001. In this final model, only Rational Processing and Emotional coping
remained significant predictors of Intelligence preference. The most important
predictor in this model was Rational Processing, which independently explained
15.0%. Overall, the four independent variables accounted for 24.5% of the variance
in Intelligence preference.
110
Thinking Styles and ILT Trait of Attractiveness
With Attractiveness as the dependent variable, a two-step hierarchical
multiple regression was conducted. Age and Gender were entered in the first step.
Experiential processing and Naïve Optimism were added in the second st step age as
both are theorised to contribute to Attractiveness trait preference. Regression
statistics are shown in Table 3.8.
Table 3.8.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Attractiveness Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .109 .012 .012
Gender .523 .107 1.772 .011
Age .003 .020 .324 .000
Step 2 .257 .066 .054
Gender .424 .087 1.425 .007
Age .006 .034 .579 .001
Experiential Processing .328 .058 .945 .003
Naïve Optimism .047 .218 3.635** .046
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
At Step 1, the overall regression model at step 1 was not significant, F(2,270)
= 1.634, p = .197. The overall regression model though did emerge as significant at
step 2 with the introduction of Experiential Processing and Naïve Optimism adding a
significant increase of 5.4% in Attractiveness variance explained in the overall
model F(4,268) = 21.563, p < .001. However, Naïve Optimism was the only
significant contributor in the final model, explaining 4.6% of the variance in
Attractiveness preference. Overall, the four independent variables accounted for
6.6% of the variance of Attractiveness.
111
Thinking Styles and ILT Trait of Dedication
With Dedication as the dependent variable, a three step hierarchical multiple
regression was conducted. Age and Gender were entered in step 1. Experiential
processing was added in step 2 as, in addition to theoretically related to Dedication,
it emerged as the strongest correlation above. In step 3, Behavioural Coping and
Naïve Optimism were entered to see if these would provide additional explanation
beyond Experiential Processing alone. Regression statistics are shown in Table 3.9.
Table 3.9.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Dedication Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .194 .038 .038**
Gender .372 .179 2.993 .032**
Age .005 .072 1.209 .005
Step 2 .306 .093 .056**
Gender .247 .119 1.985 .013*
Age .006 .074 1.273 .005
Experiential Processing .585 .244 4.070 .056*
Step 3 .331 .110 .016^
Gender .235 .113 1.893 .012^
Age .005 .063 1.070 .004
Experiential Processing .567 .236 3.909 .051**
Naïve Optimism .011 .096 1.589 .008
Behavioural Coping .006 .067 1.113 .004
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
The overall regression model at step 1 was significant, F(2,270) = 5.283, p =
.006. Gender alone significantly contributed 3.2% to the regression model at this
step. The introduction of Experiential processing in step 2 explained an additional
and significant 5.6% in Dedication preference variation, F(3,269) = 9.247, p < .001.
Gender at this step also remained a significant contributor. The introduction of Naïve
Optimism and Behavioural Coping at step 3 contributed a non-significant increase of
1.6% of variance explained. However, the overall regression model remained
112
significant, F(5,267) = 6.586, p < .001. The only significant predictor in the final
model was Experiential processing, which independently explained 5.1%. However,
Gender did approach significance (p = .059). Overall, the five independent variables
accounted for 11.0% of the variance in Dedication preference.
Discussion
Using the CEST framework (Epstein, 2003; 2014), the theory that followers
develop ideal ILTs that are congruent with their information processing style and the
degree to which they think constructively was tested. The results observed in this
study generally support this. Theoretical and practical implications are considered.
Information Processing Style – ILT Congruence
It was assumed that individuals seek to think, behave and communicate in
ways which are congruent with their preferred information style and that when they
are required to communicate outside of their preferred style, they are likely to
experience negative vibes and emotions (see Berne, 1961; McCann & Higgins,
1988). Over time followers would learn to associate particular leader traits that have
contributed to interactions and working environments that are congruent or
incongruent with how the follower tends to processes information. Specifically,
followers come to develop preferences for leader traits that are congruent with either
the emotionally driven experiential system or the affect-free rational system. What is
important to recognise here is that, due to the independence of the two systems and
different information processing rules, traits that are congruent with one system may
not be for the other. Thus, congruence is based upon the rules for each system.
113
As expected, high rational processors tended to prefer intelligent leaders – a
non-affect laden ILT trait. Particularly important was the finding that the rational
system was not significantly related to the more affect laden ILT factors (e.g.
sensitivity, attractiveness, dedication). Drawing upon the congruence argument,
followers with a strong rational processing style prefer the behaviours intelligent
leaders tend to exhibit. Specifically, such leaders are likely to engage in familiar
behaviours for the follower such as providing a logical rationale to the follower for
their decisions and actions. Such behaviours facilitate communication and
understanding for the follower as they are aligned with the processing preferences of
these followers. Indeed previous research demonstrates that the more aligned a
message is to an individual’s processing style, the more receptive that individual will
be to that message (see Rosenthal & Epstein, 2000 as cited in Epstein, 2003). In
addition, intelligent leaders are likely to facilitate work climates that are congruent
for highly rational followers. For example, intelligent leaders are likely to provide
more time and space to followers in general to consider things. Such behaviours and
climates that intelligent leaders are likely to exhibit and bring about are likely to be
aligned to followers with a high rational processing style.
As would be expected, ILT dimensions that appear to generally incorporate
affective components such as sensitivity and dedication were more likely to be
preferred by followers with a higher experiential system processing style. Sensitive
leaders are likely to engage in more empathic leader-follower interactions and
facilitate more positive participatory work climates (e.g. see Cummings et al., 2010;
Kellett, Humphrey, & Sleeth, 2002; Rotemberg & Saloner, 1993). Experiential
processing is associated with empathy and positive relationships (Pacini & Epstein,
114
1999a) suggesting that followers with high experiential processing styles will tend to
prefer a working relationship and climate that is relationship-focused.
Similarly, the traits of dedicated leaders appear more congruent with
followers who demonstrate a high experiential processing style. This suggests that
dedicated leaders, with their higher levels of motivation and enthusiasm, appeal to
the emotional nature of the experiential system. Indeed, individuals with a high need
for affect tend to be more receptive to persuasive messages that are affect-based
(Haddock, Maio, Arnold, & Huskinson, 2008). In addition, the focus on goals in
dedicated leaders may help them to be future focused, which would appeal to the
imaginative tendencies of high experiential processors. Specifically dedicated
leaders appeal to followers’ imagination by promoting a positive and better state of
reality that currently does not exist. It is important to note that transformational and
charismatic leadership theories have been conceptualised as appealing to follower
emotions (e.g. Uhl-Bien et al., 2014). Given that transformational and charismatic
leadership theories incorporate similar traits to those on the sensitivity and
dedication dimensions, the current research suggests that transformational and
charismatic leaders would appeal more strongly to followers with experiential
processing styles. Future research, therefore, may want to test this proposition.
Given that tyrannical leaders have the potential to trigger negative affect in
followers (Schyns & Schilling, 2013) it was a little surprising that no relationships
between experiential thinking and tyranny emerged. It is not immediately clear as to
why no relationship may exist although two explanations are proposed. First, the
original measure of tyranny (see Offerman et al., 1994) included traits more focused
on domination and being demanding on others. These appear to be more targeted
towards what the leader expects from their followers. The EFA conducted in the
115
current research dropped most of these types of items, as they did not contribute to a
simple factor structure. Therefore, tyranny here may not actually reflect tyrannical
behaviours per se but instead is more aligned with leader self-interest/superiority
(e.g. selfish, conceited, power-hungry). Consequently, the impact of such behaviours
may be less on followers than traits such as domineering and demanding. Second, it
may be due to the constructiveness of the processing in the experiential system
around emotions rather than a specific preference for experiential processing. This is
because preferences for the experiential system does not necessary mean that the
content of the processing in this system is adaptive. Thus, relationships are cancelled
out due to some followers’ processing information about a tyrannical leader more
adaptively but others more destructively.
As expected, followers who exhibit a high experiential processing style
tended to prefer attractive leaders. High experiential processors tend to rely more on
heuristic rules when making decisions (e.g. see Denes-Raj & Epstein, 1994; Denes-
Raj et al., 1995; Kirkpatrick & Epstein, 1992; Pacini & Epstein, 1999a, 1999b). All
things being equal, attractiveness tends to be used as a heuristic when rating others
(Verhulst et al., 2010). Consequently, physically attractive and stylish leaders are
likely to appeal to the processing tendencies of followers with high experiential
processing as these individuals tend to be more open to, and influenced by aesthetics
(Leder et al., 2004; Norris & Epstein, 2011). The regression analyses though
suggests that experiential processing is less important than Naïve Optimism for
predicting attractiveness preferences. As Naïve optimism is a constructive thinking
dimension though, this distinction is discussed in more detail in the following sub-
section.
116
Constructive Thinking – ILT Congruence
The current study also provides evidence that followers seek traits in leaders
that are congruent with their own constructive thinking. According to CEST,
individuals have tendencies to process incoming information constructively or
destructively which have subsequent implications on the emotions experienced and
evaluations of the situation and target objects (Epstein, 1998; 2001). Such tendencies
are, therefore, likely to influence how people process information regarding the
leader, which would subsequently influence their preferences for leader traits.
Various elements of constructive and destructive processing tendencies were tested
in the current research as operationalised in the CTI (Epstein, 2001).
There was limited evidence to suggest that poor emotional copers would seek
sensitive leaders – a potentially complementary trait. This may appear somewhat
inconsistent with the research conducted by Keller (1999) who found that neurotic
students preferred sensitive leaders. Exploratory analyses (see Appendix E), though,
did suggest that males who are poor emotional copers, especially those who were
sensitive to disappointment, rejection, criticism and disapproval, did seek a sensitive
leader. These differences may be due to general tendency for males to be more
individualistic and competitive (Eagly & Carli, 2007; Newman, Groom, Handelman,
& Pennebaker, 2008), the expectation that males should be strong, to not express
their feelings or emotions (e.g. see Johnson, Murphy, Zewdie, & Reichard, 2008;
Plant, Hyde, Keltner, & Devine, 2000; Williams & Best, 1990) and the tendency to
define themselves in terms of their job (e.g. being the “bread-winner”; Eagly &
Wood, 1991). These would place strong perceived demands on males in the work
context.
117
For males, therefore, a non-sensitive leader would likely create more direct
and indirect opportunities for a male follower to engage their destructive pathway
tendencies and consequently experience negative vibes and emotions. For example, a
non-sensitive leader may engage in less empathic interactions and create a more
task-focused work-climate, both of which may be interpreted as reflecting a negative
self-concept. They may even feel trapped in such roles due to the restrictive male
role expectations and worries about loss of financial security for themselves and
family if they were to leave – a version of learned helplessness (see Dweck, 1975;
Seligman & Maier, 1967). In contrast, good emotional copers, who tend to value
themselves and are less sensitive to criticism and disapproval, (Epstein, 2001) would
likely be less affected. Indeed, in some cases, some males with good emotional
coping may even associate sensitive leaders with actually undermining the
achievement of work goals as they may see the actions of sensitive leaders to be
unnecessary to the achievement of team/organisational goals.
While it may be surprising to observe good emotional copers tend to prefer
less intelligent leaders, it aligns with the notion of congruence. Theoretically,
intelligent leaders would likely be seen by poor emotional copers as being able to
compensate for their increased tendency to interpret issues and problems via
destructive cognitive pathways. Rational thinking, which itself is associated with
intelligence (Epstein, 2001) is positively associated with leader effectiveness (Cerni
et al., 2010), planning and problem solving (Epstein, 2001). Given that intelligence
was found to have negative correlations with neuroticism (Ackerman & Heggestad,
1997; Furnham et al., 1998; Zeidner & Matthews, 2000), it is plausible to suggest
that intelligent leaders may be appreciated by followers with poor emotional coping
as they help them to identify opportunities to minimise or mitigate current and future
118
workplace issues. Rational leaders, therefore, would tend to reduce the experience of
issues, and the length of time they remain cognitively prevalent, for such followers.
In addition, the intelligent leader also fulfils these followers’ needs for a stable self-
concept and for relating to others. However, their need for self-enhancement may
remain frustrated.
The second key constructive thinking dimension is behavioural coping
(Epstein, 2001). Followers with good behavioural coping appear to prefer dedicated
leaders. Good behavioural copers tend to engage in planning, positive thinking and
taking effective action (Epstein, 2001). It is, thus, an action-oriented dimension of
constructive thinking. A leader who does not, appear motivated, hard-working or sets
goals is, therefore, likely to frustrate the action oriented processing tendencies of a
follower with high behavioural coping. Thus, the tendencies of the non-dedicated
leader are incongruent with the cognitive tendencies of followers with high
behavioural coping. Consequently, over time, they would be likely to attribute the
lack of such leader behaviours with their experiences of frustration and in more
extreme cases anger at the perceived lack of direction and action by the leader.
While not specifically hypothesised, the finding that good behavioural copers
tended to prefer sensitive leaders could be considered as theoretically aligned with
the notion of congruency, in particular, the congruence around the conscientiousness
sub-scale. For example, one of the items on this sub-scale is “When I realise that I
have made a mistake, I usually take immediate action to correct it”. Such actions
indicate that high behavioural thinkers have a degree of sensitivity to things
occurring in their environment. Indeed, behavioural coping is concerned with how
effectively individuals overcome issues in the external world (Epstein, 1998; 2001).
A sensitive leader may help in this regard. Specifically, they may be sensitive to such
119
a follower’s positive intentions and behaviours and, thus, be more likely to support
them. In support, research has found that when leaders and followers show similar
levels of conscientiousness, followers tend to engage in increased in-role behaviours
(Deluga, 1998). In addition, when leaders provide support, followers tend to show
increased performance in a variety of areas (e.g. Cheung & Wong, 2011; Chiaburu,
Smith, Wang, & Zimmerman, 2014).
It is important to note that the finding of a positive relationship between
behavioural coping and sensitivity does not appear consistent with Keller’s (1999)
research, which found no significant relationship between conscientiousness and
sensitivity. There are two possibilities why this may be the case. First, as per CEST,
there are two self-concepts that exist independently in each system (Epstein, 2003).
Thus, while these may be somewhat similar in certain areas, they do not necessarily
align. For example, an individual can report being conscientious, and believe they
are, but behavioural measures suggest otherwise (e.g. see the interpretation of
Sprirrison & Gordy [1993] by Epstein, 2001). In support, the relationships observed
in previous research between conscientiousness as measured by the CTI (Epstein,
2001) and that in the NEO (Costa & McCrae, 1989; 1992) are not so large to
conclude they are measuring the same thing (r = .67; see Epstein, 1992). Second,
Keller (1999) used a student sample and, as such would likely have less experience
of business leaders that the current research sample of currently employed
individuals; indeed, the average age of the current sample was over twice that
reported in Keller’s research. The increased direct and indirect experiences with
business leaders that highly conscientiousness individuals in the current sample
would have may have contributed, over time, to stronger associations or beliefs that
sensitive leaders support, or are congruent with, their tendency to think in terms of
120
hard work, planning, and doing one's best (see Epstein, 2001). Indeed, research
shows that caring individuals also tend to be more conscientious (Epstein, 1983).
As expected, naively optimistic followers were more likely to prefer
dedicated and attractive leaders. Naïve optimists tend to endorse stereotyped ways of
thinking (Epstein, 2001) and thus may conceptualise an ideal leader as a culturally
stereotypical leader; as one who is motivated, goal focused and dedicated (e.g. see
Offermann et al., 1994) amongst other traits. Dedicated leaders, therefore, may be
seen to engage in interactions and create work-environments, which are congruent
with naively optimistic followers. This may also explain why naïve optimists prefer
attractive leaders. Specifically, naïvely optimistic followers may be more likely to
uncritically endorse the heuristic that attractiveness is an element of an effective
leader (e.g. see Epstein et al., 1996; Norris & Epstein, 2011; Pacini & Epstein,
1999b). This because Naïve Optimism indicates that such heuristic reliance will be
unrealistic to a situation, which differs to experiential processing that encompasses a
more general preference for relying on heuristics (Epstein, 2001; Norris & Epstein,
2011). Consequently, the naively optimistic follower may be more likely to
unquestionably use attractiveness as a heuristic when rating their leader when
compared to a follower who simply shows a preference towards simply a general
preference towards reliance on such heuristics.
Theoretical and practical implications
First, the research positions CEST as an important interpretative framework
for understanding individual sense-making around leaders. Unlike other specific
theoretical frameworks such as attachment theory (e.g. see Keller, 2003), the value in
121
CEST lies in the fact that it provides a coherent and inclusive framework of
personality, which is compatible with psychodynamic, cognitive and learning
theories (Epstein, 2003). For example, unlike attachment theory which is somewhat
restricted to caregivers in early childhood (e.g. see Bowlby, 1969; 1980), it is the
influence of all significant experiences throughout life which are seen to contribute
to the general properties of schemas within the experiential system (Epstein, 1998;
2003). Alternatively other theories, which are seen as directly competing and
offering differential predictions, (e.g. self-verification and self-enhancement) may
actually be overcome through the CEST framework.
Second, unlike the big-five personality traits, the ways that information
processing style and constructive thinking are conceptualised within CEST lends
them to change (see Epstein, 2003; 2014 for a discussion on techniques). ILTs might
be changed, therefore, by targeting the underpinning cognitive tendencies. If the
relationships observed are causal, then theoretically it suggests that developing
effective leadership (from a relationship based perspective such as leader-member-
exchange theory) is not only based on developing the leader but also the follower.
This has obvious practical implications in terms of leadership development
programmes whereby the follower needs to be incorporated into the leadership
development process. Currently, though, leadership interventions typically only
focus on leaders’ behaviours without acknowledging that followers have different
images of what a leader is, which subsequently influences their perceptions and
evaluations around leaders (Eden & Leviathan, 1975; Lord et al., 1984; Lord &
Maher, 1993). This is turn is likely to affect the quality of the leader-follower
relationship (Epitropaki & Martin, 2005; van Gils et al., 2010).
122
Third, there are counter-intuitive implications around selection and team
composition decisions. Alternative leadership theories and research suggests that
dedicated (e.g. charismatic leadership), sensitive (e.g. relationship-focused) and
intelligent (e.g. task-focused) leaders would be the most effective for contributing to
positive team and organisational outcomes (see Day & Stogdill, 1972; Fleishman,
1953; Fleishman & Harris, 1962; Hemphill & Coons, 1957; House, 1971; Judge et
al., 2004). Consequently, many leadership development programmes seek to select
such leaders. The current research suggests that followers who are best suited to
working with such leaders would be those who demonstrate high rational processing
styles but low emotional coping. This leader-follower combination would contribute
to more congruent and positive leader-follower relationships and thus to positive
team and organisational level outcomes (Dulebohn et al., 2012). However, high
levels of emotional coping is strongly associated with positive outcomes at the
individual level (e.g. positive self-esteem, happiness, calmness, and lower levels of
depression (see Epstein, 1987; Epstein et al., 1996; Norris & Epstein, 2011).
Therefore, placing followers with low emotional coping with intelligent leaders
creates ethical and practical issues. Specifically, intelligence in leaders contributes to
team and organisational outcomes. Therefore, choosing followers with low levels of
emotional coping to work with such leaders should further improve outcomes at the
team and organisational levels via better-quality leader-follower relationships. Yet,
while this combination benefits the team and organisation, at the individual level the
follower with poor emotional coping would tend to show high levels of destructive
cognitions and negative affect.
This counter-intuitive finding can be explained through CEST (Epstein,
2003; 2014). Specifically, both systems are independent of each other and process
123
information by different sets of rules. What is necessarily a congruent trait for one
system is not necessarily congruent with the other. However, while the two systems
are independent, they do operate in parallel and are bi-directionally interactive
(Epstein, 2014). Thus, compromises between them occur (Epstein, 2003; 2014)
suggesting that the two systems will compromise in terms of what they seek in an
ideal leader.
Limitations
The results of the current research must be interpreted within the context of
its limitation. First, the current study utilised a cross-sectional methodology, which
does not allow definitive causal statements to be made on the relationships observed
and unobserved third variables may actually be involved. Second, the current study
did not use an “implicit” measure of ILTs but rather an explicit measure. Previous
research has found different relationships between personality variables and implicit
and explicit measures of leadership styles (Schoel, Bluemke, Mueller, & Stahlberg,
2011). Some of the relationships observed, therefore, may have been influenced by
social desirability responding to some items. Research using implicit measures may,
therefore, demonstrate differential relationships compared to those observed in the
current study.
Third, situational factors were not assessed but conceptually may have
moderated the relationships observed (see Keller, 1999). However, as the target of
focus was that of an ideal leader in a more generic business setting, rather than the
participants own work environment (i.e. participants were asked to indicate the
degree to which traits are characteristic of an ideal leader in a business setting),
124
would likely result in responses less influenced by a participant’s current
organisation setting. Future research though may wish to consider controlling for the
impact of situational factors. Fourth, the nature of the CEST focuses primarily on the
experiential system (Epstein, 2003). Consequently, there is subsequently less focus
on the rational system and aspects of the rational system may be considered under
measured in the current study. At the outset of this research, this was an intentional
imbalance as, from the perspective of CEST (Epstein, 2001; 2003), research
assessing the relationships between the rational components of personality and ILT’s
have already undertaken in the literature (e.g. Keller, 1999).
Fifth, the current study adopted an approach where only one ILT dimension
was analysed at a time. This singular approach limits some of the interpretability as
results have been interpreted around the variables themselves (Foti & Hauenstein,
2007). In contrast, a pattern approach would consider the internal coherent and
consistent pattern of ILT variables that exist within followers and provide a more
holistic understanding of follower ILTs (e.g. see Foti & Hauenstein, 2007). One of
the other benefits of a pattern approach compared to a variable approach is that
patterns of ILTs may provide more stable prediction of what contributes to effective
leader-follower relationships (e.g. see Mumford et al., 2000b; Yukl, 2002).
Consequently, future research may wish to adopt a pattern approach. It is important
to note that the pattern approach is considered as a complementary approach to the
variable approach, not as a substitute. Indeed, Foti and Hauenstein (2007) state that
“the pattern approach can complement the variable approach” (p. 353).
Consequently, the current results are still important for extending ILT prediction
research.
125
CHAPTER 4
IFT Congruency: Exploring the Relationships between Leaders’ Information
Processing Style, Constructive Thinking, and Implicit Followership Theories
If followers’ information-processing style and degree of constructive thinking
are related to their implicit ideal leadership theories (see the previous chapter) then it
seems reasonable to propose that we may also observe similar relationships between
these characteristics in leaders and their Implicit Followership Theories (IFTs). IFTs
refer to those schemas that contain information on the traits and behaviours of
followers (Sy, 2010). From a theoretical perspective, IFTs are understood as an
important missing element in helping to understand leader-follower relationship
dynamics in a more holistic manner (e.g. Avolio et al., 2009; Sy, 2010; van Gils et
al., 2010). This is because they act as knowledge structures that specify and organise
information about the traits and behaviours of followers (Sy, 2010). Once an
individual is attributed with a follower status, the observer’s perceptions and
judgements are filtered through their IFT which, in turn, influences their behaviours
towards that individual (Lord & Maher, 1993; Poole et al., 1989; Sy, 2010).
While empirical research on IFTs has only recently begun to emerge (e.g. Sy,
2010; Whiteley et al., 2012), similar concepts have been discussed over the decades
(e.g. see Eden, 1990; Goodwin, Wofford, & Boyd, 2000; McGregor, 1960; Wofford
& Goodwin, 1994). For example, McGregor (1960) delineated two types of
managers – Theory X and Theory Y. Theory X managers are seen to hold more
negative beliefs about the nature of employees. For these managers, employees are
inherently lazy, unmotivated, will try to avoid work and are motivated for selfish
reasons (e.g. money). In contrast, Theory Y managers tend to hold more positive
beliefs in that they see employees as generally self-motivated and, provided the
126
opportunity, will seek to do well in their tasks. Actions of the manager become
aligned with these beliefs (e.g. see Finman, 1973; Neuliep, 1987; 1996). For
example, Theory X managers may introduce monetary rewards based upon work
output and/or introduce organisational systems and processes, which closely monitor
employee behaviour (McGregor, 1960). Theory Y managers though would tend to
promote more democratic and trusting work environments.
It was not until recently though that an IFT measure was developed (see Sy,
2010). Six core IFT factors: (1) industry; (2) good citizen; (3) enthusiasm; (4)
incompetence; (5) conformity; (6) insubordination (Sy, 2010) were found to make up
most leaders’ IFTs. The first three factors were found to load on a second-order
factor which Sy (2010) labelled followership prototype. The final three loaded on to
a followership anti-prototype second-order factor (Sy, 2010). While the average
ideal IFT appears to contain higher levels of the first three factors (Sy, 2010), it is
important to recognise that, at the individual level, some leaders endorse more the
anti-prototypical traits (e.g. Carsten et al., 2010; Henry, 1997; Sy, 2010). Indeed, the
quantitative results of Sy (2010) indicate there are some leaders who prefer more
passive and conformist follower styles. This is also supported by qualitative studies
(e.g. Carsten et al., 2010; Henry, 1997). For example, Henry (1997) found that some
leaders see the role of followers as simply to carry out leaders’ instructions without
question – a more conformist or compliant expectation. Carsten and colleagues
(2010) also found that some leaders construct followers as passive and obedient. In
contrast, other leaders constructed the role of a follower as taking a more proactive
role. These findings lend themselves to the question of why do leaders differentially
construct the role of a follower?
127
This is an important consideration as IFTs have implications for leadership
and organisational outcomes (Sy, 2010; Whiteley et al., 2012). For example, leaders
who hold positive IFTs (i.e. industry, enthusiasm and good citizen) tend to have
followers who report higher levels of well-being, an increased liking for the leader,
better relationships with the leader and increased performance expectations (Sy,
2010; Whiteley et al., 2012). Sy (2010) also found that leaders’ followership
prototype was positively related to their followers’ job satisfaction and trust in the
leader. In contrast, leaders who held more anti-prototypical traits (i.e. incompetence,
conformity and insubordination) were more likely to have followers report the
opposite. Looking at the theoretical antecedents of leaders’ IFTs, therefore, becomes
important.
However, given that IFTs have only recently moved onto the research
agenda, it should not be surprising that a scarcity of studies exists on their
antecedents (c.f. Derler & Weibler, 2014). The available limited research indicates
that female leaders show higher preferences for good citizen traits and behaviours in
followers compared to male leaders’ preferences (Derler & Weibler, 2014). The
same research also found leaders’ perceptions of market conditions (i.e. dynamic vs.
competitive) and internal-coordination mechanisms (e.g. degree of centralisation and
formalisation) were related to preferences for specific IFTs. Kruse and Sy (2011)
also found that increasing positive affect resulted in a corresponding increase in the
expression of prototypical followership theories. To the author’s knowledge, though,
no research has investigated whether leaders’ personality characteristics relate to
their IFTs. This includes processing styles (Norris & Epstein, 2011) and constructive
thinking (Epstein, 2001). It is logical to propose, however, that if followers’
processing styles and constructive thinking are related to their ideal ILTs (see the
128
previous chapter) then these characteristics in leaders may also be related their ideal
IFTs. Specifically, leaders should develop preferences for IFTs that fit or are
congruent with, their information-processing style and constructive thinking.
Therefore, a brief overview of person-environment theory is now presented.
Person-environment fit
As noted in the previous chapter, person-environment fit (P-E fit) is typically
assumed to occur when there is a compatibility, or congruence, between an
individual and the work environment (Kristof-Brown et al., 2005). Although there
have been broad ways of conceptualising and measuring fit (see Caplan, 1987; Judge
& Ferris, 1992; 1993; Kristof-Brown et al., 2005), the needs-supply definition
(Caplan, 1983; 1987; Kristof, 1996) is most relevant here and refers to where a
follower meets or addresses a leader’s needs, values, and preferences.
Theory Development
Similar to the theory presented in the previous chapter, it is assumed that
individuals experience positive vibes (e.g. wellbeing, calmness and positive
anticipation) and emotions (e.g. happiness) when they think, behave and
communicate in ways which are congruent with their information processing style
and constructive thinking tendencies (see Berne, 1961; Epstein, 2003). When the
environment requires or pressures an individual to deliberate, act or converse in
ways that are inconsistent or incongruent with their preferred style, they will
experience negative vibes (e.g. agitation, irritation, tension, frustration, disquietude,
129
queasiness, edginess, and apprehension) and emotions (e.g., fear, anxiety, worry; see
Epstein, 2003; McCann & Higgins, 1988).
The nature (e.g. form, content, emotiveness) and frequency of
communication between leaders and followers is assumed to be more underpinned,
or influenced, by leaders’ own personality due to their positional power (e.g. see
Scholl, Ellemers, Sassenberg, & Scheepers, 2015; Tost et al., 2013; Watson, 1982).
In dyadic communication, those with the same personality traits tend to share a
variety of common reference points in how they perceive, understand, and act in
reaction to the same situation (Schoon, 2008; Yeakley, 1982). Consequently, the
dyadic partners should be able to interact and communicate more effectively than
those who do not share the same personality traits.
Given the dyadic nature of the leader-follower relationship, leaders and
followers who share the same information processing style (i.e. congruent) should be
able to interact and communicate more successfully and with less effort than those
who do not. More congruent leader-follower interactions should result in a tendency
for the leader to experience positive vibes and emotions as they are not pressured or
required to communicate outside of their preferred style (see Berne, 1961; Epstein,
2003). In an incongruent leader-follower interaction, though, leaders theoretically
would tend to show increased attention towards those followers who do not share the
same information-processing style (e.g. see Bargh, 1982). This may be due to more
time being required for leaders’ communication to be understood, be acted upon, or
even agreed to, by the follower due to the differences between their cognitive frames
of reference (e.g. Bradley & Herbert, 1997; Higgins, King, & Mavin, 1982; McCann
& Higgins, 1988; see also descriptions of personality type interactions by Kroeger et
130
al., 2002). From the perspective of the current research, such outcomes would result
in leaders experiencing negative vibes and emotions.
The experience by leaders of positive/negative vibes and emotions during
congruent/incongruent leader-follower interactions provide the learning opportunity
to associate particular follower traits and behaviours with their respective positive or
negative affective experience. Consequently, over time, leaders should develop
preferences (i.e. ideal IFTs) and dislikes for particular follower traits and behaviours,
which correspond to these positive and negative experiences. Drawing upon these
lines of reasoning, several hypotheses regarding the relationships between leaders’
information processing style and constructive thinking with their ideal IFT can be
made.
Hypothesis development: Information Processing Style and IFTs
Those high in rational processing are deep thinkers, tend to develop clear,
rational and explainable reasons for decisions, and enjoy thinking in abstract terms
(Pacini & Epstein, 1999a). High rational processors are also more amenable to
actuarial and objective material than high experiential processors (Rosenthal &
Epstein, 2000 as cited in Epstein, 2003). They also tend to be more problem-solution
focused (Epstein et al., 1996). The incompetence dimension of IFTs contains the
items of inexperienced and uneducated (Sy, 2010). These items appear congruent,
albeit in a negative manner, with the cognitive and behavioural tendencies of high
rational processors. Specifically, high rational processors may have come to prefer
followers who are trained and proficient in their job role and are generally more
educated. Such follower traits may facilitate more informed and intelligent leader-
follower interactions that are congruent with leaders who have more rational
131
processing styles. Consequently, this should result in leaders experiencing positive
vibes and emotions. In contrast, leaders with a low rational processing style may be
more cognitively motivated to develop preferences for more uneducated and
inexperienced followers. This is because they are less able to question leaders’ ideas
and directions due to lower job-related knowledge and general ability. In support,
inexperienced followers have been found to be more susceptible to leader influence
than experienced followers suggesting that inexperienced followers are less likely to
question the leader (Chong & Wolf, 2009). Thus, it is hypothesised:
Hypothesis 1: a negative relationship between leaders’ rational thinking and
preference for incompetence in followers will exist…
Individuals high on experiential processing show higher levels of empathy
and are better at forming and maintaining positive social relationships (Norris &
Epstein, 2011; Pacini & Epstein, 1999a). High-quality working relationships depend
on a high level of trust between leaders and followers (Schriesheim, Castro, &
Cogliser, 1999). Those high on experiential processing, therefore, may develop
preferences for followers who show traits conducive to the building of strong,
trusting and successful working relationships and friendships at work. Such
outcomes appear to be associated with the IFT dimension of good citizen, which
contains the items of loyal, reliable and team player (Sy, 2010). For example,
synonyms of loyal and reliable include being trustworthy and dependable (Loyal,
n.d; Reliable, n.d.). In addition, being a team player is also described in terms of
dependability and building effective working relationships with others to achieve
some goal (Driskell, Goodwin, Salas, & O’Shea, 2006). Thus, it is hypothesised:
132
Hypothesis 2: a positive relationship between leaders’ experiential processing and
their preference for good citizen will emerge…
Insubordinate followers have traits such as being bad-tempered, rude and
arrogant (Sy, 2010). Such traits and behaviours can be described as socially
inappropriate as they break generally accepted social norms of behaviour (DeBono,
Shmueli & Muraven, 2011). Breaking of social norms can increase the chance of
conflict (Pruitt, 1998). Therefore, insubordination in followers may increase the
chance for leader-follower conflict. Conflict has the potential to elicit negative affect
(e.g. Fulk & Wendler, 1982). Affect is a core element of the experiential system
(Epstein, 2003; 2014). Thus, high experiential processors, especially those high on
emotionality, may develop a particular dislike for insubordinate followers as they
would be more likely to elicit negative vibes and emotions in the leader. In contrast,
leader-follower conflict is less likely when the latter engages in more respectful
behaviours. Consequently, they are less likely to elicit negative affect in the leader.
Therefore, it is hypothesised that:
Hypothesis 3: a negative relationship between leaders’ experiential processing and
preference for insubordination in followers will exist.
Hypothesis development: Constructive Thinking and IFTs
Positive organisational outcomes require followers to be active in decision-
making processes and to positively support the leader to affect change and achieve
the desired goals (Chaleff, 1995; 2003; 2009; Kelly, 1988; 1992). For example,
Kelly (1988; 1992) emphasised that the most effective followers are those who
133
demonstrate the courage to voice antithetical positions to leaders if needed (e.g. if
there are issues with leaders’ actions) but in a respectful and positive way (Kelly,
1988; 1992). It is clear though that not all leaders perceive such feedback in the same
way (Grant et al., 2009). Threats to self-esteem, of which someone who voices a
differing opinion to oneself is a key source, can evoke anxiety in the receiver
(Epstein, 2001; Eysenck, MacLeod, & Mathews, 1987; Hodges, 1968). Followers
who, therefore, speak up and question the status quo and decisions of a leader have
the potential to cause conflict, uncertainty and elicit negative emotions in a leader
(e.g. Argyris & Schön, 1978). For example, leaders may see followers questioning
their decisions as a threat to their power, status and competence (e.g. Grant et al.,
2009) and, therefore, experience increased anxiety (Carver et al., 1985; Meyer &
Starke, 1982; Sachs, 1982; Swann & Read, 1981). This can even motivate some
leaders to avoid or dismiss the criticism or feedback and even attack the messenger’s
credibility (Ilgen et al., 1979; Morrison & Milliken, 2000; Kluger & DeNisi, 1996).
Emotional coping is about how people deal with frustration, disappointment,
rejection, criticism and disapproval (Epstein, 2001). Leaders with good emotional
coping, therefore, may be better able to deal (i.e. cognitively and emotionally) with
those followers who speak up and question them. For example, a leader with good
emotional coping may be less likely to consider a follower who voices an antithetical
position as a threat to self-esteem. Consequently, leaders with poor emotional coping
may develop preferences for followers who simply accept and follow their decisions
and directions without question as they are less likely to create cognitively and
emotionally threating situations. The conformity IFT factor, which contains the items
of follows trends, easily influenced and soft-spoken, appears to lend itself to more
134
passive behaviours in followers which would be less likely to create such threat
situations. Therefore, it is hypothesised that:
Hypothesis 4: a negative relationship between leaders’ emotional coping and their
preference for conformity in followers will emerge.
Leaders who are high in behavioural coping are more likely to have
automatic thoughts to events that facilitate effective action such as engaging in hard
work, effort, and planning (Epstein, 2001). They also tend to focus more on
achieving outcomes rather than get caught up worrying about things (e.g. deadlines;
Epstein, 2001). The industry, enthusiasm and conformity IFT factors appear to be
congruent with such tendencies in leaders with high behavioural coping, albeit for
different reasons. The industry factor contains the items of productive, hardworking
and goes above and beyond (Sy, 2010). Such behaviours in followers may be
perceived by leaders with high behavioural coping as positive as they are interpreted
to be supportive of the leader and team/organisational goals. Passive followers, in
contrast, may be more likely to frustrate and even anger (non-satisfying need states)
the leader (e.g. see Maier & Verser, 1982; Watson & Michaelsen, 1984). Thus, it is
hypothesised:
Hypothesis 5: a positive relationship between leaders’ behavioural coping and their
preference for industry in followers will emerge…
Good behavioural copers are also more confident in their abilities to be
successful when taking on challenges and tend not to be pejorative, but rather focus
135
on the effectiveness of particular behaviours in others (Epstein, 2001). In the context
of the work-environment, the effectiveness of followers’ behaviours, therefore, may
become emphasised for leaders with high levels of behavioural coping. The traits
and behaviours of conformist followers (i.e. follows trends, easily influenced, and
soft-spoken) do not appear to lend themselves to proactivity as passivity and/or
conformity can undermine organisational success (Chaleff, 1995; 2008; Crant, 2000;
Kelley, 1988; Sy, 2010). For example, Crant (2000) states that effective followership
requires followers to occasionally challenge the status quo through actively adapting
to the current situational demands they find themselves in order to bring about
improved outcomes. Given the focus on overcoming problems and being action-
oriented for leaders with high levels of behavioural coping, their experiences of more
passive and conformist followers, therefore, may particularly frustrate, or in some
cases even anger them. This is because they may see follower passivity as
contributing to undermining goal achievement. Consequently, they would have
developed preferences for non-conformist followers. Thus, it is hypothesised:
Hypothesis 6: a negative relationship between leaders’ behavioural coping and
their preference for conformity in followers will emerge.
Categorical thinkers tend to believe that others are in error for simply taking
different positions to themselves (Epstein, 2001). They are also quick to experience
annoyance and anger when their expectations are violated (Epstein, 2001). High
categorical thinkers may develop preferences for conformist followers as such
individuals would be more inclined to simply follow leaders’ directions and not
question them (i.e. items of easily influenced and follow trends; Sy, 2010). This is
136
because non-conformist followers may be more likely to query leaders’ ideas and
take a different perspective, which consequently could be perceived as a potential
threat to leaders’ abilities and their self-esteem (e.g. see Argyris & Schön, 1978;
Carver et al., 1985; Fuller & Marler, 2009; Grant et al., 2009; Meyer & Starke, 1982;
Sachs, 1982; Swann & Read, 1981). In contrast, conformist followers would be less
likely to trigger this threat reaction in leaders as they may be less likely to voice
different perspectives. In support, employees with a low proactive personality (which
contains some traits similar to those in the conformity factor) tend to engage in less
voice behaviour and taking charge (Fuller & Marler, 2009). Therefore, it is
hypothesised that:
Hypothesis 7: a positive relationship between leaders’ categorical thinking and
their preference for conformity in followers will emerge.
Naïve optimists think in simplistic and stereotyped ways and are
unrealistically optimistic about outcomes (Epstein, 2001). Consequently, they may
conceptualise an ideal follower as one that is culturally stereotypical. The cultural
stereotype that appears most associated with followers is that of a hard-worker who
is productive and enthusiastic (Berg, 1998; Sy, 2010). The former two traits appear
highly similar to the IFT factor of industry (i.e. productive, hardworking, goes above
and beyond) while the latter enthusiasm (i.e. excited, outgoing, happy; Sy, 2010).
Therefore, it is hypothesised that:
Hypothesis 8: a positive relationship between leaders’ naïve-optimism and their
preference for industry in followers will emerge…
137
Hypothesis 9: a positive relationship between leaders’ naïve-optimism and their
preference for enthusiasm in followers will emerge.
Followers though are also stereotypically defined in more negative terms
such as deferent, passive and obedient (Berg, 1998; Carsten et al., 2010). Therefore,
leaders who are naïve optimists may also seek more conformist followers. Therefore,
it is also hypothesised that:
Hypothesis 10: a positive relationship between leaders’ naïve-optimism and their
preference for conformity in followers will emerge.
Method
Participants
A total of 133 managers participated. Participants were employed in a wide
variety of organisations and industries. There were 77 females (57.9%) and 56 males
(42.1%). The average age was 44.27 years (SD = 11.06). The average length of time
an individual had been employed in their organisation was 100.14 months (SD =
102.43). Differences between males (M = 43.91, SD = 10.12) and females (M =
44.53, SD = 11.70) on age was not significant as indicated by an independent
samples t-test, t(131) = .319, p = .750. There was also no difference between males
(M = 100.63, SD = 84.17) and females (M = 99.79, SD = 114.43) on time in current
organisation, t(131) = .046, p= .963.
138
Measures
Implicit Followership Theory Questionnaire (Sy, 2010) – adapted. The
Implicit Followership Theory Questionnaire (see Appendix C; Sy, 2010) is designed
to measure an individual’s IFT. In the original version, individuals are asked to rate
on a 7-point Likert scale the degree to which 18 traits are characteristic (1 = Not at
all Characteristic; 7 = Extremely Characteristic) of “a follower in a business setting”.
Reliability ranges between .75 (acceptable) to .91 (excellent) depending on the factor
(George & Mallery, 2003; Sy, 2010). Similar to the rationale presented in the
previous chapter, though, the target of focus (e.g. follower, ideal follower) can
influence ratings (e.g. see Offerman et al., 1994). Therefore, the measure was
adapted to target “an ideal follower in a business setting” as it was theorised that
followers who match leaders’ ideal IFTs would be more theoretically relevant targets
for the purpose of the current study. Similar to the ILT measure in the preceding
chapter, a 10-point Likert Scale was also adopted to help overcome potential
ceiling/floor effects for this change (i.e. increase ability for more nuanced responses
at top and bottom of the scale).
Rational-Experiential Multimodel Inventory (REIm; Norris & Epstein,
2011). The REIm (Norris & Epstein, 2011) is a 42 item measure which asks
individuals to respond on a 5-point Likert scale (1 = Definitely False; 5 = Definitely
True) the degree to which they use a rational (12 items) or experiential processing
style (30 items). The experiential scale consists of 3 sub-scales (10 items each) to
measure its different aspects: (1) intuitive; (2) emotionality and; (3) imagination. The
intuitive facet is designed to assess the ability and engagement of an individual’s use
of intuitive judgements (e.g. “I often go by my instincts when deciding on a course of
139
action”.) The emotionality sub-scale assesses the intensity, frequency, duration, and
favourable attitude towards strong affect (e.g. “My emotions don’t make much
difference in my life”; reversed scored). The imagination sub-scales assesses the
engagement in, and appreciation of, imagination, aesthetic productions, and imagery
(e.g. “Sometimes I like to just sit back and watch things happen.”) Higher scores
indicate a higher preference for processing on that particular dimension. All main
scales in the current study showed acceptable to excellent levels of reliability (range
.70 to .88; Kline, 2000).
Constructive Thinking Inventory (CTI; Epstein, 2001). The constructive
and destructive components of the experiential system were assessed by the 108-item
constructive thinking inventory (CTI, Epstein, 2001). The constructive components
of the experiential system include global constructive thinking, emotional coping,
and behavioural coping. The destructive components are personal superstitious
thinking, categorical thinking, esoteric thinking, and naïve-optimism (Epstein, 2001).
The reader is referred to Table 1.3 presented in Chapter 1 for main-scale and sub-
scale definitions. Participants respond on a 5-point Likert scale based on how much a
statement is true of them (1 = Definitely False; 5 = Definitely True). Raw scores are
entered into an electronic scoring programme (Epstein & PAR Staff, 2008) and
converted to T-Scores based upon gender norms.
Procedure
The same website that was developed to support and promote the research
described in Chapter 3 was used as both research studies were conducted at the same
140
time. Participants were also recruited via the same means. Please refer to Chapter 3’s
procedure section for more information.
The two-stage procedure described in the previous chapter was used in the
current study to control for some of the effects of common-method bias (see
Podsakoff et al., 2003). In stage one, participants completed the REIm (Norris &
Epstein, 2011), the adapted Implicit Followership Theory Questionnaire (Sy, 2010)
and other demographic information. Those who completed stage one were invited
two days later to complete stage two via email. In stage two, participants completed
the CTI (Epstein, 2001) only. The average time between the completion of stage 1
and stage 2 was 5.05 days (SD = 3.46, range 2 - 22). Participants could choose to
receive written feedback based upon their responses to the CTI (Epstein, 2001).
Results
Common method bias
To assess common method bias, a principal components analysis was used
with direct oblimin rotation (see Podsakoff et al., 2003). A total of 43 components
with Eigenvalues greater than 1 were identified. These accounted for 89.47% of the
total variance. The first factor did not account for the majority of the variance
(14.12%). Harman’s single-factor analysis indicated that a single factor did not
account for the majority of the variance (14.12%). These results suggest common
method was not a significant cause for concern.
141
Exploratory Factor Analysis – IFT Measure
As the current measure used an adapted version of the IFT measure and no
prior assumptions or hypotheses were made regarding what factors would make up
“ideal” followers (see Finch & West, 1997), an exploratory factor analysis (EFA)
rather than a confirmatory factor analysis (CFA), was used to investigate the
underlying factor structure of the ILT items. There were 7.38 subjects per item,
which is higher than the general minimum rule of 5 subjects-to-variables ratio (see
Bentler & Chou, 1987; Bryant & Yarnold, 1995). Hatcher (1994) also specifies that
a minimum number of 100 individuals as desirable for factor analysis. All items
correlated at least .3 or above with at least one other item. The Kaiser-Meyer-Olkin
measure of sampling adequacy was .689, above the recommended value of .6, and
Bartlett’s test of sphericity was highly significant, χ2 (153) = 816.34, p < .001. The
diagonals of the anti-image correlation matrix were also all over .5, supporting the
inclusion of each item in the factor analysis. Given the sample size and overall
indicators, factor analysis was conducted with all 18 items.
As the primary purpose was to identify the key IFT factors and compute
composite scores for each, principal components analysis was used (see Abdi &
Williams, 2010). As the final factors were expected to be correlated (see Sy, 2010),
the oblimin rotation was used to obtain a simple and interpretable factor solution
(Yaremko, Harari, Harrison, & Lynn, 1986). A total of 4 items were eliminated
through an iterative process as they did not contribute to either a theoretically
interpretable simple factor structure and, with the exception of one item
(uneducated), failed to meet a minimum criterion of having a primary factor loading
of .4 or above, and no cross-loading of .3 or above. The uneducated item was
retained despite cross loading onto a conformity component due to a much higher
142
primary loading on an incompetence component which was theoretically
interpretable (see Fabrigar, Wegener, MacCallum, & Strahan, 1999).
A principal-components factor analysis of the remaining 14 items indicated
five theoretically interpretable factors, which explained 68.75% of the variance. The
original industry and good citizen scales found by Sy (2010) did not emerge as
separate factors in the EFA. Instead, four of the original six items for these scales
loaded onto a new component, which was called committed due to the nature of the
loading items. These final five factors, therefore, were insubordination, committed,
incompetence, conformity and enthusiasm. All items had primary loadings over .6
and, except for the uneducated item, no items had cross-loadings above .3. The
factor loading matrix for this final solution is presented in Table 4.1. Descriptive
statistics by overall group and gender for the new IFT factors and inter-correlations
are shown in Table 4.2. Reliability of new IFT scales is also shown on the diagonals.
As can be seen, there were some reliability issues with the enthusiasm component
(<.6; George & Mallery, 2003; Kline, 2000). However, it did meet the minimum
level of .5 suggested for research purposes (Kline, 2000). The new scale was also
slightly more reliable than the original three items (.57) specified by Sy (2010).
Furthermore, Cronbach’s alpha provides the lower-bound estimate of reliability
(Tavakol & Dennick, 2011) Therefore, the enthusiasm component was retained to
explore the hypothesised relationships. However, it is important to recognise that
authors recommend at least .7 as significant random error exists when reliability is
low, which may result in interpretation issues (e.g. see Bland & Altman, 1997;
Tavakol & Dennick, 2011). Consequently, this scale should be interpreted with some
caution due to this low reliability. As the industry and good citizen factors did not
143
Table 4.1.
Direct Oblimin Rotated Pattern Matrix of the 14-Item Adapted IFT Measure (n = 133) Items Insubordination Committed Incompetence Conformity Enthusiasm
Bad-Temper .869 -.022 -.012 .078 -.103
Rude .793 -.031 -.107 -.046 .090
Arrogant .760 -.015 .037 -.102 .032
Goes above and beyond .021 .888 -.017 -.063 .086
Productive -.164 .842 -.011 .040 .104
Reliable -.042 .676 .037 .072 -.027
Loyal .117 .630 .021 -.109 -.233
Slow .142 .027 -.858 .236 .135
Uneducated -.136 -.122 -.700 -.331 -.050
Inexperienced .088 .013 -.610 -.104 -.242
Follows trends -.010 .011 .045 -.867 -.072
Easily influenced .116 .043 -.071 -.855 .099
Outgoing .103 -.072 .079 -.062 -.822
Excited -.141 .137 -.149 .092 -.802
Variance Explained 24.52% 18.42% 9.50% 8.81% 7.50%
Note: Factor loadings above ±.300 are bolded to aid readability.
144
Table 4.2.
Means, Scale Reliabilities, and Inter-correlations of Scales on 18 Item Adapted IFT Measure IFT Factor Means (SD) 1 2 3 4 5
Overall
(n = 133)
Male
(n = 56)
Female
(n = 77)
1. Insubordination 1.33 (.55) 1.32 (.47) 1.34 (.61) (.76) -.27** .29** .31** .03
2. Committed 8.67 (1.00) 8.51 (1.07) 8.78 (.94) (.76) -.10 .01 .27**
3. Incompetence 2.71 (1.24) 2.69 (1.20) 2.73 (1.27) (.61) .33** .24**
4. Conformity 4.16 (2.12) 3.87 (1.86) 4.37 (2.29) (.73) .22*
5. Enthusiasm 6.48 (1.87) 6.91 (1.59) 6.16 (2.00) (.58)
Notes: *Indicates p < .05; **Indicates p< .01; Cronbach’s alpha reliabilities are shown on the diagonal in brackets. All inter
correlations and reliability coefficients as based on overall group data.
145
separately emerge in the EFA, hypotheses 2, 2a, 7,7a, 8, 10 and 10a could not be
directly tested.
Data screening & assumption checking
Any data point that was more than 1.5 times the Interquartile range (IQR)
above or below the first or above the third quartile was classified as an outlier (see
Tukey, 1977). Due to the small sample size, a decision was made to truncate to
outliers to the nearest highest/lowest non-outlier score (see Osborne & Overbay,
2004). Thus the relative order of scores is maintained but distribution issues are
reduced and statistical power is retained. A total of 35 outliers across all scales were
recoded with a mean of 1.09 items per factor. Inspection of the histograms and Q-Q
plots indicated some minor normality issues but the inferential statistics used are
generally quite robust against this (Tabachnick & Fidell, 2007). Insubordination
though showed a strong positive skew, which was not improved by an inverse log
transformation (see Osborne, 2002). Therefore, hypotheses involving
insubordination were assessed by non-parametric procedures where applicable
(Allen & Bennett, 2010). Inspection of bivariate scatter plots for remaining IFT
factors indicated the assumptions of both linearity and homoscedasticity had been
met.
Descriptive Statistics, Scale Reliabilities and Scale Inter-correlations
Descriptive statistics for the REIm at both the overall group and by gender
are shown in Table 4.3. An independent samples t-test was used to compare mean
rational and experiential processing style between males (n = 56) and females (n =
146
77). Levene’s test was significant for rational processing (p = .005) indicating equal
variance assumption had been violated and thus an adapted t-test was used. The
Levene’s test for all other factors did not reach significance. There were no
significant differences between males and females on rational processing, t(101.049)
= 1.370, p = .174, experiential processing, t(131) = .696, p = .488, intuition, t(131) =
1.250, p = .214, and imagination, t(131) = .069, p = .945. Females though, showed
significantly higher emotionality than males, t(131) = 2.101, p = .038.
Table 4.3.
Descriptive statistics on REIm dimensions for overall group and as a factor of
gender REIm Factor Means (SD)
Overall
(n = 133)
Male
(n = 56)
Female
(n = 77)
1. Rational 3.96 (.57) 4.04 (.64) 3.90 (.51)
2. Experiential 3.51 (.35) 3.48 (.32) 3.53 (.38)
3. Intuition 3.61 (.49) 3.67 (.47) 3.56 (.49)
4. Emotionality 3.29 (.54) 3.18 (.49) 3.38 (.56)
5. Imagination 3.62 (.60) 3.62 (.47) 3.62 (.68)
An independent samples t-test was used to compare the means of females (n
= 77) and males (n = 56) on each IFT factor. Levene’s test was significant for
incompetence (p = .035) and conformity (p = .034) indicating equal variance
assumption had been violated. Thus adapted t-tests were used for these. No
significant differences were observed on committed, t(131) = .212, p = .832,
incompetence, adapted t(122.443) = .190, p = .850, or conformity, adapted t(129.24)
= 1.400, p = .164. However, males showed higher levels on enthusiasm than females,
t(131) = 2.315, p = .022. No significant difference was found on insubordination
between males (Mean Rank = 69.68) and females (Mean Rank = 65.05) as indicated
by a Mann-Whitney U-test, U = 2006.00, z = .783, p = .434.
147
Hypothesis Testing
As described in chapter 3, while SEM would provide a more powerful way of
assessing the hypotheses, it is estimated that 14 parameters would need to be
specified in the model, which would mean that the recommended sample size would
be 350 (see Nachtigall et al., 2003). As the current sample size does not meet this
sample size requirement, a decision was made to use correlations and Hierarchical
Multiple Regression for hypothesis testing.
Bivariate Pearson’s product-movement correlation coefficients were
calculated between the REIm and CTI with the four normally distributed IFT
dimensions: committed, incompetence, conformity and enthusiasm. Due to the non-
normal data, Spearman’s rho was used to assess relationships involving the
insubordination IFT dimension. Fifteen individuals did not complete the second
stage and, therefore, analyses using the CTI are based on a sample of 118 leaders.
Table 4.4 presents the correlational results.
REIm and IFT hypothesised relationships
Hypothesis 1 was supported as indicated by a significant negative
relationship between rational processing and incompetence. Hypothesis 3 was not
supported as demonstrated by the non-significant negative relationship between
experiential processing and insubordination.
148
Table 4.4.
Correlations between REIm (n=133), CTI (n=118) and each IFT Dimension
IFT Dimensions
Insubordination˄ Committed Incompetence Conformity Enthusiasm
Demographics
Gender -.07 .13 .02 .12 -.20*
Age .03 .17* .03 -.20* -.03
REIm
Rational .10 .03 -.32** -.24** -.15
Experiential -.07 .08 .20* .21* -.02
CTI
Global Constructive
Thinking -.07 .06 -.15 -.39** .16
Emotional Coping -.06 .01 -.12 -.31** .19*
Behavioural Coping -.13 .15 -.21* -.44** .01
Personal
Superstitious
Thinking
-.04 .10 .17 .26** .07
Categorical Thinking -.02 .11 .32** .35** .09
Esoteric Thinking .13 -.02 .37** .20* .13
Naïve-Optimism -.08 .24** .13 .14 .26**
Notes: *Indicates p<.05; **Indicates p<.01; ˄Relationships indicated by Spearman’s rho. Results
reported for Enthusiasm should be interpreted with caution due to low reliability.
CTI and IFT Hypothesised Relationships
Hypothesis 4 was supported as indicated by a significant negative
relationship between emotional coping and preference for followers showing
conformity. Hypothesis 6 was supported as indicated by the negative relationship
between behavioural coping and conformity. Hypothesis 7 was supported as shown
by the positive relationship between categorical thinking and conformity. Support for
hypothesis 9 was provided by the significant positive relationship between naïve-
optimism and enthusiasm. Hypothesis 10 was not supported as demonstrated by the
non-significant positive relationship between naïve-optimism and conformity at the
group level.
149
Non-hypothesised REIm and CTI main scale relationships with IFT dimensions
Several non-hypothesised REIm main-scale with IFT relationships emerged.
First, a significant positive correlation between experiential processing and
incompetence emerged. Second, there was a significant negative relationship
between rational processing and conformity. Finally, there was a significant positive
relationship between experiential processing and conformity. Several non-
hypothesised relationships between the CTI main-scale and IFT dimensions also
emerged. First, a significant negative correlation emerged between global
constructive thinking and conformity. Second, a significant positive correlation was
observed between emotional coping and enthusiasm at the group level. Third, a
significant negative correlation emerged between behavioural coping and
incompetence. Fourth, there was a significant positive correlation between
categorical thinking with incompetence. Fifth, esoteric thinking showed a significant
positive correlation with both incompetence and conformity at the group level.
Finally, a significant positive correlation between naïve-optimism and committed
emerged.
Hierarchical Regression
To further examine the relationships between constructive thinking and
information processing with ILT dimensions, hierarchical multiple regression
analyses were conducted. Bivariate scatter plots were obtained to assess and satisfy
linearity and homoscedasticity for all regressions. The variables were all
demonstrated to be linearly related. Residual scatter plots also demonstrated that the
assumption of homoscedasticity had been satisfied. The correlations of the variables
150
were assessed for multicollinearity but none was detected as all correlations were
below 0.9 (Tabachnick & Fidell, 2007). All analyses are based upon n = 118. A
dummy variable was used to assess the effects of gender in the model1. As no
significant relationships between any of the independent variables and
insubordination emerged, no regression model was calculated.
Information Processing/Constructive Thinking and IFT Trait of Committed
With Committed as the dependent variable, a two-step hierarchical multiple
regression was conducted. Age and Gender were entered in Step 1. Naïve Optimism
was added in Step 2 as, it emerged as a significant predictor at the correlational
stage. There is also a theoretical rationale for its inclusion. Specifically, naïve
optimists tend to endorse stereotyped ways of thinking (Epstein, 2001).
Consequently, they may conceptualise an ideal follower as one that is culturally
stereotypical; as one who is hard working and committed (e.g. see Engle & Lord,
1997; Sy, 2010) amongst other traits. Committed followers, therefore, may be seen
to engage in interactions and create work-environments, which are congruent with
the processing tendencies of naively optimistic leaders. Regression statistics are
shown in Table 4.5.
1Where male = 0 and female = 1.
151
Table 4.5.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Committed Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .246 .060 .060*
Gender .156 .077 .851 .006
Age .021 .233 2.582* .054
Step 2 .328 .108 .047*
Gender .183 .090 1.020 .008
Age .019 .207 2.318* .042
Naïve Optimism .022 .219 2.455* .047
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
The overall regression model at Step 1 was significant, F(2,115) = 3.691, p =
.028. Age alone significantly contributed 5.4% to the regression model at this step.
The introduction of Naïve Optimism in Step 2 explained an additional and
significant 4.7% in Dedication preference variation, F(3,114) = 4.578, p = .005. Age
at this step also remained a significant contributor. Overall, the three independent
variables accounted for 10.8% of the variance in Committed.
Information Processing/Constructive Thinking and IFT Trait of Incompetence
With Incompetence as the dependent variable, a two-step hierarchical
multiple regression was conducted. Age and Gender were entered in Step 1. Rational
processing was added in Step 2 as it is both theoretically related to incompetence as
well as emerging as having a significant relationship in the correlational analyses. In
Step 3, Categorical Thinking, Esoteric Thinking, Behavioural Coping and
Experiential Processing were entered to explore whether these added any additional
explanation to the variance in Incompetence. Regression statistics are shown in
Table 4.6.
152
Table 4.6.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Incompetence Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .103 .011 .011
Gender .146 .057 .617 .003
Age .010 .086 .927 .007
Step 2 .364 .132 .122**
Gender .056 .022 .250 .000
Age .003 .027 .308 .001
Rational Processing -.787 -.356 -4.000** .122
Step 3 .483 .233 .100**
Gender -.134 -.052 -.593 .002
Age .002 .016 .180 .000
Rational Processing -.486 -.220 -2.133* .032
Experiential
Processing .347 .099 1.106 .009
Behavioural Coping .010 .059 .566 .002
Categorical Thinking .029 .192 1.985^ .027
Esoteric Thinking .036 .229 2.390* .040
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
The overall regression model at Step 1 was non-significant, F(2,115) = .619,
p = .540. The introduction of Rational Processing in Step 2 explained an additional
and significant 12.2% in Incompetence preference, F(3,114) = 5.800, p = .001. In
Step 3, the addition of Categorical Thinking, Esoteric Thinking, Behavioural Coping
and Experiential Processing explained an an additional and significant 12.2% in
Incompetence preference, F(7,110) = 4.770, p < .001. However, the only significant
predictors in the final model were Rational Processing and Esoteric Thinking, which
independently explained 3.2% and 4.0% of the variance in Incompetence. It should
be noted that categorical thinking did approach significance (p = .050). Overall, the
seven independent variables accounted for 23.3% of the variance in Incompetence.
153
Information Processing/Constructive Thinking and IFT Trait of Conformity
With Conformity as the dependent variable, a five-step hierarchical multiple
regression was conducted. Age and Gender were entered in Step 1. Emotional
coping, behavioural coping and categorical thinking were added in Step 2 as they
were all theoretically related to incompetence as well as emerging as having a
significant relationship in the correlational analyses. In Step 3, rational processing
was entered. While not specifically presented in the introduction, theoretically, it
may be that less conformist followers, who are harder to influence and are more
likely to question the status quo (e.g. see Blass, 2009; Milgram, 1965), logically are
more likely to critique leaders’ ideas and be more willing to stand up to them. Such
behaviours in followers may appeal to leaders with high rational processing styles
due to their preferences for in-depth thinking and enjoyment of taking on challenging
issues (Epstein, 1998; Norris & Epstein, 2011). Specifically, non-conformist
followers would be more likely to critique and provide feedback on leaders’ ideas,
which provides input to the rational system to process – a positive communication
experience for the rational system. In support, the correlational analysis
demonstrated a significant negative relationship between rational processing and
conformity. In Step 4, the remaining variables, which had shown a significant
correlational relationship with conformity were entered to explore whether these
added any additional explanation to the variance in Conformity. Regression statistics
are shown in Table 4.7.
154
Table 4.7.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Conformity Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .220 .048 .048^
Gender .599 .137 1.510 .019
Age -.033 -.172 -1.887 .029
Step 2 .516 .267 .218**
Gender .213 .049 .573 .002
Age -.026 -.133 -1.557 .016
Emotional Coping .018 .084 .742 .004
Behavioural Coping -.111 -.388 -3.777** .093
Categorical Thinking .066 .258 2.672** .047
Step 3 .519 .269 .003
Gender .210 .048 .562 .002
Age -.029 -.148 -1.667^ .018
Emotional Coping .017 .079 .700 .003
Behavioural Coping -.102 -.357 -3.161* .066
Categorical Thinking .062 .242 2.433* .039
Rational Processing -.245 -.065 -.657 .003
Step 4 .553 .306 .036
Gender .268 .061 .722 .003
Age -.032 -.166 -1.877^ .023
Emotional Coping .012 .057 .452 .001
Behavioural Coping -.110 -.384 -3.329** .071
Categorical Thinking .087 .340 3.125** .063
Rational Processing -.291 -.077 -.773 .004
Experiential
Thinking .846 .142 1.581 .016
Personal
Superstitious
Thinking
-.063 -.226 -1.757^ .020
Esoteric Thinking .001 .003 .034 .000
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
The overall regression model at Step 1 only approached significance,
F(2,115) = 2.925, p = .058. The introduction of Emotional Coping, Behavioural
Coping and Categorical Thinking in Step 2 explained an additional and significant
21.8% in Incompetence preference, F(5,112) = 8.143, p < .001. The key contributors
at this step were Behavioural Coping and Categorical Thinking, which independently
explained 9.3% and 4.7% in Conformity variance explained respectively. Emotional
coping did not emerge as a significant contributor in this model. In Step 3, the
155
addition of rational processing added a non-significant .3% (p = .513) of variance
explained, F(6,111) = 6.823, p < .001. In Step 4, the addition of Experiential
Processing, Personal Superstitious Thinking and Esoteric Thinking added a non-
significant 3.6% increase in Conformity variance explained. The final model
revealed only Behavioural Coping (7.3%) and Categorical Thinking (6.3%) were
significant independent contributors, although both Age (p = .063) and Personal
Superstitious Thinking (p = .082) did approach significance. Overall, the final model
30.6% of the variance in Conformity.
Information Processing/Constructive Thinking and IFT Trait of Enthusiasm
With Enthusiasm as the dependent variable, a three-step hierarchical multiple
regression was conducted. Age and Gender were entered in Step 1. Naïve Optimism
was added in Step 2. In Step 3, Emotional Coping was entered to explore whether
this added any additional variance explained. Regression statistics are shown in
Table 4.8. The overall regression model at Step 1 was significant, F(2,115) = 3.485,
p = .034. Gender was the only significant contributor though at Step 1,
independently explaining 5.6% of the variance in Enthusiasm. The introduction of
Naïve Optimism in Step 2 explained an additional and significant 5.8% in
Incompetence preference, F(3,114) = 4.931, p = .003. Gender also remained a
significant contributor at this step, independently contributing 4.9% in variance
explained. In Step 3, Emotional Coping explained a non-significant increase of 2.7%
in Enthusiasm variance explained, F(4,113) = 4.685, p = .002. In this final model,
Gender and Naïve Optimism were the only remaining significant predictors which
independently explained 4.5% and 5.3% of the variance in Enthusiasm respectively.
Overall, the final model accounted for 14.2% of the variance in Enthusiasm. It
156
should be noted here though that the results reported for Enthusiasm should be
interpreted with some caution due to low reliability.
Table 4.8.
Unstandardised (B) and Standardised (β) Regression Coefficients, and Squared Semi-Partial
Correlations (sr2) For Each Predictor Predicting Enthusiasm Trait Preference
Independent Variables B β t sr2 R R2 ∆R2
Step 1 .239 .057 .057*
Gender -.844 -.237 -2.618* .056
Age .005 .030 .337 .001
Step 2 .339 .115 .058**
Gender -.791 -.222 -2.516* .049
Age .000 .001 .011 .000
Naïve Optimism .043 .243 2.726** .058
Step 3 .377 .142 .027^
Gender -.758 -.213 -2.434* .045
Age -.007 -.047 -.510 .002
Naïve Optimism .041 .233 2.647** .053
Emotional Coping .030 .173 1.899^ .027
*Indicates p < .05; **Indicates p < .01; ˄Approaching significant p < .10
Discussion
The current study extends theory on leader IFT development. The primary
focus was to ascertain whether leaders develop preferences for traits and behaviours
in followers that are congruent with their information processing style (Norris &
Epstein, 2011) and degree of constructive thinking (Epstein, 1998; 2001). There was
general support for this proposition. The discussion presents theoretical and practical
implications of the findings.
Information Processing Style – IFT Congruence
Leaders are understood to develop preferences for traits and behaviours in
followers that have contributed to their experience of positive vibes and emotions
157
(see Epstein, 2003). In contrast, they learn to disfavour those that have historically
resulted in the experience of negative vibes and emotions. Leader-follower
interactions, therefore, are conceptualised as learning contexts. Specifically, when
followers’ traits and behaviours facilitate leaders to think, behave and communicate
in ways that are congruent (implicitly or explicitly) with their information processing
style, they experience positive affect (see Berne, 1961; McCann & Higgins, 1988).
Alternatively, they experience negative affect when the traits and behaviours of
followers require leaders’ to process information and communicate in ways not
aligned to their information processing style. In both contexts, leaders learn to
associate the specific follower traits and behaviours with these positive and negative
affective experiences.
As hypothesised, leaders with a high rational processing style tended to
prefer more competent followers (or more specifically less incompetent followers).
Incompetence includes traits such as uneducated and inexperienced. More
experienced followers and those who are educated are likely to hold more job-related
knowledge and a general cognitive ability to support increased performance and
creativity in their job role (Amabile, 1998; Quińones, Ford, & Teachout, 1995;
Tierney & Farmer, 2002; Weisberg, 1999). Such traits in followers will probably be
more congruent with the cognitive and behavioural tendencies of leaders with high
rational processing styles. For example, more experienced and educated followers
should be better at providing a leader with valuable insights and perspectives into
job/task related issues and decisions than less competent followers. Consequently,
rational leaders would develop preferences for competent followers.
Interestingly, leaders with a high rational processing style tended to prefer
less conformist followers. Whilst not specifically hypothesised, this finding is still
158
aligned with the notion of congruence. Specifically, it is likely that less conformist
followers, who are harder to influence and are more likely to question the status quo
(e.g. see Blass, 2009; Milgram, 1965), logically are more likely to critique leaders’
ideas and be more willing to stand up to them. Such behaviours in followers may
appeal to leaders with high rational processing styles due to their preferences for in-
depth thinking and enjoyment of taking on challenging issues (Epstein, 1998; Norris
& Epstein, 2011). Specifically, non-conformist followers would be more likely to
critique and provide feedback on leaders’ ideas, which provides input to the rational
system to process – a positive communication experience for the rational system.
Consequently, leaders with high rational processing styles may be inclined to
develop ideal IFTs that are congruent with the ability for followers to challenge
leaders’ positions and decisions. Thus, followers can help leaders think more deeply
about issues (e.g. provide new evidence or alternative perspectives). Consequently,
the leader-follower relationship for leaders with high rational processing is about
reinforcing their rational processing preferences. However, it is important to
recognise that rational processing provided no significant increase in explaining
conformity preference beyond Behavioural coping and Categorical Thinking when it
was entered into the hierarchical regression analyses.
Unexpectedly, leaders with high experiential processing tend to show a
preference for less competent followers. Experiential processing is associated with
the development and maintenance of effective relationships (Norris & Epstein,
2011). It may be that the nature of more competent followers increases the chance
for leader-follower conflict. Specifically, competent followers show increased
general and job-related knowledge which, while positive in some regards, can result
in an amplified opportunity for the follower to identify and voice issues and
159
problems with the decisions, ideas, and directions of their leader. This is due to more
nuanced job/task understanding and a general cognitive ability to think through
problems. This can increase the chance of leader-follower conflict (Argyris & Schön,
1978) and, thus, the potential to experience negative vibes and emotions during such
interactions (e.g. Carver et al., 1985; Meyer & Starke, 1982; Sachs, 1982; Swann &
Read, 1981). Consequently, leaders with high experiential processing styles may
come to develop preferences for less competent followers who would tend not to
question or create opportunities for such conflict and instead maintain a positive
working relationship.
Interestingly, leaders with high experiential processing styles also tended to
prefer more conformist followers. This was not initially expected, but may be due to
how these leaders tend to interpret the actions conformist followers. Specifically,
given the importance of emotionality in the experiential system, leaders who
experience strong emotions may be especially susceptible to interpreting the actions
of non-conformist followers as a potential threat (e.g. triggers their competitiveness;
e.g. see Burris, 2012). For example, less conformist followers may be more likely to
offer conflicting viewpoints to the leader and be harder to influence. Therefore,
leaders who tend to experience strong emotions, may show an increased chance that
non-conformist followers will engage in behaviours that have the potential to trigger
a threat reaction in them. This can result in the experience of intense negative affect
(e.g. frustration, anger, anxiety, and fear) for the leader. Consequently, they would
tend to develop preferences for more conformist followers who are less likely to
psychologically threaten them. However, while both incompetence and conformity
both showed significant positive relationships with experiential processing provided,
they provided no significant increase in explaining either incompetence or
160
conformity preferences when entered into their respective hierarchical regression
analyses suggesting that other variables are more important from a theoretical
perspective.
Constructive Thinking – IFT Congruence
Perhaps the most striking findings of the current study were those related to
leaders’ preferences for conformist followers. The broad finding that leaders with
high levels of global constructive thinking, especially the constructive dimensions of
emotional and behavioural coping, tend to prefer less conformity in their followers
appears to be quite important for the wider leadership and followership theoretical
fields. Effective followership (Kelley, 1988; 1992; 2008), courageous followership
(Chaleff, 1995; 2003; 2008; 2009) and proactive behaviour literature (Crant, 2000)
all propose that the followers’ willingness to challenge leaders is one of the key
dimensions of effective followership. However, leaders do not interpret such
behaviours in the same way and limited theory and research exists as to why this is
the case (see Grant & Ashford, 2008; Grant et al., 2009).
One explanation the current study lends itself to is that leaders who are more
constructive thinkers are generally more cognitively and emotionally open (i.e.
“courageous” in the language of Chaleff [2009]) to those followers who are willing
to speak up and question the status quo (i.e. non-conformist/proactive). Given the
cognitive and interpretative processes involved in constructive thinking (see Chapter
1 and Epstein, 1998 for overview), leaders with good constructive thinking may be
less likely to process the actions of proactive followers along a destructive pathway
which would generally result in the experience of negative vibes and emotions – a
161
negative need state (see Epstein, 1998; 2014). In contrast, leaders who tend to
process information along more constructive pathways would be more protected
from the experience of such negative affect. Indeed, previous research has found that
even if the actions of a proactive employee are intended to benefit their organisation,
some leaders see these actions as personally threatening and/or as an act of defiance
(Burris, 2012). The current research, therefore, positions CEST (Epstein, 1998;
2003) as a potential theoretical framework for understanding how and why cognitive
and emotional tendencies in leaders may stifle or support proactive followership
behaviours.
The fact that leaders with poor emotional coping tended to prefer conformist
followers reinforces the above theoretical position. Followers who question such
leaders’ actions, and offer alternative perspectives, have the potential to elicit doubt,
anxiety, fear or confusion in them (Argyris & Schön, 1978; Carver et al., 1985;
McCann & Higgins, 1988; Meyer & Starke, 1982; Sachs, 1982; Swann & Read,
1981). These behaviours also have the potential to trigger conflict between the leader
and follower (Argyris & Schön, 1978). All of these are unsatisfying need states for
these leaders (i.e. the need to reduce unpleasant feelings; Epstein, 2014). Therefore,
it appears leaders with poor emotional coping may be more likely to interpret the
behaviours of non-conformist followers as a threat and consequently trigger the
experience of negative affect. Thus, they would develop preferences for more
conformist followers.
Given that emotional coping is concerned with how effectively individuals
deal with the inner world of feeling (Epstein, 1998), the focus of emotional coping
from the perspective of leaders is more about the level of perceived internal threat
being triggered by non-conformist followers (e.g. see Grant et al., 2009).
162
Behavioural coping though is concerned with how effectively individuals overcome
issues in the external world (Epstein, 1998; 2001). From the perspective of leaders’
(whose role tends to be focused on achieving performance goals), therefore, the
content of processing may be more about the degree of perceived external threat
attributed to a follower for potentially undermining the achievement of valued
outcomes. In the context of conformist followers, leaders with good behavioural
coping may, through these experiences, attribute the passivity of conformist
followers as a threat to undermining valued team or organisational goals. Proactive
personality literature generally supports these premises (e.g. Erdogan & Bauer, 2005;
Fuller & Marler, 2009; Zhang, Wang, & Shi, 2012). For example, Zhang and
colleagues (2012) found that followers who exhibited lower levels of proactive
personality in comparison to their leader showed lower quality relationships and
poorer work outcomes. Furthermore, the congruence between leaders and followers
in terms of their proactive personality resulted in a tendency to share the goal of
improving the work-environment.
Similar to how conformist followers are seen, it may be that an incompetent
follower may be perceived by leaders with good behavioural coping as a threat to
achieving valued team/organisational goals. This is due to the followers’ lack of
experience, job-related and/or general knowledge and tendency to be more indolent.
These behaviours may particularly frustrate leaders who are action-oriented and
focused on solving problems (i.e. good behavioural copers). Indeed, meta-analytical
research has found that work experience is positively related to both objective and
subjective measures of job performance (Quińones et al., 1995). Furthermore, as
previously mentioned, when followers show lower levels of proactive behaviours
compared to their leader, the latter reports a lower quality relationship and poorer
163
follower performance (Zhang et al., 2012). These findings suggest that judgements
around the followers’ performance are based on the leaders’ own problem-focused
and behavioural action tendencies. Consequently, interactions between leaders with
good behavioural coping with incompetent/competent followers over time would
become associated with their experience of negative/positive vibes due to their
ability to support/not support the achievement of valued team/organisational
outcomes. This would result in a learned preference for more competent followers.
As expected, leaders with high levels of categorical thinking tended to prefer
more conformist followers. Categorical thinkers tend to make more broad
generalisations about groups of people (Epstein, 1998; 2001). Clear distinctions
between one’s own group and another can result in an individual seeing other groups
as a potential threat (Epstein, 2001). Categorical thinkers also experience irritation
and anger more rapidly when things do not match their expectations and tend to see
others who hold different perspectives as in fault (Epstein, 1998; 2001). In the
context of the leader role, leaders with high categorical thinking may result in non-
conformist followers being perceived as in error and/or as a potential threat source.
In either case, these are likely to bring about a degree of irritation, frustration, anger
or anxiety for leaders with high categorical thinking – negative need states (see
Epstein, 1998). Leaders with high categorical thinking, therefore, would tend to
develop preferences for conformist followers as they would be more inclined to
simply follow the directions of leaders and not question the status quo (i.e. easily
influenced and follow trends). The proactive followership literature as already
described generally supports this proposition (i.e. all things being equal, proactive
behaviour can be interpreted by leaders in different ways; see Argyris & Schön,
1978; Fuller & Marler, 2009; Grant & Ashford, 2008; Grant et al., 2009; Zhang et
164
al., 2012). Indeed, unlike their positive intention, some managers can interpret
followers’ proactive behaviours quite negatively (e.g. see Argyris & Schön, 1978;
Carver et al., 1985; Fuller & Marler, 2009; Grant et al., 2009; Meyer & Starke, 1982;
Sachs, 1982; Swann & Read, 1981). This suggests that the general trust leaders have
in people is a potential key factor in how effective follower behaviours will be
perceived by them (e.g. see Chaleff, 1995; 2003; 2008; Kelley, 1988). These
findings are also aligned with the descriptions of Theory X and Theory Y managers
(see McGregor, 1960). Specifically, that Theory X managers are less trusting of their
employees to do the right thing compared to Theory Y managers.
Leaders with high categorical thinking also showed increased endorsement
for less competent followers. More experienced followers and those who are
educated are likely to hold more job-related knowledge and a general cognitive
ability, which support an increase in performance and creativity (Amabile, 1998;
Quińones et al., 1995; Tierney & Farmer, 2002; Weisberg, 1999). As previously
discussed, one of the potential outcomes of this is an amplified opportunity for
followers to identify issues and problems with the decisions and ideas of their leader.
Consequently, competent followers may be more likely to question leaders and their
directions due to their more nuanced understandings of, and ability to break down
into their component parts, tasks and problems. As already noted, categorical
thinkers also tend to see others in error when they hold different perspectives and are
quicker to experience irritation and anger (Epstein, 1998; 2001). Interactions with
more competent followers, therefore, provide the opportunity for them to take a
different perspective to the leader. Consequently, leaders with high categorical
thinking may tend to perceive such followers as in error, which consequently results
in the experience of negative affect (e.g. frustration and/or anger).
165
As expected, naively optimistic leaders tended to endorse preferences for
enthusiastic followers. Naïve optimists tend to endorse commonly accepted ideas
and beliefs and are generally highly-spirited individuals who like people and are
liked by others (Epstein, 1998). Naïvely optimistic leaders thus would tend place a
stronger emphasis on positive relationships in the working environment. The
enthusiasm factor contains items that appear to support the development of positive
relationships (i.e. outgoing and excited). Specifically, the nature of enthusiasm bears
a resemblance to descriptions of extraverts (see Costa & McCrae, 1997; Eysenck &
Eysenck, 1975). Extraverts are described as enjoying human interaction and being
enthusiastic, talkative and outgoing. Extraversion is positively related to lower levels
of conflict (Barrett & Pietromonaco, 1997) and with having more positive
relationships both inside (Robertson & Kinder, 1993) and outside of work (Lopes,
Salovey, & Straus, 2003; White, Hendrick, & Hendrick, 2004). Enthusiastic
followers, therefore, are likely to be seen as supporting the development of positive
leader-follower working relationships. These outcomes would be congruent with the
valued behaviours of naïvely optimistic leaders. Consequently, such leaders would
tend to develop preferences for enthusiastic followers. Due to the low reliability
observed on the enthusiasm dimension though, it is possible that alternative
explanations exist. For example, naively optimistic leaders may specifically prefer
outgoing followers rather than enthusiastic followers per se, on the assumption that
these two traits are not necessarily underpinned by the same enthusiasm construct.
While not as strong as the other results, it appears relevant to mention here
that strong emotional copers also tend to prefer enthusiasm in their ideal followers.
The internal focus of emotional coping (Epstein, 2001) may indicate that poor
emotional copers tend to associate an enthusiastic follower as a threat. Emotional
166
coping is positively related to extraversion (Epstein, 1983) and strongly negatively
correlated with neuroticism (Epstein, 1992). As noted above, the enthusiasm IFT
factor is similar to the traits that describe extraverted individuals (see Costa &
McCrae, 1997; Eysenck & Eysenck, 1975). Extraverted individuals tend to be more
forceful and use more assertive conflict resolution styles (Boora & Shanti, 2010;
Vestewig & Moss, 1976). Neurotic individuals, in contrast, are less assertive
(Vestewig & Moss, 1976) and tend to use more avoidant conflict styles as they see
conflict as a threatening event and consequently experience high levels of anxiety
and fear (Rahaman, Mollah, & Uddin, 2010). Hence, for leaders with poor emotional
coping, the more assertive and intense nature of enthusiastic followers could be
particularly threatening and anxiety provoking. Consequently, over time, leaders
with poor emotional coping would come to develop preferences for less enthusiastic
followers. However, it is important to note that, most likely due to a small sample
size, emotional coping only approached significance in the regression analyses
suggesting that Naïve Optimism may be more theoretically relevant.
It was interesting to find that leaders who are low in esoteric thinking tend to
prefer competent followers. Indeed, the finding that esoteric thinking independently
contributed beyond rational thinking to predicting preferences for incompetence in
followers is important when one recognises that high esoteric thinkers tend to be less
likely to use engage their rational system to critically consider issues (Epstein, 1998).
This reduced engagement of critical faculties would help to explain their preference
for leaders who are less competent. Specifically, a competent follower would be
more likely to question a leader’s decisions and action due to increased knowledge
via education and experience, which would likely increase the need for the leader to
engage their rational system to critically assess the follower’s input. From the
167
perspective of a leader with high esoteric thinking, this would not align with their
preferences or tendencies for non-rational thought. Consequently, they would tend to
develop preferences for incompetent followers who are less likely to be in a position
(in terms of knowledge and confidence) to question them.
Further theoretical and practical implications
There are a number of additional theoretical and practical implications of the
current findings. First, Kruse and Sy (2011) demonstrated that manipulating affect
can influence the expression of IFTs. Specifically, increased positive affect increased
people’s expression of prototypical followership theories. However, given the more
transient nature of affect, these changes are likely to be somewhat short-lived.
Consequently, once the affect has worn off, people are likely to revert to their
original IFTs. From the perspective of CEST though, affect is an outcome of
underlying cognitive processes occurring in the experiential system (Epstein, 1998;
2003; 2014). Changes in these underlying cognitive processes, therefore, should
provide more stable and long-lasting changes in IFTs. This is because they change
the underlying cognitive processes that contribute to a leader’s affective experiences.
The CEST personality framework and the findings by Kruse and Sy (2011) in the
context of the current study suggest that targeting the cognitive processes in the
experiential system would result in respective changes to leaders’ IFTs. In support,
Cerni and colleagues (2010) have shown that developmental interventions targeting
the improvement of information processing style and constructive thinking in leaders
can increase subordinate rating of transformational leadership. Other research has
further demonstrated a positive relationship between leaders’ prototypical
followership theories and transformational leadership (Duong, 2011). These two
168
studies are highly suggestive that manipulation of the cognitive processes in the
experiential system is casually linked to leaders’ IFTs.
Second, similar to Chapter 3, the findings here have implications for team
composition decisions. Situational leadership (Blanchard, 1985; Blanchard et al.,
1985; Chemers, 1997; Hersey & Blanchard, 1969; 1977; Northouse, 2007) and the
path-goal theory of leadership (Chemers, 1997; House, 1971; 1996; House &
Mitchell, 1975) argue that leaders need to adapt to the personality and ability of
followers to bring about the best outcomes. For example, situation leadership models
argue for the need to adapt their use of directive (or task behaviours) and supportive
(or relationship) behaviours based upon followers’ maturity (psychological and job)
levels (Chemers, 1997; Graeff, 1983; Northouse, 2007). Job maturity, defined as the
follower’s level of experience, knowledge and understanding of the task (Chemers,
1997), appears to be quite similar in nature to the incompetence IFT trait (i.e.
uneducated, inexperienced, slow). This suggests that high rational leaders would be
less suited to those followers with low job maturity as they would show lower
experience, knowledge and understanding of a task all of which are less preferred in
a rational leader.
Related to above are implications for relationship based perspectives of
leadership (e.g. leader-member-exchange, see Dansereau et al., 1975; Graen & Uhl-
Bien, 1995; relational view of leadership, see Hollander, 1971; 1992; 2012). These
theories generally recognise the leader as a specific individual in a relationship with
their followers and that leadership is the process which occurs between them. Both
IFTs and ILTs act as a framework for interpreting the behaviours of followers and
leaders respectively and for guiding behaviours in relation to them (see Lord &
Maher, 1993; Sy, 2010; Whiteley et al., 2012). The current theory argues that leaders
169
IFTs and followers ILTs, develop, in part, through their experiences with followers
and leaders respectively. These experiences though are filtered through their
cognitive tendencies (i.e. information processing style and degree of constructive
thinking). Furthermore, these tendencies influence how people generally respond to
daily events, including those in the workplace. This, therefore, conceptually
positions CEST and implicit leadership and followership theories within a
potentially important theoretically framework for understanding how relationships
between leaders and followers develop.
Limitations
Due to the similarities in theory and research design between the current and
previous study, the current research has similar limitations. First, the use of explicit
measures to assess implicit theories have been acknowledged as a potential issue in
that the measure used to assess the concept are not aligned (Uhlmann et al., 2012).
The core issue is that an individual’s awareness of what is being measured can
influence how they respond. Consequently, relationships measured using implicit
measures may be different to those found here. It is important to note though that
Epitropaki and colleagues (2013) recently stated that “both indirect and direct
measures can advance our understanding of how controlled and uncontrolled
processing shapes leader–follower processes and outcomes” (p.866). Indeed, the use
of a general working population made the use of an indirect measure less suitable
from a logistical perspective. Therefore despite the potential methodological
mismatch, the current study still is relevant and important. Future research though
may wish to use an implicit IFT measure to assess whether similar findings emerge.
170
Second, situational factors were not assessed but conceptually may have
moderated the relationships observed (e.g. see Derler & Weibler, 2014). However, it
would seem that the use of a more abstract target (i.e. ideal follower in a business
setting) would be less influenced by their current organisation culture. Indeed, as the
current findings emerged in a diverse set of currently employed managers using two
reliable and valid personality measures, the results would appear as potentially
generalisable to a variety of business settings.
Third, while adequate for the analyses undertaken, the sample size was
somewhat low. Therefore, some of the relationships may be obscured due to a lack
of power to detect significant relationships. Fourth, as CEST focuses primarily on
the experiential system (Epstein, 2003), the contributions of the rational system
could be considered somewhat under-represented. As noted in the previous chapter
though, from the perspective of CEST (Epstein, 2001; 2003), research assessing the
relationships between the rational components of personality and ILT’s have already
undertaken in the literature (e.g. Keller, 1999).
Fifth, the singular variable approach, which was adopted limits some of the
interpretability of the findings (Foti & Hauenstein, 2007). Instead a pattern approach
could provide more insight as it can be considered as offering a more holistic
understanding of leader IFTs (e.g. see Foti & Hauenstein, 2007). Specifically, the
pattern of the relationships between the IFT dimensions that a leader holds rather
than any one specific IFT dimension. As noted in the previous chapter, this can be
considered a complementary approach to the variable approach, not as a substitute
(Foti & Hauenstein, 2007). Consequently, the current results are still important for
extending ILT theory and research. However, future research may wish to consider a
pattern approach.
171
CHAPTER 5
Leaders’ Information Processing Style and Constructive Thinking as Follower
LMX Antecedents: The Moderating Influence of Followers’ Implicit Ideal
Leadership Theory
In Chapter 3 it was found that followers’ information processing style and
constructive thinking are related to their implicit ideal leadership theories (herein
shorted to IILT for simplicity). In Chapter 4, it was found that leaders’ information
processing style and constructive thinking are related to their implicit ideal
followership theories (herein shortened to IIFT for simplicity). The theoretical
approach adopted in these studies was that IILTs and IIFTs, in part, through
followers’ and leaders’ experiences with leaders and followers respectively.
Specifically, experiences between leaders and followers are filtered through each of
their own information processing tendencies (Norris & Epstein, 2011) and the
different factors of constructive thinking (e.g. emotional coping; Epstein, 2001). It is
assumed that individuals experience positive vibes and emotions when such
interactions are experienced in ways that are congruent with their information
processing style and constructive thinking tendencies (see Berne, 1961; Epstein,
2003). In contrast, they experience negative vibes or emotions when interactions
require or pressure them to deliberate, act or converse in ways which are not
congruent with these aspects of the self (e.g. see Epstein, 2003; McCann & Higgins,
1988). These emotional experiences provide the learning opportunity for both
leaders and followers to associate particular follower or leader traits and behaviours
with their experience of positive or negative affective experiences which results in
the development of leaders’ IIFTs and followers’ IILTs.
172
Given the notion of leader-follower interaction in this theoretical
underpinning, the findings presented in the previous two chapters are important
when viewed through relationship based perspectives of leadership (e.g. leader-
member exchange; see Dansereau et al., 1975; Graen & Uhl-Bien, 1995). Drawing
upon leader-member exchange, a relationship oriented theory of leadership (see
Dansereau et al., 1975; Graen et al., 1973; Graen & Uhl-Bien, 1995), the next two
chapters seek to test two general models for understanding how leader and follower
information processing styles and constructive thinking relate to follower and leader
perceptions of the relationship respectively. In addition, given that implicit
followership and leadership theories act as frameworks for interpreting the
behaviours of followers and leaders respectively and for guiding behaviours in
relation to them (see Lord & Maher, 1993; Sy, 2010; Whiteley et al., 2012), these
models also seek to understand whether these relationships are moderated by
leaders’ IIFTs or followers’ IILTs. This is based on the assumption these will
influence how followers and leaders perceive, encode and interpret the dyadic
partners behaviours.
Leader-Member Exchange (LMX) Theory
Leader-follower relationships have been observed as “central to
organizational functioning” (Dulebohn et al., 2012, p. 1744). Consequently, Leader-
Member Exchange (LMX) with its focus on leader-follower relationships has
garnered much research attention in the leadership domain and has gone through
many theoretical refinements (Graen & Uhl-Bien, 1995). LMX is a social exchange
perspective as it assumes a give-and-take of valued emotional and tangible resources
173
between a leader and follower (Blau, 1964; Graen & Scandura, 1987; Liden, et al.,
1997; Sparrowe & Liden, 1997). It explicitly recognises that followers are a
heterogeneous group who can perceive, interpret and react in very different ways to
the same things and that managers can act in different ways with each follower
(Dansereau et al., 1975; Graen et al., 1973). Specifically, LMX takes the position
that leaders and followers negotiate different types of exchange relationships which
fall on a continuum from low-quality or purely contractual based (e.g. giving of
specified monetary reward in exchange for completion of agreed tasks) to high-
quality relationships characterised by mutual trust, professional respect, liking and
reciprocal influence (Dienesch & Liden, 1986; Gerstner & Day, 1997; Liden &
Maslyn, 1998). Followers who have low-quality relationships with leaders are seen
as the ‘out-group’ whereas those with high-quality relationships are seen as the ‘in-
group’ (Liden & Maslyn, 1998). While the latter group of followers have better
relationships, leaders have correspondingly higher performance and loyalty
expectations (Hollander & Offermann, 1990).
According to LMX theory, effective leadership occurs when the relationship
moves to a mature level (Graen & Uhl-Bien, 1991). The Leadership Making Model
(Graen & Uhl-Bien, 1991; 1995) provides both a description of how leaders and
followers move into this mature stage in addition to prescribing how leaders can
improve each individual leader-follower dyadic relationship to bring this about
(Graen & Uhl-Bien, 1995). According to this model, the relationship goes through
three stages: (1) stranger stage; (2) acquaintance stage; and (3) mature partnership
stage (Graen & Uhl-Bien, 1995). The stranger stage occurs when leaders and
followers first meet. The lack of knowledge each holds on the other results in
interactions based more upon their respective jobs and social roles – a more
174
transactional or contractual exchange of resources. An initial ‘offer’ from either
party (but typically leaders due to their more powerful position) is made at the end of
the stranger stage to improve the work-relationship (Graen & Uhl-Bien, 1995). If
accepted and reciprocated by the partner, then the relationship moves into the
acquaintance stage.
During the acquaintance stage, there is a continuing exchange of resources or
contributions between leaders and followers but these move beyond simple
contractual exchanges to include more socially valued resources (e.g. emotional
support; Graen & Uhl-Bien, 1991; 1995). However, the resources are still somewhat
limited as each party is essentially continuing to test the other’s commitment to the
relationship. Assuming each is committed, these continuing exchanges help to
positively reinforce the relationship by meeting each other’s needs. In other words,
when one member of the dyad addresses the other’s needs then this will be
reciprocated as per the norm of reciprocity (Blau, 1964).
During the mature partnership stage, the limited exchanges observed in the
acquaintance stage become highly established (Graen & Uhl-Bien, 1995). The
relationship is highly interdependent and of high-quality (Dienesch & Liden, 1986;
Gerstner & Day, 1997; Liden & Maslyn, 1998). Leaders and followers show high
levels of interest in helping each other meet their needs and goals. Each party though
must continue to perceive the other as equally or exceeding the individuals’ own
contributions to the relationship otherwise the relationship may become jeopardised
(see Buunk, Doosje, Jans, & Hopstaken, 1993; Walster, Walster, & Berscheid,
1978). This applies to all the Leadership Making Model stages. This has the
consequence that not all leader-follower relationships will reach the mature
175
partnership level as partners may not reciprocate (or, at least, be perceived to
reciprocate) each other’s contribution.
There is now substantial evidence demonstrating that mature leader-follower
partnerships contribute to positive individual and organisational outcomes (e.g.
Dulebohn et al., 2012; Erdogan & Liden, 2002). When high-quality relationships
exist, there is less turnover, increased performance evaluations, increased
organisational commitment in both followers and leaders, less job stress, and greater
innovation and creativity (Dulebohn et al., 2012; Duchon, Dunegan, & Uhl-Bien,
1992; Erdogan & Liden, 2002; Graen & Uhl-Bien, 1995; Judge & Ferris, 1993;
Liden et al., 1993). In addition, followers work harder (Basu & Green, 1997; Duchon
et al., 1986), engage in more effective communication (Fairhurst, 1993; Fairhurst &
Chandler, 1989), show increased organisational commitment (Nystrom, 1990) and
increased performance (Dulebohn et al., 2012; Seers & Graen, 1984). Furthermore,
many of these outcomes are observed across different cultures and nations (Anseel &
Lievens, 2007; Eden, 1993; Erdogan et al., 2006; Schyns et al., 2005; Wakabayashi
& Graen, 1984), which demonstrates that relationship quality is a highly
generalisable construct (van Gils et al., 2010). Given the wide-ranging important
organisational outcomes, it becomes important to understand LMX’s antecedents.
Relative to its outcomes, LMX antecedents have been less researched
(Mahsud et al., 2010). A recent meta-analysis by Dulebohn and colleagues (2012)
identified several follower characteristics, which were related to LMX. The
competence and ability of followers were positively related to their reported LMX.
Followers’ personality traits of agreeableness, conscientiousness, extraversion,
internal locus of control, and positive affectivity were also all found to positively
relate to their perceptions of LMX quality. Such traits/behaviours are likely to be
176
viewed by leaders as positive and productive as they help leaders move towards
valued goals (e.g. meeting sales targets). Correspondingly, leaders seek to further
develop and maintain positive relationships with these types of followers to retain
these benefits. In support, conscientiousness is one of the most reliable indicators of
job performance (Barrick & Mount, 1991), while agreeableness is associated with
helping behaviours, cooperation (Graziano et al., 2007) and reciprocity (Perugini et
al., 2003). Conscientiousness and agreeableness in followers, therefore, are likely to
increase leaders’ respect and trust resulting in more effective working relationships.
Leader characteristics that were found to positively influence follower LMX include
extraversion, agreeableness, transformational leadership and contingent reward
behaviour (Dulebohn et al., 2012). Follower ILTs have also been found to predict
their LMX rating (e.g. Bass & Avolio, 1989; Eden & Leviatan, 1975; 2005; Lord et
al., 1984; Rush et al., 1977; Schyns & Meindl, 2005; Uhl-Bien et al., 2014).
Linking Cognitive Experiential Self Theory and LMX
Although the above demonstrates links between leaders and followers’
personality traits and LMX, no research to the author’s knowledge has considered
whether information processing style or the experiential aspects of personality relate
to LMX. There are, however, theoretical reasons to believe that both leaders’ and
followers’ information processing styles (Norris & Epstein, 2011) and the degree
they think constructively (Epstein, 2001) should be linked to LMX. A person’s
information processing style and constructive thinking have significant influences on
individuals’ subsequent behaviours across a wide range of contexts and situations
due to these factors having wide-ranging connections throughout the self-concept
177
(see Epstein, 2003; 2014). Indeed, Cognitive-Experiential Self Theory (CEST) states
that most of our behaviours are influenced by the schemas in the experiential system
(Epstein, 1994; 2001).
Consequently, the general stability of these personality constructs should
have a strong influence on the general behaviours that are manifested in leader-
follower relationships. In support, previous research demonstrates links between
information processing style and constructive thinking with specific leader and
follower behaviours (e.g. Cerni et al., 2012; Curtis & Lee, 2013). For example,
leaders with high rational information processing and good behavioural coping tend
to use more rational influencing tactics and more effective conflict handling
techniques (e.g. integrating ideas; Cerni et al., 2012; Curtis & Lee, 2013). Such
behaviours will be observed and evaluated by their dyadic counterpart in the leader-
follower relationship. Therefore, it is proposed that leaders’ and followers’
information processing style and constructive thinking will predict their dyadic
partner’s perception of relationship quality.
However, implicit leadership theory (ILT; Eden & Leviathan, 1975; Lord &
Maher, 1990; 1993) and implicit followership theory (IFT) (Sy, 2010) stipulate that
the observer of the behaviour (i.e. the dyadic counterpart) will draw upon their
implicit theories of leaders and followers to guide these observations and
evaluations. This suggests that followers and leaders may evaluate the same
behaviours in their dyadic counterpart quite differently due to differences in role
expectations. This in turn would likely result in different judgements of the same
relationship from the perspective of leaders and followers. Yet this explicit
recognition is not typical in the LMX literature. In fact, a particularly pertinent issue
in LMX literature is that relationship quality has typically been measured from the
178
followers’ perspective on the assumption that this relationship will be similarly
experienced by both leaders and followers (van Gils et al., 2010; Wilson et al.,
2010). This is unsurprising given that such an approach is aligned with the premise
in LMX theory that the relationship is experienced similarly by both leaders and
followers (van Gils et al., 2010). However, research strongly suggests that both
parties typically experience the same relationship in substantially different ways
(Gerstner & Day, 1997; Sin et al., 2009; van Gils et al., 2010; Wilson et al., 2010;).
For example, meta-analytical research demonstrates that correlations between
leaders’ and followers’ LMX are actually quite low (.29-.37; Gerstner & Day, 1997;
Sin et al., 2009). If the assumption that the relationship is experienced similarly by
leaders and followers is correct, then a much stronger correlation should be expected.
As the evidence does not support this, it becomes important to theorise what may be
contributing to the difference.
van Gils and colleagues (2010) argue that the key issue concerns what is
exchanged between leaders and followers. They state “both dyadic partners are
likely to perceive the contribution each person makes to the relationship based on
their expectations for the particular role of the person” (p. 339). Both ILTs and IFTs
are based in role theory as they specify, and organise information around, the traits
and behaviours of both leaders and followers and serve to guide the observers’
interpretations of their respective actions, what a perceiver attends to, what they
encode and the information they retrieve when recalling leader or follower related
information (Eden & Leviathan, 1975; Epitropaki & Martin, 2004; 2005; Lord &
Maher, 1990; 1993; Lord et al., 1984; Phillips & Lord, 1982; Poole et al., 1989; Sy,
2010). van Gils and colleagues (2010), therefore, draw upon ILTs and IFTs as
potential constructs to explain LMX disagreement. Specifically, IFTs and ILTs
179
should influence how leaders and followers perceive, interpret and behave when
interacting with their dyadic counterpart.
While van Gils and colleagues (2010) are correct to draw attention to the role
IFTs and ILTs play in influencing both the leaders and followers perception of the
dyadic relationship, they do not appear to explicitly consider the target schema (e.g.
leader, ideal leader, follower, ideal follower) the perceiver will draw upon to guide
their observations, expectations and interpretations. ILT and IFT ratings differ
depending on the target schema used (Offerman et al., 1994; Sy, 2010). For example,
the targets of leaders and supervisors are typically rated as more tyrannical than the
target of effective leaders (Offerman et al., 1994). These differences should,
therefore, influence how they perceive and make subsequent judgements of the
observed leader or follower (see Bass & Avolio, 1989; Lord et al., 1984; Rush et al.,
1977). Consequently, it is necessary to theorise what the target of the ILT or IFT
schema is used in everyday work-setting as this is likely to have implications on the
LMX rating by the perceiver.
It is proposed that both leaders and followers will, in general, draw upon their
IIFT or IILT respectively rather than a more general implicit followership or
leadership theory to perceive, interpret, and evaluate their dyadic counterpart. IIFTs
or IILTs would be more likely to take into consideration the personal preferences of
the follower or leader which a more general ILT or IFT would not. This is suggested
by the theory and findings presented in Chapters 4 and 5 which imply that an
individual leader’s or follower’s ideal dyadic partner (i.e. their IIFT or IILT
respectively) would account for their own personal cognitive-emotional tendencies
(i.e. their information processing style and degree of constructive thinking). In
contrast, the more universal targets of a “leader” or “follower” would theoretically
180
be less likely to take these preferences into account as the focus is more on the
cultural/stereotypical beliefs of what leaders tend to be in general (see Lord et al.,
1984; Offerman et al., 1994). In an everyday leader-follower relationship, given that
both leaders and followers are interdependently influencing each other’s thoughts,
emotions and behaviours (Hollander, 1992), the use of an ideal leader or ideal
follower target appears to be more likely to be drawn on when making judgements
about their dyadic partner.
Based on the theory and research presented above, the remainder of this
chapter and following chapter present two studies which seek to: (1) explore whether
leaders and followers information processing styles and degree of constructive
thinking relate to their partner’s (i.e. follower or leader respectively) LMX rating;
and (2) see whether these relationships are moderated by followers’ IILTs and
leaders’ IIFTs respectively. Conceptually, this is presented in Figures 5.1a and 5.1b.
Drawing on person-environment fit (P-E fit) theory (see previous chapters; Kristof-
Brown et al., 2005), at the most basic level it is anticipated that follower IILTs and
leader IIFTs which are aligned or are congruent with their dyadic partners
information processing style and constructive thinking will result in the hypothesised
relationships (specific hypotheses based on these general models are outlined below
and in the next chapter) emerging as stronger. To aid readership, the current chapter
focuses on the how leaders’ information processing styles and constructive thinking
relate to their followers’ LMX perceptions and whether the followers’ IILT
moderates this relationship (see Figure 5.1a). The following chapter focuses on how
followers’ information processing styles and constructive thinking relate to their
leaders’ LMX and whether the leaders’ IIFT moderates this (see Figure 5.1b).
181
Figures 5.1a. and 5.1b. Graphical Representation of Proposed Moderation Models.
Follower LMX Dependent Variable
Drawing upon the general moderation model proposed in Figure 5.1a,
specific hypotheses for leaders’ information processing and constructive thinking
dimensions and their relationships to followers’ LMX are expanded on below. To
maintain a consistent theoretical approach across all studies in this dissertation, the
IILT dimensions observed in Chapter 3 were used in this study. These dimensions
included Sensitivity, Tyranny, Intelligence, Attractiveness and Dedication. Table 5.1
reintroduces the reader to the 22 items found to make up these dimensions.
182
Table 5.1.
Implicit Ideal Leadership Theory Dimensions and Items from Exploratory Factor
Analysis conducted in Chapter 3 Ideal ILT Factor Items in Factor
Sensitivity Sensitive; Compassionate; Sympathetic; Warm; Sincere
Tyranny Selfish; Manipulative; Power-Hungry; Conceited; Loud
Intelligence Intellectual; Knowledgeable; Clever; Intelligent
Attractiveness Well-dressed; Well-groomed; Classy
Dedication Motivated; Dedicated; Goal-Oriented; Enthusiastic; Hard-
working
Hypothesis development: Leader Information Processing and Follower LMX
High rational processors are deep thinkers, tend to develop clear, rational and
explainable reasons for decisions, and enjoy thinking abstractly (Pacini & Epstein,
1999a). High rational processors also tend to be more problem-solution focused
(Epstein et al., 1996). Being problem-solution focused is vital for leaders to
overcome complex social problems, which arise in organisations (Mumford,
Zaccaro, Harding, Jacobs, & Fleishman, 2000a). Rational processing, therefore, may
help to support leaders in navigating the complex organisational social environment
in order to bring about more effective and positive working relationships. In support,
leaders with high rational information processing styles correspondingly use rational
influencing tactics and integrative conflict handling techniques (Cerni et al., 2012;
Curtis & Lee, 2013). Leaders who adopt integrative conflict handling styles tend to
have followers who report higher quality relationships with the leader (Green, 2008).
Leader rational information processing, therefore, should be positively related to
followers’ LMX.
As per the general moderation model presented, the items, which make up the
intelligence dimension of IILTs (see Table 5.1) appear to be most aligned with
valuing (or not) the behavioural tendencies of leaders with high levels of rational
processing (e.g. rational decision making and communication tactics). In support,
183
followers with high levels of rational processing tend to prefer more intelligent
leaders (see Chapter 3) suggesting preferences for similar traits and behaviours as
themselves. Therefore, it is hypothesised that:
Hypothesis 1: a significant positive relationship will exist between leaders’ rational
processing and followers’ LMX…
Hypothesis 1a: ...and this relationship will be stronger for followers’ with higher
preferences for leader intelligence.
High experiential processors tend to be superior at forming and maintaining
positive social relationships (Norris & Epstein, 2011; Pacini & Epstein, 1999a).
Empathy is an important aspect of developing relationships in all contexts including
those at work as it helps individuals to take another’s perspective and re-experience
their emotions (Katz, 1963; Yukl, 1998). Empathy importantly requires imagination
as this helps an individual take another’s perspective (Kellett et al., 2002; Rudebeck,
2002). Imagination is a component of experiential processing (Norris & Epstein,
2011). Therefore, high experiential processing in leaders may help improve leader-
follower relationships. In support, the strongest correlates of follower LMX tend to
occur when leaders use more relationship-oriented behaviour such as supporting and
developing their followers (Yukl & Fu, 1999; Yukl, O’Donnell, & Taber, 2009).
The ILLT dimension that appears to contain those items most congruent with
valuing or emphasising relationship based leader behaviour is sensitivity (see Table
5.1). This is because the leaders’ traits and behaviours associated with sensitivity
184
(e.g. compassionate, warmth, sincerity) are highly similar to those proposed in
conceptualisations of empathy (see Davis, 1983). Consequently, the behavioural
tendencies of leaders with high experiential processing styles should be more
congruent with those followers who endorse high levels of sensitivity in their ideal
leader. It is, therefore, hypothesised that:
Hypothesis 2: a significant positive relationship will exist between leaders’
experiential thinking and followers’ LMX…
Hypothesis 2a: …and this relationship will be stronger for followers with higher
preferences for leader sensitivity.
Hypothesis development: Leader Constructive Thinking and Follower LMX
As noted in previous chapters, emotional coping is a core dimension of
constructive thinking and is about how individuals deal with frustration,
disappointment, rejection, criticism and disapproval (Epstein, 2001). These are
typical situations that most leaders will face in the workplace. Examples include
frustration and disappointment at not achieving a sales target or having a proposal
rejected by a client. Good emotional copers tend to be able to cope with these
situations without excessive distress (Epstein, 2001). In contrast, poor emotional
copers will tend to take these personally, allow them to undermine their sense of
worth and/or dwell on them to a disproportionate degree. For example, leaders with
poor emotional coping may interpret feedback from one of their followers as
reflecting their (in)competence rather than as a potential source of help. Indeed, in
185
some cases, they may simply avoid or dismiss follower feedback due to the negative
impacts on the self (e.g. see Grant et al., 2009; Ilgen et al., 1979; Kluger & DeNisi,
1996; Menon, 2001; Morrison & Milliken, 2000). High levels of emotional coping,
therefore, should help support leaders to develop high-quality relationships as they
should be able to manage their emotional experiences more effectively when
interacting with others (e.g. low anxiety and an increased focus on the needs of the
follower). In support, research demonstrates that good emotional copers tend to
develop more favourable relationships (Epstein & Meier, 1989). They are also more
optimistic and less introverted (Epstein, 1992) that in turn are both associated with
higher levels of LMX (Murphy & Ensher, 1999). Furthermore, high levels of
neuroticism, of which emotional coping has strong positive relationships with, is
positively related to an avoidant conflict management style (Antonioni, 1998;
Epstein, 2001). Those who use an avoidant conflict management style tend to
experience increased relationship conflict and stress (Friedman, Tidd, Currall, &
Tsai, 2000) which is likely a result of their sensitivity to threat.
The traits and behavioural tendencies of leaders with poor emotional coping
as described above may be less likely to be perceived by followers seeking leader
sensitivity as congruent with this ideal. This is primarily because such traits and
behaviours in these leaders are unlikely to promote or support positive and
supportive interpersonal interactions/communication (e.g. see Cheek & Melchior,
1990; Daly & Stafford, 1984; Melchior & Cheek, 1990; Strand, 2006). For example,
leaders who tend to avoid or dismiss followers’ input and feedback are unlikely to be
ascribed as sensitive or compassionate as they do not seek to take account of
followers’ views and feelings. In support, employee voice literature demonstrates
that being able to feel heard and supported can help people manage their tasks and
186
feelings more effectively (Garon, 2012; Wood, 2008). Furthermore, followers
(especially males) who endorse high sensitivity in their ideal leader tend to show
lower levels of emotional coping themselves (see Chapter 3). This is suggestive that
they tend to seek compensatory leader traits and behaviours to overcome the
follower’s low self-esteem and tendency to experience negative affect. The more
avoidant and pessimistic tendencies of leaders with poor emotional coping are
unlikely to provide this compensatory requirement or support. Consequently, the
proposed positive relationship between leader emotional coping and followers’ LMX
may be moderated by the degree to which followers endorse sensitivity in their ideal
leader.
Hypothesis 3: a significant positive relationship will exist between leaders’
emotional coping and followers’ LMX…
Hypothesis 3a: …and this relationship will be stronger for followers with higher
preferences for leader sensitivity.
High behavioural copers are more likely to have automatic thoughts to
events, which facilitate effective action such as engaging in hard work, effort, and
planning (Epstein, 2001). They also tend to be positive, optimistic thinkers who
focus more on achieving outcomes rather than worrying about things such as
deadlines (Epstein, 1992; 1998; 2001). Furthermore, they tend to be non-judgmental
and instead focus on the effectiveness of particular behaviours in others (Epstein,
1998; 2001). In leaders, all these traits and behaviours are likely to support the
187
development of positive and supportive working relationships with their followers.
For example, being non-judgmental is a key factor in helping facilitate others to
express their true positions and feelings without fear of being judged (e.g. see Chan
& Treacy, 1996; May, Gilson, & Harter, 2004; Nembhard & Edmondson, 2006;
Rogers, 1959; Walumbwa & Schaubroeck, 2009). This openness and safety should
help followers develop trust in their leaders (e.g see Colquitt, Scott, & LePine, 2007;
Gao, Janssen, & Shi, 2011; Mayer, Davis, & Schoorman, 1995; Rousseau, Sitkin,
Burt, & Camerer, 1998). Behavioural coping is also associated with an integrative
style of conflict-handing (Cerni et al., 2012) which, in turn, is associated with
higher-quality relationships (Green, 2008). In addition, positivity and optimism have
been found to support leaders’ ability to communicate effectively, empathise with
their team, and facilitate a climate of innovation and creativity (Dulebohn et al.,
2012; Prati, Douglas, Ferris, Ammeter, & Buckley, 2003; Scott & Bruce, 1994).
Given these links, high behavioural coping in leaders may support the development
and maintenance of high-quality relationships with followers.
It would seem logical to expect that the optimistic and action-oriented nature
of leaders with high levels of behavioural coping would, in particular, appeal to
those followers who endorse high levels of dedication in their ideal leader (see Table
5.1). For example, the items of motivated and enthusiastic would align with the
positive and optimistic nature of leaders with high behavioural coping. Alternatively,
a leaders’ tendency to be conscientious would likely resonate with the dedicated and
hard-working items. The increased tendency for such leaders to be non-judgmental,
though, is also likely to appeal to followers who endorse high levels of sensitivity in
their ideal leaders. This is because being non-judgmental should facilitate
psychologically safe environments in which followers can voice their needs and
188
provide input (e.g. see Botero & Dyne, 2009; LePine & van Dyne, 2001). The
proposed relationship between leaders’ behavioural coping and followers’ LMX,
therefore, should be moderated by the degree to which followers endorse both
dedication and sensitivity in their ideal leaders. Based upon the above, it is
hypothesised that:
Hypothesis 4: a significant positive relationship will exist between leaders’
behavioural coping and followers’ LMX…
Hypothesis 4a and 4b: …and this relationship will be stronger for followers with
higher preferences for leader dedication and sensitivity.
Categorical thinkers believe there is only one right way to do something –
their way (Epstein, 1998; 2001; Epstein & Meier, 1989). Consequently, they tend to
see others in error for simply taking a different position. They are also quick to
experience annoyance and anger when their expectations are violated (Epstein, 1992;
2001) and tend to be more judgmental, intolerant and uncaring towards others
(Epstein, 1992; 2001). Given such tendencies, it was unsurprising that leaders with
high categorical thinking tend to seek more conformist followers (i.e., do not
question the status quo) as they are less likely to question or criticise their leaders. In
contrast, followers who do question or criticise would tend to elicit negative vibes
and emotions in these leaders that are negative need states (see Chapter 4). Such
behavioural tendencies and negative follower beliefs are unlikely to promote the
189
development and maintenance of high-quality leader-follower relationships for a
number of reasons.
First, the ability to trust others is an important factor in interpersonal
cooperation (Kramer & Tyler, 1996; Lewicki, Tomlinson, & Gillespie, 2006;
McAllister, 1995) and the resolution of conflict (Deutsch, 1973; Lewicki et al.,
2006). It also helps individuals take risks and accept vulnerability (Gao et al., 2011).
Therefore, leaders who tend to be generally distrustful of people should be less likely
to develop high-quality leader-follower relationships. Second, those who are
judgmental are unlikely to promote a psychologically or emotionally safe
environment in which others can express their true positions or feelings due to
anxiety and fear of being judged (Chan & Treacy, 1996; May et al., 2004; Nembhard
& Edmondson, 2006; Rogers, 1959; Walumbwa & Schaubroeck, 2009). Finally, the
tendency for leaders with high categorical thinking to endorse conformity (see
Chapter 4) is unlikely to promote positive and supportive interactions and work-
environments for followers (e.g. see theory and research on Theory X and Y
managers; Finman, 1973; McGregor, 1960; Neuliep, 1987; 1996). In support, leaders
who hold more anti-prototypical follower beliefs, of which conformity in one
element, tend to have followers who report lower levels of LMX (Sy, 2010). Based
on these reasons, high categorical thinking in leaders is likely to undermine the
development of a high-quality leader-follower relationship as perceived by the
follower.
This relationship may be moderated by the degree to which followers endorse
sensitivity in their ideal leader. From the perspective of such a follower, the
behavioural tendencies of leaders with high categorical thinking as described above
are unlikely to be perceived or judged as congruent with their preferences. For
190
example, leaders who are intolerant, uncaring and cognitively/emotionally closed to
others’ needs and perspectives are likely to be perceived quite negatively by
someone who seeks compassion and warmth in their ideal leader. It is, therefore,
proposed that:
Hypothesis 5: a significant negative relationship will exist between leaders’
categorical thinking and followers’ LMX…
Hypothesis 5a: …and this relationship will be stronger for followers with higher
preferences for leader sensitivity.
Naïve optimists think in positive and optimistic ways but which differs from
positive thinking as it is unrealistic (Epstein, 2001). Positivity, though, is one of the
most effective communication techniques for influencing others (Cameron, 2008;
Norman, Avolio, & Luthans, 2010) and is a highly valued personal attribute in
general society (Collinson, 2012). Indeed, positivity and optimism can help leaders
construct powerful and attractive visions of the future which can help reassure and
calm followers into thinking that ‘everything will be ok’ (Peters & Austin, 1985).
The positivity associated with naïve-optimism, therefore, should help support the
leader develop positive leader-follower relationships. In support, naïve-optimists
tend to be well liked and are popular in politics (Epstein, 1998; 2014). Naïve
optimists also tend to think in simplistic and stereotyped ways (Epstein, 2001). For
example, they may believe that everyone should try to look happy despite feeling sad
or that all people are good at heart. Such beliefs in leaders are likely to reinforce
191
positive leader-follower interactions as they help promote positivity and optimism.
In support, naïve-optimism, especially the sub-scale of pollyanna-ish thinking, is
positively associated with positive affective traits such as happiness and extraversion
while negatively related to negative affect and neuroticism (Epstein, 1992 as cited in
Epstein, 2001). Consequently, there should be a positive relationship between
leaders’ naïve-optimism and LMX.
This relationship may be moderated by followers who endorse high levels of
attractiveness in their ideal leader. Leaders with high levels of naïve-optimism
appear to engage in behaviours which could be interpreted as impression
management. Impression management is a goal-directed behaviour which aims to
influence the perceptions held by others by regulating and controlling what
information is disseminated during social interactions (Leary & Kowalski, 1990).
For example, the tendency to believe that everyone should try to look happy despite
feeling sad suggests that how one presents oneself may be important. Visual
appearance and outward behaviours are key factors in how people make implicit and
explicit judgements about others (Adam & Galinsky, 2012; Johnson, Schofield, &
Yurchisin, 2002; Rosenberg, Kahn, Tran, & Le, 1991; Slepian, Ferber, Gold, &
Rutchick, 2015). The IILT trait of attractiveness includes the items of well-dressed,
well-groomed, and classy (see Table 5.1). These appear to be congruent with the
theorised self-presentation tendencies of leaders with high levels of naïve-optimism.
Consequently, followers who endorse high levels of attractiveness in their ideal
leader should see such leaders more positively than those who endorse lower levels.
In addition to attractiveness, the proposed relationship may also be
moderated by dedication. It may be that the positivity of leaders with high levels of
naïve-optimism is congruent to some degree with those expectations of followers
192
who endorse high levels of dedication in their ideal leader. For example, these
leaders may say all the right things and appear enthusiastic, motivated and even
provide a positive vision of the future which motivates followers who seek high
levels of dedication. For followers, such behaviours, therefore, would likely resonate
with their preferences for motivated and enthusiastic leaders. Based upon the above,
it is hypothesised that:
Hypothesis 6: a significant positive relationship will exist between leaders’ naïve-
optimism and followers’ LMX…
Hypothesis 6a & 6b: …and this relationship will be stronger for followers with
higher preferences for leader attractiveness and dedication.
Method
Participants
Data was collected from employees and their line managers. These
participants came from a small boutique legal firm, a local branch of a national union
as well as direct participation (i.e. direct participation by individual managers who
then invited their team members to participate). The legal firm was recruited through
professional contacts of the researcher. The national union requested to participate
via a website, which had been set up to support the organisational participation (i.e.
promote the research and provide information), which was advertised via
professional social media (e.g. LinkedIn) and university alumni email distribution
lists, in addition to professional contacts of the researcher. In regard to direct
193
participation, managers and their direct-reports were also eligible to participate.
Where managers indicated they would like to participate, they were asked to send an
invite to their direct reports, which contained a link to a survey specifically
developed for direct reports. The final sample consisted of 106 followers (68.9%
female, 31.1% male) and 26 line leaders (65.4% male, 34.6% female). Average
follower age was 39.40 (SD = 11.09; range 18-62). Average leader age was 41.27
(SD = 8.93; range 26-67). The average length of time a follower and leader had been
working together was 39.09 months (SD = 46.57 range 1-300).
Measures - Leader
Rational-Experiential Multimodel Inventory (REIm; Norris & Epstein,
2011). The REIm (Norris & Epstein, 2011) is a 42 item measure which asks
individuals to respond on a 5-point Likert scale (1 = Definitely False; 5 = Definitely
True) the degree to which they use a rational (12 items) or experiential processing
style (30 items). The experiential scale consists of 3 sub-scales (10 items each) to
measure its different aspects: (1) intuitive; (2) emotionality and; (3) imagination. The
intuitive facet is designed to assess the ability and engagement of an individual’s use
of intuitive judgements (e.g. “I often go by my instincts when deciding on a course of
action”.) The emotionality sub-scale assesses the intensity, frequency, duration, and
favourable attitude towards strong affect (e.g. “My emotions don’t make much
difference in my life”; reversed scored). The imagination sub-scales assesses the
engagement in, and appreciation of, imagination, aesthetic productions, and imagery
(e.g. “Sometimes I like to just sit back and watch things happen.”) Higher scores
indicate a higher preference for processing on that particular dimension.
194
Constructive Thinking Inventory (CTI; Epstein, 2001). The constructive
and destructive components of the experiential system were assessed by the 108-item
constructive thinking inventory (CTI, Epstein, 2001). The constructive components
of the experiential system include global constructive thinking, emotional coping,
and behavioural coping. The destructive components are personal superstitious
thinking, categorical thinking, esoteric thinking, and naïve-optimism (Epstein, 2001).
The reader is referred to Table 1.3 presented in Chapter 1 for main-scale and sub-
scale definitions. People respond on a 5-point Likert scale how much a statement is
true of them (1 = Definitely False; 5 = Definitely True). Raw scores are entered into
an electronic scoring programme (Epstein & PAR Staff, 2008) and converted to T-
Scores based upon gender norms.
Measures - Follower
Implicit Leadership Theory Questionnaire (Offerman et al., 1994) –
adapted. The current study used the Implicit Leadership Theory Questionnaire
utilised in the previous studies. Individuals were asked to rate on a 10-point Likert
scale (1 = Not at all Characteristic; 10 = Extremely Characteristic) the degree to
which the traits were characteristic of “their ideal leader in a business setting”.
However, for the reasons described in the introduction, only the 22 items that loaded
on the five IILTs dimensions (sensitivity, tyranny, intelligence, attractiveness and
dedication; see Table 5.1) were used. Higher scores are assumed to indicate a
preference for a particular leader trait/behaviour.
195
Multi-dimensional Leader-Member Exchange Questionnaire (LMX-
MDM; Liden & Maslyn, 1998). The Multidimensional Leader-Member Exchange
Questionnaire (LMX-MDM; Liden & Maslyn, 1998; see Appendix D) is an
extensively used measure of LMX. Followers are asked to respond on a 7-point
Likert scale (1 = Strongly Disagree, 7 = Strongly Agree) on how much they agree a
statement reflects their perception of four separate, but interrelated, dimensions in
the exchange relationship with their leader: (1) affect; (2) loyalty; (3) contribution
and; (4) professional respect. Affect refers to the affection felt for a leader or
follower based primarily upon interpersonal attraction as opposed to professional
values. An example item is “I like my supervisor very much as a person”. Loyalty
refers to the belief that the other member of the dyad will express public support for
them. An example item is “My supervisor would come to my defence if I were
attacked by others”. Contribution refers to the perceptions held by the individual of
the support and involvement they give to the relationship beyond their contractual
obligations. An example item is “I do work for my supervisor that goes beyond what
is specified in my job description”. Professional respect refers to the perceived
reputation of the other member of the dyad. An example item is “I admire my
supervisor’s professional skills”. All sub-scales contain 3 items. An overall LMX
score was calculated by taking the average of all 12 items. Higher scores indicate
higher quality relationships. It should be noted that the name of a follower’s actual
leader was presented as opposed to the generic word of “supervisor”.
196
Procedure & Analytical Approach
The same two-stage procedure used in the previous two studies (Chapters 3
and 4) was used to help control for the effects of common-method bias (see
Podsakoff et al., 2003). In stage one, leaders completed the REIm (Norris & Epstein,
2011) while followers completed the adapted Implicit Leadership Theory
Questionnaire (Offerman et al., 1994). Both also completed some demographic
items. Leaders and followers who completed stage one were invited two days later to
complete stage two via email. In stage two, leaders completed the CTI (Epstein,
2001) while followers completed the LMX questionnaire (Liden & Maslyn, 1998).
The average time for leaders between the completion of stage one and stage two was
4.65 days (SD = 3.11, range 2-14). The average time for followers between the
completion of stage one and stage two was 3.98 days (SD = 2.65, range 2-14). Both
leaders and followers could choose to receive written feedback based upon their
responses to the REIm (Norris & Epstein, 2011) and CTI (Epstein, 2001). In
addition, in instances where at least 5 followers rated a leader, the leader could also
choose to receive leadership feedback based upon their followers’ responses to the
LMX-MDM (Liden & Maslyn, 1998).
Despite recognising that the organisational behaviour occurs within a
complex and dynamic organisational system, a key issue in the academic literature is
a general failure to account for the multi-level dynamics of these systems (Hitt,
Beamish, Jackson, & Mathieu, 2007). Organisations are typically hierarchical in
nature with individuals being nested in work-groups, which in turn are nested in
departments and so on (Hofmann & Gavin 1998)1. It is important, therefore, to
recognise and account for how these levels influence the dependent variable(s).
1 For simplicity, the current study chooses to examine only two levels (i.e. follower and leader).
197
Specifically, the extent that individuals (i.e. followers) within a group (i.e. work
team) share common experiences (i.e. with the same leader), we would expect their
scores on the outcome variable (i.e. LMX) to be correlated across members of the
group, which violates the independence assumption of many statistical models
(Bauer, Preacher, & Gil, 2006; Hofmann, 1997). Therefore typical analytical
assessment techniques (e.g. regression, path-analysis) are likely to result in biassed
model estimates (Geiser, Bishop, Lockhart, Shiffman, & Grenard, 2013).
Specifically, standard errors of the model parameters are likely to be underestimated
due to an overestimation of effective sample size caused due to clustering of the
data. This increases the chance of Type I errors (Luke, 2004). HLM7 is a statistical
modelling programme capable of appropriately analysing cross-sectional, multi-level
data (Raudenbush, Bryk, Cheong, Congdon, & Toit, 2011).
Results
For each manager, there was an average of 4.08 followers reporting to them.
As data was collected from multiple sources (i.e. managers and followers) and across
two stages separated by at least 2 days, common method bias was not assumed to be
an issue (see discussion by Podsakoff et al., 2003). Any data point that was more
than 1.5 times the Interquartile range (IQR) above or below the first or above the
third quartile was classified as an outlier (see Tuckey, 1977). Due to the small
sample size, outliers were truncated to the nearest highest/lowest non-outlier score
(see Osborne & Overbay, 2004). Thus, statistical power is retained as no individuals
are dropped in the analyses, but distribution issues are reduced. A total of 6 outliers
across all scales were recoded with a mean of .16 items per factor. Inspection of the
198
histograms and Q-Q plots indicated some minor normality issues but the inferential
statistics used are generally quite robust against this (Tabachnick & Fidell, 2007).
Table 5.2 presents means, standard deviations, and scale reliabilities for all leader
and follower variables used in the current study. All main scales showed acceptable
to excellent levels of reliability (range .74 to .92; Kline, 2000). While some of the
correlations on the CTI were very high (e.g., emotional coping and behavioural
coping r = .88), due to licencing restrictions, it was not possible to conduct a
confirmatory factor analysis to assess whether the items produced the expected
factor structures. Based on previous research (see Epstein, 2001; Epstein & Meier,
1989), it is assumed for current purposes that these factors can be treated as
theoretically separate.
Hypothesis Testing
The first step in the analysis was to determine if there were systematic
between-cluster variances in the dependent variable (i.e., follower LMX) by
calculating an inter-class correlation (ICC; Luke, 2004). An ICC ranges from 0 to
1.00 and is a measure of the degree of dependence of observations (Geiser et al.,
2013). For current purposes, 0 would indicate that LMX ratings are independent of
the leader – an unlikely scenario given the nature of the study. In contrast, higher
ICCs would indicate LMX ratings are dependent on the leader. An alternative
interpretation is that the ICC is the expected correlation between LMX scores for
two individuals who report to the same leader (Geiser et al., 2013). The ICC
indicated that 40.2% of the variance in follower LMX was between groups. While
there are no firm rules around the degree to which to interpret this figure, Aguinis,
199
Gottfredson and Culpepper (2013) state that ICC’s as small as .10 may indicate level
two variables explain variance across groups. Given the observed ICC was above
this threshold, multi-level modelling was seen as suitable for assessing the
hypotheses.
A similar statistical process to that described by Harris, Harvey and Kacmar
(2011) was used. A single leader’s information processing style or constructive
thinking dimension was used as the predictor variable for multiple followers.
Consequently, there is no within-leader variance in the predictor variable (i.e. all
ICCs are 1.0). All variance is between supervisors. Each hypothesis test was
modelled separately. To test a specific hypothesis, three statistical steps were
undertaken. In the first step, the level two variable (i.e. relevant leader information
processing dimension or constructive thinking dimension) was entered to test main
effect hypotheses (i.e. hypotheses 1 to 6). In the second step, the proposed moderator
was entered. This was included as previous research suggests that individuals use
their ILTs to “fill in” the blanks when rating leaders (e.g. Bass & Avolio, 1989; Eden
& Leviatan, 1975; 2005; Lord et al., 1984; Rush et al., 1977; Schyns & Meindl,
2005; Uhl-Bien et al., 2014). Consequently, followers’ IILTs may also predict their
LMX score. In the third step, a cross-level interaction variable was formed by
multiplying the level two and proposed level one moderator variables. Significant
findings at this step would support the presence of a moderating effect (i.e.
hypotheses 1a to 6b). Grand centring was undertaken on all independent and
moderator variables (see Aiken & West, 1991; Geiser et al., 2013; Hofmann &
Gavin, 1998). In each approach, R2 is interpreted as the proportional reduction in
prediction error over a comparison model (Luke, 2004). Specifically, it is the
percentage change in the variance in the residuals between the tested and comparison
200
model on the assumption that better models should decrease the variance in the
residuals. Where interactions were flagged as significant, simple slopes (as
recommended by Muthen, 2015) were plotted at both one standard deviation above
and below the mean of the proposed moderator and independent variable to
investigate whether the moderation effect was in the expected direction (see Aiken &
West, 1991; Schubert & Jacoby, 2004; Stone & Hollenbeck, 1989).
Hypotheses Related to Leaders’ Information Processing Dimensions
The results of the HLM analysis investigating hypotheses 1 (main effect of
leader rational processing) and 1a (moderating effect of follower intelligence IILT)
are presented in Table 5.3. As can be seen in step one, rational processing was not
significantly related to follower LMX. Consequently, hypothesis 1 was not
supported. Step two demonstrates that follower intelligence IILT was not
significantly related to their LMX rating. Step three also demonstrates that there was
no significant interaction between leader rational processing and follower
intelligence. Therefore, hypothesis 1a was not supported.
201
Table 5.2.
Descriptive statistics, inter-correlations and scale reliabilities for all manager (n=26) and follower (n=106%) main variables used in the study. Mean SD 1 2 3 4 5 6 7 8 9 10 11
Manager Variables
REIm
1. Rational 4.03 .48 .84 .10 .46** .50** -.65** .46** .03 .05 .14 .16 .24*
2. Experiential 3.59 .36
.74 -.05 .23* -.29** .55** -.08 -.02 .04 -.05 .20*
Constructive Thinking
3. Emotional Coping 48.15 14.67
N/A^ .88** -.77** .35** -.06 .31** .26* .27** .39**
4. Behavioural Coping 47.58 9.45
N/A^ -.78** .51** -.08 .21* .24* .17 .38**
5. Categorical Thinking 37.62 8.48
N/A^ -.50** .07 -.20* -.29** -.23* -.40**
6. Naïve-Optimism 38.65 7.49
N/A^ -.06 .11 .01 .09 .41**
Follower Variables
IILT%
7. Sensitivity 7.55 1.49
.85 .38** .31** .45** .17
8. Intelligence 7.52 1.75
.89 .52** .68** .16
9. Attractiveness 5.35 2.43
.89 .61** .05
10. Dedication 8.42 1.19
.89 .20
LMX
11. Overall 5.59 0.94
.92 %Note that descriptive statistics for the IILT dimensions are based upon responses from 100 followers rather than 106.
^Reliabilities cannot be calculated due to the Constructive Thinking Inventory (Epstein, 2001) being a controlled test. It is therefore assumed that
these scales are reliable.
202
Table 5.3.
Hypotheses 1 and 1a - Hierarchical linear modelling results predicting followers’
LMX with leaders’ rational processing and followers’ intelligence as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Rational (A) .151 .159 -.983
Moderator
Intelligence (B) .029 -.636
Interaction Term
A×B .157
Level 1 variance of residuals .481 .429 .431 .441
Variance change N/A .052 -.002 -.010
ΔR2 N/A .108 -.005 -.023
^p<.10; *p<.05; **p<.01
The results of the HLM analysis investigating hypotheses 2 (main effect of
leader experiential processing) and 2a (moderating effect of follower sensitivity
IILT) are presented in Table 5.4. As can be seen, in step one, experiential processing
was not significantly related to follower LMX. Consequently, hypothesis 2 was not
supported. Step two suggested that follower sensitivity IILT was not significantly
related to their LMX rating although it did approach significance. Step three also
demonstrates no significant interaction between leader experiential processing and
follower sensitivity. Therefore, hypothesis 2a was not supported.
Table 5.4.
Hypotheses 2 and 2a - Hierarchical linear modelling results predicting followers’
LMX with leaders’ experiential processing and follower sensitivity as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Experiential (A) .243 .337 -.399
Moderator
Sensitivity (B) .112^ -.247
Interaction Term
A×B .098
Level 1 variance of residuals .481 .428 .400 .396
Variance change N/A .053 .028 .004
ΔR2 N/A .110 .065 .010
^p<.10; *p<.05; **p<.01
203
The lack of significance between experiential processing and LMX was
surprising given the integral links between the experiential system and the
development of positive relationships (e.g. Norris & Epstein, 2011; Pacini &
Epstein, 1999a). As previously described, imagination is an important factor in
empathy (Katz, 1963; Yukl, 1998) which, in turn, is key to developing effective
relationships (Kellett et al., 2002; Rudebeck, 2002). Imagination is a sub-dimension
of experiential processing. Therefore, an exploratory analysis was undertaken using
the imagination sub-scale (M = 3.60, SD = .64, reliability = .76). This analysis is
presented in Table 5.5. As can be seen, imagination emerged as a significant positive
predictor of follower LMX. Therefore, some partial support was found for
hypothesis 2. There was also a significant interaction between leaders’ imagination
and followers’ sensitivity IILT. To explore this, simple slopes analysis was
undertaken and is presented in Figure 5.2. It was observed that the slope at 1SD
beneath the mean of the moderator was not significant (t = .735, p = .464). The slope
at 1SD above the mean though was significant (t = 3.801, p < .001). This provides
partial support for hypothesis 2a.
Table 5.5.
Exploratory Hierarchical linear modelling for interaction between leaders’
imagination and followers’ sensitivity for predicting followers' LMX Baseline Step 1 Step 2 Step 3
Independent Variable
Imagination (A) .392* .460** -.800
Moderator
Sensitivity (B) .122* -.538^
Interaction Term
A×B .163*
Level 1 variance of residuals .481 .434 .406 .383
Variance change N/A .047 .028 .023
ΔR2 N/A .098 .065 .057
^p<.10; *p<.05; **p<.01
204
Figure 5.2. Moderating effect of follower sensitivity IILT (1±SD) on the relationship
between leader imagination (1±SD) and follower LMX.
Hypotheses Related to Leaders’ Constructive Thinking Dimensions
The results of the HLM analysis investigating hypotheses 3 (main effect of
leader emotional coping) and 3a (moderating effect of follower sensitivity IILT) are
presented in Table 5.6. As can be seen, in step one, emotional coping was positively
related to follower LMX and was significant. Consequently, hypothesis 3 was
supported. Step two demonstrated that follower sensitivity IILT was positively
related to their LMX rating and was significant. Step three also demonstrates that
there was a significant interaction between leader emotional coping and follower
sensitivity. To determine support for the interaction hypothesis, simple slopes
analysis was undertaken and is presented in Figure 5.3. It was observed that the slope
at 1SD beneath the mean of the moderator was not significant (t = .600, p = .550).
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Imagination
Low Sensitivity High Sensitivity
205
The slope at 1SD above the mean though was significant (t = 5.948, p < .001). This
provides support for hypothesis 3a.
Table 5.6.
Hypotheses 3 and 3a - Hierarchical linear modelling results predicting followers’
LMX with leaders’ emotional coping and followers’ sensitivity as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Emotional Coping (A) .027** .029** .024**
Moderator
Sensitivity (B) .121* -.596*
Interaction Term
A×B .012**
Level 1 variance of residuals .481 .449 .417 .384
Variance change N/A .032 .032 .033
ΔR2 N/A .067 .071 .079
^p<.10; *p<.05; **p<.01
Figure 5.3. Moderating effect of follower sensitivity IILT (1±SD) on the relationship
between leader emotional coping (1±SD) and follower LMX.
The results of the HLM analysis investigating hypotheses 4 (main effect of
leader behavioural coping) and 4a (moderating effect of follower dedication IILT)
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Emotional Coping
Low Sensitivity High Sensitivity
206
are presented in Table 5.7. As can be seen, in step one, behavioural coping was
positively related to follower LMX and was significant. Consequently, hypothesis 4
was supported. Step two demonstrates that follower dedication was not significantly
related to their LMX rating. Step three though demonstrates a significant interaction
between leader behavioural coping and follower dedication. To determine support
for the interaction hypothesis, simple slopes analysis was undertaken and is
presented in Figure 5.4. It was observed that the slope at 1SD beneath the mean of
the moderator was not significant (t = .672, p = .503). The slope at 1SD above the
mean though was significant (t = 5.848, p < .001). This provides support for
hypothesis 4a.
Table 5.7.
Hypotheses 4 and 4a - Hierarchical linear modelling results predicting followers’
LMX with leaders’ behavioural coping and followers’ dedication as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Behavioural Coping (A) .031** .030** -.240**
Moderator
Dedication (B) .133 -1.608**
Interaction Term
A×B .033**
Level 1 variance of residuals .481 .439 .407 .315
Variance change N/A .042 .032 .092
ΔR2 N/A .087 .073 .226
^p<.10; *p<.05; **p<.01
207
Figure 5.4. Moderating effect of follower dedication IILT (1±SD) on the relationship
between leader behavioural coping (1±SD) and follower LMX.
The results of the HLM analysis investigating hypothesis 4b (moderating
effect of follower sensitivity ideal ILT on the proposed relationship between leader
behavioural coping and follower LMX) is presented in Table 5.8. As already shown
above, step one revealed a significant, positive relationship between leader
behavioural coping and follower LMX. Step two suggested that follower sensitivity
IILT was positively and significantly related to their LMX rating. Step three also
demonstrates that there was a significant interaction between leader behavioural
coping and follower sensitivity. To determine support for the interaction hypothesis,
simple slopes analysis was undertaken and is presented in Figure 5.5. It was
observed that the slope at 1SD beneath the mean of the moderator was not significant
(t = .427, p = .671). The slope at 1SD above the mean though was significant (t =
5.529, p < .001). This provides support for hypothesis 4b.
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Behavioural Coping
Low Dedication High Dedication
208
Table 5.8.
Hypothesis 4b - Hierarchical linear modelling results predicting followers’ LMX
with leaders’ behavioural coping and followers’ sensitivity as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Behavioural Coping (A) .031** .035** -.147**
Moderator
Sensitivity (B) .120* -1.200**
Interaction Term
A×B .023**
Level 1 variance of residuals .481 .439 .410 .344
Variance change N/A .042 .029 .066
ΔR2 N/A .087 .066 .161
^p<.10; *p<.05; **p<.01
Figure 5.5. Moderating effect of follower sensitivity IILT (1±SD) on the
relationship between leader behavioural coping (1±SD) and follower LMX.
The results of the HLM analysis investigating hypotheses 5 (main effect of
leader categorical thinking) and 5a (moderating effect of follower sensitivity IILT)
are presented in Table 5.9. As can be seen, in step one, categorical thinking was
negatively related to follower LMX and was significant. Consequently, hypothesis 5
was supported. Step two also shows that follower sensitivity IILT was positively and
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Behavioural Coping
Low Sensitivity High Sensitivity
209
significantly related to their LMX rating. Step three further demonstrates a
significant interaction between leader categorical thinking and follower sensitivity.
To determine support for the interaction hypothesis, simple slopes analysis was
undertaken and is presented in Figure 5.6. It was observed that the slope at 1SD
beneath the mean of the moderator was not significant (t = .402, p = .688). The slope
at 1SD above the mean though was significant (t = 5.222, p < .001). This provides
support for hypothesis 5a.
Table 5.9.
Hypotheses 5 and 5a - Hierarchical linear modelling results predicting followers’
LMX with leaders’ categorical thinking and followers’ sensitivity as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Categorical Thinking (A) -.041* -.044** .148*
Moderator
Sensitivity (B) .120* .862**
Interaction Term
A×B -.024**
Level 1 variance of residuals .481 .444 .414 .361
Variance change N/A .037 .030 .053
ΔR2 N/A .077 .068 .128
^p<.10; *p<.05; **p<.01
The results of the HLM analysis investigating hypotheses 6 (main effect of
leader naïve-optimism) and 6a (moderating effect of follower attractiveness IILT)
are presented in Table 5.10. As can be seen, in step one, naïve-optimism was
positively related to follower LMX but only approached significance. Consequently,
hypothesis 6 was rejected. Step two also suggests that follower attractiveness IILT
did not significantly contribute to predicting their LMX rating. However, step three
demonstrates a significant interaction between leader naïve-optimism and follower
attractiveness. To determine support for the interaction hypothesis, simple slopes
210
Figure 5.6. Moderating effect of follower sensitivity IILT (1±SD) on the
relationship between leader categorical thinking (1±SD) and follower LMX.
analysis was undertaken and is presented in Figure 5.7. It was observed that the slope
at 1SD beneath the mean of the moderator was not significant (t = .726, p = .470).
The slope at 1SD above the mean though was significant (t = 4.405, p < .001). This
provides support for hypothesis 6a.
Table 5.10.
Hypotheses 6 and 6a - Hierarchical linear modelling results predicting followers’
LMX with leaders’ naïve-optimism and followers’ attractiveness as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Naïve-Optimism (A) .039^ .039^ -.093*
Moderator
Attractiveness (B) .000 -.928**
Interaction Term
A×B .022**
Level 1 variance of residuals .481 .439 .440 .389
Variance change N/A .042 -.001 .051
ΔR2 N/A .087 -.002 .116
^p<.10; *p<.05; **p<.01
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Categorical Thinking
Low Sensitivity High Sensitivity
211
Figure 5.7. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader naïve-optimism (1±SD) and follower LMX.
The results of the HLM analysis investigating hypotheses 6b (moderating
effect of follower dedication IILT on the relationship between leader naïve-optimism
and follower LMX) are presented in Table 5.11. As demonstrated previously, in step
one, naïve-optimism was positively related to follower LMX but only approached
significance. Step two also demonstrates that follower dedication IILT did not
significantly contribute to their LMX rating. However, step three demonstrates a
significant interaction between leader naïve-optimism and follower dedication. To
determine support for the interaction hypothesis, simple slopes analysis was
undertaken and is presented in Figure 5.8. It was observed that the slope at 1SD
beneath the mean of the moderator was not significant (t = .585, p = .560). The slope
at 1SD above the mean though was significant (t = 4.015, p < .001). This provides
support for hypothesis 6b.
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Naive Optimism
Low Attractiveness High Attractiveness
212
Table 5.11.
Hypothesis 6b - Hierarchical linear modelling results predicting followers’ LMX
with leaders’ naïve-optimism and followers’ dedication as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Naïve-Optimism (A) .039^ .037 -.171*
Moderator
Dedication (B) .138 -.940*
Interaction Term
A×B .026**
Level 1 variance of residuals .481 .439 .409 .396
Variance change N/A .042 .030 .013
ΔR2 N/A .087 .068 .032
^p<.10; *p<.05; **p<.01
Figure 5.8. Moderating effect of follower dedication IILT (1±SD) on the relationship
between leader naïve-optimism (1±SD) and follower LMX.
Discussion
Based on the recognition that leaders and followers do not usually experience
their dyadic relationship in the same way (see Gerstner & Day, 1997; Sin et al.,
2009; van Gils et al., 2010; Wilson et al., 2010), the current chapter began by
introducing a general theoretical framework which integrated CEST (Epstein, 1998,
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Naive Optimism
Low Dedication High Dedication
213
2003), ILTs (Lord & Maher, 1993) and IFTs (Sy, 2010) to help explain leader and
follower LMX ratings. At the most basic, information processing style and
constructive thinking were theorised to play influential roles in the interactions
between leaders and followers. This is due to these constructs spreading widely
throughout an individual’s self-concept and thus influencing a wide range of
people’s behaviours across a variety of contexts and situations (Epstein, 2014). This,
therefore, should include interactions between leaders and followers.
Given the different roles of leaders and followers, the general model also
drew upon existing ILT and IFT theory (see Lord & Maher, 1993; Sy, 2010) to
propose that each member of the leader-follower dyad would evaluate the other’s
behaviour and actions from different perspectives. Specifically, it was proposed that
followers and leaders would draw upon their IILT or IIFT respectively when making
judgements and evaluations of the dyadic partner as these would take account some
of the observer’s personal preferences and tendencies (i.e. their information
processing style and constructive thinking). Drawing on a theoretical basis of person-
environment fit (see Kristof-Brown, 1996; Kristof-Brown et al., 2005), it was
generally proposed that leaders’ or followers’ who have information processing
styles or constructive thinking traits which are congruent with their dyadic partners’
IILT or IIFT should act to increase this partners’ LMX rating.
The remainder of chapter explored one-half of this proposal. Specifically,
how leaders’ information processing style and level of constructive thinking relate to
followers’ perceptions of LMX and whether followers’ IILT moderate this. In
general, while some hypotheses did not emerge, the results provide support for the
moderation model presented in Figure 5.1a. Theoretical and practical implications of
this part of the model are discussed below.
214
Leaders’ information processing and followers’ LMX
It was surprising that no significant positive relationship emerged between
leader rational processing and follower LMX, although it was in the expected
direction. Furthermore, intelligence did not emerge as a moderator between leaders’
rational processing and followers’ perceptions of the quality of the leader-follower
relationship. This appears to contradict research conducted in the transformational
leadership domain which has found that a positive relationship between leaders’
rational processing and their own ratings of transformational leadership (Cerni et al.,
2008; King, 2012). However, the findings do align with more recent research in this
area which demonstrates that this relationship does not emerge when
transformational leadership is rated from followers’ perspectives (Simoens, 2014).
One potential explanation is that rational processing is not seen or
experienced in the same way by followers as it is leaders. Indeed, other ratings of an
individual’s rational processing tend to be lower than self-ratings (Norris & Epstein,
2011) suggesting that rational processing in leaders does not necessarily lend itself to
‘intelligent’ behaviours when viewed from the perspective of followers. For
example, the more delayed and considered approach of highly rational leaders may
be interpreted as confusion or indifference to an issue from the perspective of
followers. Such behaviours, therefore, may not necessarily lead to follower
attributions that such leaders are intelligent. Instead, the perception of ‘intelligent’
leader behaviours by followers may be more aligned with the tendencies described in
global constructive thinking – that of flexible thinkers who can adapt to meet the
needs of the situation (Epstein, 2001; 2014). From the perspective of followers,
therefore, it may be the processing occurring in leaders’ experiential systems which
is more important due to its impact on a wide range of leader behaviours. This is
215
aligned with the notion in CEST that the experiential system processes information
much more rapidly than the rational system and, thus, tends to influence initial
behaviour (Epstein, 2003; 2014), which would include that occurring in leader-
follower interactions.
For leaders, though, it appears that rational processing is important. Rational
system is logical and deals more in abstraction than the experiential system (Norris
& Epstein, 2011). It is also a delayed system that acts to correct initial errors made
by the experiential system (Pacini & Epstein, 1999a; 1999b; Epstein, 2003). High
rational processors tend to take the time to consider things. From the perspective of
leaders, such processing most likely helps them to deal with issues and problems
they experience in the work-place. Consequently, this would explain the positive
relationship which emerges between rational processing and self-rating of
transformational leadership (see Cerni et al., 2008; King, 2012).
Leaders’ experiential processing did not positively predict followers’ LMX.
In addition, followers’ sensitivity IILT did not moderate this relationship. Given its
fundamental associations with the development of positive interpersonal
relationships (Norris & Epstein, 2011; Pacini & Epstein, 1999a), this was surprising.
Exploratory analysis, though, revealed that higher levels of imagination did tend to
result in followers reporting higher quality relationships with their leaders.
Furthermore, this relationship was stronger for followers who endorsed higher levels
of sensitivity in their ideal leader. Imagination, therefore, appears to, at least partly,
contribute to leader behaviours that are congruent with followers who seek
sensitivity in their ideal leader. This is likely due to the fact that imagination is an
important element of empathy as it helps an individual to take the perspective of
another so that they can understand how they may be affected (Gaesser, 2013).
216
These findings are generally aligned with other research, which identifies empathy as
an important factor in effective leadership (e.g. see Bass, 1985; Walumbwa, Avolio,
Gardner, Wernsing, & Peterson, 2008).
The sensitivity IILT factor contains items (see Table 5.1) which are similar to
descriptions of empathy. From the perspective of followers who seeks high
sensitivity in their ideal leader, the ability of leaders to empathise with their needs
and perspectives would be very important. For example, such followers would likely
appreciate leaders who are sensitive to their work-loads, their personal needs,
emotional issues and career goals. This is not to say that those who generally endorse
low levels of sensitivity in their ideal leader do not also seek empathy. Indeed, as
observed in the current study most people endorse relatively high levels of
sensitivity in their leaders. However, the results demonstrate that LMX ratings are
not just dependent on how leaders act but also on how followers interpret these
behaviours.
Leaders’ constructive thinking and followers’ LMX
In a similar vein to the discussion above around leader imagination,
emotional coping in leaders also appears to help them be more open to their
followers’ perspectives and needs. Specifically, high levels of emotional coping
appears to help leaders focus on the validity of followers’ criticism, input or
perspective rather getting caught up in the impact of such negative feedback on their
own self-esteem resulting in attempts to defend themselves (e.g. the leader feels their
competence is “attacked” resulting in psychological and/or behavioural defences
being marshalled directly or indirectly towards the follower; e.g. see Grant et al.,
217
2009; Ilgen et al., 1979; Morrison & Milliken, 2000; Kluger & DeNisi, 1996;
Menon, 2001). This aligns with findings in Chapter 5 in which leaders with high
levels of emotional coping tended to endorse lower levels of conformity in their ideal
followers. It was suggested that leaders with good emotional coping are generally
more cognitively and emotionally open to those followers who are willing to speak
up and question the status quo (see also Burris, 2012; Chaleff, 2009).
The openness and non-defensiveness of leaders with high levels of emotional
coping appear to be particularly congruent with followers who endorse high
sensitivity in their ideal leader. Specifically, it appears to buffer leaders from
engaging in cognitive/behavioural defence mechanisms which in turn provides an
increased internal cognitive/emotional space for the leader to focus their attention on
the follower (e.g. open to alternative perspectives and needs of the follower even if
they may not be aligned with that of the leader; e.g., Bushman & Baumeister, 1998;
Rubin & Hewstone, 1998). This is a more indirect approach to empathy than that
described for imagination. However, the resultant behaviours would tend to be more
valued (i.e. more congruent) with the traits and behaviours that followers with high
sensitivity seek in leaders compared to those who endorse low levels of sensitivity.
As expected, leaders’ behavioural coping positively predicted followers’
LMX. This finding can be partly interpreted through the broaden-and-build theory of
positive emotions (Fredrickson, 1998; 2001; 2003; Fredrickson & Branigan, 2005).
According to this theory, positive emotions help broaden people’s awareness,
thinking and actions which result in increased personal and social resources
(Fredrickson, 2003). These resources act as a repertoire of skills and knowledge from
which they can adapt to new contexts in the future. In other words, positive emotions
contribute to a “‘upwards spiral’ towards optimal functioning and enhanced
218
emotional wellbeing” (Fredrickson, 2003, p.169). While the resources which can be
generated are varied (e.g. see Isen & Daubman, 1984; Isen & Means, 1983), some of
the most important for current purposes include a tendency to be more open to
information (Estrada, Isen, & Young, 1997), to take account of others’ needs to a
greater degree (Waugh & Fredrickson, 2006), and being more open to diverse ideas,
options (Kahn & Isen, 1993) and critical feedback (Raghunathan & Trope, 2002).
Common to each of these is an increased attention and openness to others needs’ and
perspectives which may be contrary to the individuals own. In the context of the
current study, therefore, it appears that the leaders with good behavioural coping
helps them be more open to the needs of, and input from, followers. The result is an
increased opportunity to address these needs and input. Given such tendencies, it is
not surprising, therefore, that leaders with high behavioural coping skills were
particularly valued by followers who seek high levels of sensitivity in their ideal
leader. This is likely due to the increased value that such followers place on leader
empathic behaviours.
As predicted the relationship between leaders’ behavioural coping and
followers’ LMX was also moderated by the followers’ dedication IILT. High levels
of behavioural coping in leaders should help support them engage in planning and
taking quick, considered and effective action to overcome workplace problems and
issues (Epstein, 2001). Leaders with high behavioural coping would also tend to be
less preoccupied with overcoming problems and instead ‘just get on with it’ (i.e.
positive thinking; see Epstein, 2001). Furthermore, broaden-and-build theory and
research would suggest that leaders with good behavioural coping would hold an
increased pool of knowledge and skills to apply to novel workplace problems (see
Estrada et al., 1997; Fredrickson, 1998; 2001; 2003; Fredrickson & Branigan, 2005;
219
Kahn & Isen, 1993). Such proactive behaviour in a leader, in addition to an increased
skill and knowledge base to draw upon, is likely to be highly valued by followers
who endorse high levels of dedication (e.g. dedicated; enthusiastic; hard-working),
which consequently increases positive beliefs about the leader and their relationship
with them. In contrast, leaders who are not motivated or hard-working are likely to
frustrate such followers resulting in a more negative perception of the leader and the
relationship.
Unsurprisingly, the cognitive and behavioural tendencies of leaders with high
categorical thinking do not tend to facilitate the development of positive and
supportive leader-follower relationships. This is further undermined for followers
who endorse high levels of sensitivity in their ideal leader. Specifically, the negative
impact of leaders’ categorical thinking appears to be stronger for followers who
endorse high levels of sensitivity in their ideal leader. Categorical thinkers tend to
see others who hold different perspectives to themselves as in error, be somewhat
less trusting of others and more judgmental (Epstein, 1998; 2001). They are also
quick to experience anger and annoyance when things do not go according to their
expectations (Epstein, 1998). Being closed to alternative perspectives, judgmental
and quick to anger is unlikely to facilitate a psychologically safe work environment
for followers to engage in proactive communication behaviours (e.g. followers
express constructive suggestions for change; Botero & Dyne, 2009; LePine & van
Dyne, 2001). This is because these followers may feel threatened by the reactions
(e.g. anger, frustration) of leaders with high levels of categorical thinking to their
input and/or feedback (see Chan & Treacy, 1996; May et al., 2004; Nembhard &
Edmondson, 2006; Rogers, 1959; Walumbwa & Schaubroeck, 2009). Alternatively,
followers may perceive that providing input and/or feedback is simply a waste of
220
time and energy as it will not be considered or acted upon (e.g. see Ellis & Dyne,
2009). In both cases, leader-follower communication is undermined.
In support, communication issues have consistently been shown to
undermine the development of positive and effective working relationships and
environments (e.g. see Garon & Ringl, 2004; Gwartney-Gibbs & Lach, 1994;
Keashly, Trott, & MacLean, 1994). In addition, leaders with higher levels of
categorical thinking tend to seek higher levels of conformity in followers (see
Chapter 4). High levels of conformity beliefs are associated with an increased
likelihood that leaders will resort to using more directive control techniques (e.g.
close supervision and extrinsic rewards) inherent to their more powerful and
authoritative position to gain compliance (see Finman, 1973; Neuliep, 1987; 1996;
McGregor, 1960). Such approaches tend to be associated with lower levels of LMX
(Şahin, 2012). Taken as a whole, it appears that categorical thinking, as defined in
CEST (Epstein, 2001), contributes to leaders having negative beliefs about the nature
of followers which results in more judgmental, less open and less trusting
relationships as seen from followers’ perspectives. For most followers, these would
be negative experiences, but for those who endorse high levels of sensitivity in their
ideal leader, these behaviours appear to be experienced particularly negatively.
The non-significant but positive main effect relationship between leaders’
naïve-optimism and followers’ LMX warrants some discussion. Exploratory
analyses (see Appendix F) suggest that this non-significance may be due to
differences in the nature of the naïve-optimism sub-scales when interpreted in the
context of the work-environment and work relationships. Over-optimism is about
unrealistic optimism and can result in individuals failing to acknowledge potential
problems and unpleasant realities (Collinson, 2012; Epstein, 1998; 2014).
221
Consequently, they are at an increased risk of failing to take action as they simply
expect that all issues will work out. For example, these leaders may underestimate
how much a project will cost or how long it will take to finish. Therefore, in the
context of the work-environment, over-optimism in leaders may tend to be seen
somewhat negatively from the perspective of followers. For example, feedback from
followers about issues or problems may fail to be recognised or even acknowledged.
However, stereotypical thinking and pollyanna-ish thinking are about
positive, albeit unrealistic and simplistic, beliefs. Such beliefs may help support the
development of leader-follower relationships as they help reinforce positive
expectations of how they and others should behave during their interactions. Indeed,
pollyanna-ish thinking is the sub-scale which shows the strongest relationships with
caring, agreeableness and positive affect (Epstein, 1983) all of which should support
positive and supportive interactions with followers. Consequently, the different
elements of naïve-optimism should be considered in isolation to each other, rather
than as a whole, when seeking to understand what is contributing to followers’
perceptions of LMX. This conclusion is aligned to Epstein (2014) who states that
“the positive and the unrealistic components of naïve-optimism cancel each other out
with respect to adjustment” (p.168).
While there was no main effect of naïve-optimism, it appears that naïve-
optimism in leaders is important for followers who endorse high levels of
attractiveness and dedication as expected. Naïve optimists appear to engage in
behaviours that are similar to those falling under certain areas of impression
management (e.g. appearing happy even when one feels sad). In particular, how they
appear to others. Consequently, these leaders are likely to control their visual
appearance and presentation such as their clothing. For followers who endorse high
222
levels of attractiveness, such tendencies of leaders are likely to be congruent with
followers who endorse high levels of attractiveness. For followers who endorse high
levels of dedication, it appears that the positivity and optimism of such leaders
appeal to their desire for their ideal leader to be motivated and enthusiastic.
Other comments and practical implications
Whilst not focused upon in the introduction, the current research also aligns
with other research that has demonstrated links between follower ILTs and LMX
(e.g. Bass & Avolio, 1989; Eden & Leviatan, 1975; 2005; Lord et al., 1984; Rush et
al., 1977; Schyns & Meindl, 2005; Uhl-Bien et al., 2014). Specifically, those who
endorsed sensitivity in their ideal leader tended to rate the relationship with the
leader more highly. Shondrick and Lord (2010) draw upon adaptive resonance theory
(see Grossberg, 2013) to describe an information processing approach whereby
resonance (i.e. where an external stimulus such as a leader is matched to a specific
cognitive category) creates difficulty in distinguishing between the ILT and the
actual leader’s behaviours. Specifically, when the leader category is made salient for
the follower, they actively processes and manipulate information about that leader
through the information and characteristics contained in their ILT. The result is
essentially that followers use their ILTs to “fill in” the blanks when rating leaders
(e.g. Bass & Avolio, 1989; Eden & Leviatan, 1975; 2005; Lord et al., 1984; Rush et
al., 1977; Schyns & Meindl, 2005; Uhl-Bien et al., 2014). In the current case
therefore, it appears that when the leader is cued in the mind of a follower, the
follower’s sensitivity IILT positivity contributes to LMX rating as the actual traits
and behaviours of the leader are distorted through the follower’s sensitivity IILT
223
(e.g. remembering sensitive leader behaviours as occurring when none have actually
occurred).
While broaden-and-build theory of positive emotions originally focused
primarily on how positive emotions help broaden attention (e.g. Fredrickson, 1998;
2001), more recent additions to this theoretical framework recognise that negative
emotions narrow one’s attention; to the extent to which negative emotions can lead
to downward spirals (Garland et al., 2010; Rathunde, 2000). Such downward spirals
result in attention turning towards the self and rigid, defensive behaviours being
marshalled (Garland et al., 2010). Some of the findings and discussion presented in
regards to leaders’ information processing and constructive thinking can be
summarised through this extended conceptualisation. Specifically, it can be
conceptualised that low levels of constructive thinking (i.e. emotional and
behavioural coping) and high levels of destructive thinking, in general1, narrow
leaders’ attentions away from their followers’ needs and perspectives. Instead,
leaders’ focuses move inwards as they seek to protect themselves from perceived
attacks on themselves and their capabilities (e.g. follower providing feedback and
input on a leader's ideas or direction) and/or the experience of negative vibes and
emotions (e.g. high categorical thinker acting in ways to remove their annoyance or
anger). From the perspective of Chaleff (2009), such leaders are not courageous to
the input and feedback of followers.
The current findings have important implications with regards to how
practitioners approach leadership. It is important to remember the distinction
1 It should be noted that naïve-optimism is more complex due to its inherent links with the experience
of positive affect yet less objectively conscientious behaviour (Epstein, 2001; 2014) which may
undermine positive leader-follower relationship development as described in-text. Therefore naïve-
optimism should be interpreted somewhat more cautiously.
224
between the terms ‘leader’ and ‘leadership’ as these have practical implications for
how we approach leadership in the workplace. Leader development is focused only
on leaders themselves with the aim of changing their cognition or behaviour to
match a specific leadership model (Day, 2001). In contrast, leadership development
is contextually based as it recognises the wider environment in which leadership
occurs including the follower, task and organisational culture. Most current
mainstream approaches are typically leader development initiatives and are not
actually leadership development as they are normally described or marketed (Day,
2001). Focusing on only one-half of the leadership equation fails to recognise that
followers have different perspectives of an ideal leader and that the same leader
behaviours may be interpreted and evaluated in very different ways by followers.
Indeed, LMX theory specifically states that followers are not a homogeneous set of
individuals who react and act in the same ways (Dansereau et al., 1975; Graen et al.,
1973). Practitioners, therefore, may benefit from adopting a true leadership
development approach, which specifically incorporates followers as well as leaders.
In addition to developing leaders’ information processing styles and constructive
thinking (see Cerni et al., 2010), this approach would bring followers’ attention to
their active role in leadership. Specifically, how their IILTs influence their
judgements of, and relationships with, leaders and the potential reasons for this (e.g.
see Chapter 3; see also Lord & Maher, 1993; Sy, 2010; Whiteley et al., 2012).
Bringing attention to followers’ IILT and the reasons why they may hold a
particular IILT may be beneficial as by Kruse and Sy (2011) demonstrating that IFTs
can be changed by manipulating an individual’s affective state. Theoretically, this
should also apply to the manipulation of IILTs. However, due to the more transient
nature of affect, these changes are likely to be somewhat short-lived. From the
225
perspective of CEST though, affect is an outcome of underlying cognitive processes
occurring in the experiential system (Epstein, 1998; 2003; 2014). Changes in these
underlying cognitive processes, therefore, should provide more stable and long-
lasting changes in followers ILTs. By targeting change to both followers’ IILT and
leader cognitive processes, the research would suggest that followers’ LMX can be
improved. This should result in a variety of positive individual, team, and
organisational outcomes (see meta-analysis by Dulebohn et al., 2012).
Limitations
As with all studies, consideration of limitations must be made. First, it is
important to acknowledge that an alternative interpretation could generally be
applied to the current findings. Specifically, as followers’ IILTs are related to their
information processing styles and constructive thinking (see Chapter 3), it is possible
that it is a congruence between followers’ and leaders’ information processing styles
and constructive thinking that is contributing to followers’ LMX rating rather than
the congruence between the leader’s information processing styles and constructive
thinking and the followers IILT. Indeed, research demonstrates that when leaders
and followers are congruent in terms of various personality factors, followers are
more likely to report higher LMX (see Schaubroeck & Lam, 2002; Strauss, Barrick,
& Connerley, 2001; Zhang, et al., 2012). To extend upon the current research,
therefore, incorporating the followers’ information processing style and constructive
thinking would help to tease these relationships apart. Three outcomes could be
expected from this. First, it is the congruence between leaders and followers’
information processing style and constructive thinking only, which contribute to
226
their LMX rating. Second, the interpretation applied to the current research (i.e.
congruence between the leader’s information processing style and constructive
thinking with the follower’s IILT. Third, and probably most likely, that both
congruence between leaders and followers’ information processing style and
constructive thinking, in addition to congruence between the leader’s information
processing style and constructive thinking with the follower’s IILT independently
contribute to the follower’s LMX rating.
Second, the somewhat small sample size may have reduced the power to find
significant relationships. Nevertheless, as the current study utilised quite a broad
sample of leaders and followers across a wide range of industries and jobs, the fact
that significant results were observed and were all in the expected directions
provides respectable evidence that the proposed general model is suitable for
understanding how personality and social roles influence working relationships.
Furthermore, the broad range of participants suggests the results are somewhat
generalisable to a wide range of organisational settings.
Third, the use of an explicit measure to assess an implicit construct has been
acknowledged as a potential issue in that the measure and the construct are not
aligned (Uhlmann et al., 2012). However, Epitropaki and colleagues (2013) note that
both explicit and implicit measures are useful for understanding the role of
leadership schemas in the workplace. Therefore, the findings of the current study still
are relevant and important. Future research though may wish to use an implicit
measure to further investigate whether similar relationships emerge.
227
CHAPTER 6
Followers’ Information Processing Style and Constructive Thinking as Leader
LMX Antecedents: The Moderating Influence of Leaders’ Implicit Ideal
Followership Theory
In the previous chapter, it was argued that the information processing style
and constructive thinking of leaders and followers are antecedents to followers’ and
leaders’ perceptions of the leader-follower relationship respectively. Drawing upon
implicit leadership theory (ILT; Eden & Leviathan, 1975; Lord & Maher, 1990;
1993) and implicit followership theory (IFT; Sy, 2010), it was further argued that
followers and leaders will draw upon their implicit ideal leadership (IILT) and
implicit ideal followership theories (IIFT) respectively when interpreting and
evaluating their dyadic partners’ behaviours. This was on the basis that these would,
in part, account for the observers’ characteristics and preferences – namely their
information processing style (Norris & Epstein, 2011) and constructive thinking
(Epstein, 2001).
Corresponding to these arguments, two general research models were
developed (see Figures 5.1a and b) to help explain why leaders and followers do not
typically experience their relationship in the same way (Gerstner & Day, 1997; Sin et
al., 2009). The previous chapter focused on testing the propositions of one of these
models. Specifically, how leaders’ information processing styles and constructive
thinking related to followers’ LMX perceptions and whether the latters’ IILT
moderated these relationships (see Figure 5.1a). It was found that leaders’
imagination and a variety of constructive thinking dimensions were related to
followers’ LMX. Furthermore, followers’ IILTs moderated these relationships. The
228
results were interpreted as demonstrating that follower LMX ratings are not only
dependent on leaders’ behaviours but also on how followers’ interpret these.
This chapter focuses on the second model (see Figure 5.1b). Specifically,
how followers’ information processing styles and constructive thinking relate to
leaders’ LMX and whether the latters’ IIFTs moderate this. As the theory presented
in the previous chapter also underpins the current chapter, the reader is referred to
the first half of the introduction of Chapter 5 for a more detailed overview. However,
as there has been much less focus on what resources followers provide to leaders
(Wilson et al., 2010), this chapter begins with a consideration of what resources
leaders may value in order to aid hypothesis development.
What Resources Can Leaders Receive From Followers?
The dyadic nature of the leader-follower relationship requires us to consider
what leader needs may be met (or not) from the resources that may (or may not) be
provided by followers. Unfortunately, as noted above, unlike resources that leaders
provide to followers, there has been much less theory and research on what followers
can provide to leaders (Wilson et al., 2010). According to Foa and Foa (1974; see
also Wilson et al., 2010) though there are six general categories of resources, which
can be exchanged in social relationships: (1) money; (2) goods; (3) services; (4)
status; (5) information; and (6) affiliation. Given that LMX is a social-exchange
based theory (see Blau, 1964; Graen & Scandura, 1987; Liden et al., 1997; Sparrowe
& Liden, 1997), these six general categories were seen as a suitable basis for
identifying what resources leaders may receive from followers and, therefore, served
to guide main effect hypothesis development.
229
As per the moderation theory presented in the previous chapter, we must also
consider how the leaders’ IIFT may serve to moderate the perceived value of these
resources (see discussion by Foa & Foa, 1974). The IIFT dimensions observed in
Chapter 4 were used to facilitate interpretation. These dimensions are shown in
Table 6.1 and include insubordination, committed, incompetence, conformity and
enthusiasm. These were conceptualised as potential moderators when developing the
hypotheses presented below.
Table 6.1.
IIFT Factors and Items from EFA conducted in Chapter 4 IIFT Factor Items in Factor
Insubordination Bad-Temper; Rude; Arrogant
Committed Goes above and beyond; Productive; Reliable; Loyal
Incompetence Slow; Uneducated; Inexperienced
Conformity Follows Trends; Easily Influenced
Enthusiasm Outgoing; Excited
Hypothesis development
Followers’ Information Processing and Leaders’ LMX
While leaders’ rational thinking was not related to followers’ LMX rating
(see Chapter 5), there is a theoretical reason to consider that followers’ rational
information processing will relate to leaders’ LMX. High rational processors are
deep thinkers, tend to develop clear, rational, and explainable reasons for decisions,
and enjoy thinking abstractly (Norris & Epstein, 2011; Pacini & Epstein, 1999a).
Correspondingly, high rational processors are more likely to adopt a rational
influencing style (Cerni et al., 2012). A rational processing style in followers,
therefore, should help support them present rational and logical arguments when
interacting with leaders. According to Oc and Bashshur (2013), the use of rational
230
arguments and evidence by followers is the most effective approach when attempting
to persuade leaders. This is due to the followers assumed closeness to important and
relevant information to the situation under focus. Therefore, followers’ rational
information processing should be positively related to leaders’ perceptions of
followers’ competence and thus improve the leader-follower relationship. In support,
use of rational persuasion tactics by followers is related to better performance
assessments (Higgins, Judge, & Ferris, 2003) and promotability (Thacker & Wayne,
1995). High rational processors also tend to be more problem-solution or action
focused (Epstein et al., 1996). From the perspective of leaders, such tendencies are
likely to be evaluated positively (i.e. valued) as it may help the leader move towards
their own goals (e.g. increase sales, more efficient processes). This suggests that
followers’ rational information processing would be positively related to leaders’
LMX ratings.
As per the moderation model presented in Figure 5.1b (see Chapter 5), this
relationship may be moderated by the degree to which leaders endorse conformity in
their ideal follower. The research undertaken in Chapter 4, and the wider literature,
strongly suggests that a sizeable minority endorse conformity as part of followers’
roles (see Grant et al., 2009; McGregor’s, 1960; Sy, 2010). Followers that like to
think about things, present logical and coherent arguments, and take a more active
role (i.e. a high rational processor), does not appear to be congruent with leaders who
endorse passive follower beliefs. Instead, such tendencies appear to be more
congruent with leaders who believe that their ideal follower should question the
status quo and be confident in standing up to them when required. Consequently, it
can be hypothesised that:
231
Hypothesis 1: a significant positive relationship will exist between followers’
rational processing and leaders’ LMX…
Hypothesis 1a: ...and this relationship will be stronger for leaders who prefer non-
conformity in their followers.
The experiential system is more strongly associated with the development
and maintenance of positive social relationships (Norris & Epstein, 2011; Pacini &
Epstein, 1999a). Empathy is an important aspect of developing relationships in all
contexts, including those at work, as it helps individuals to take another’s
perspective and re-experience their emotions (Katz, 1963; Yukl, 1998). Imagination,
a component of experiential processing (Norris & Epstein, 2011), is an important
factor in this process (Kellett et al., 2002; Rudebeck, 2002; Smyth, 1996). High
levels of experiential processing, therefore, should increase the opportunity and
ability for followers to address the role and socio-emotional needs of leaders. This
should subsequently improve leaders’ perceptions of their relationship with these
followers.
This relationship may be moderated by the degree to which leaders’ endorse
insubordination in their ideal follower. The items that make up the insubordination
IIFT factor (see Table 6.1) appear to be congruent, or more specifically incongruent,
with the development of positive and supportive working relationships. For example,
employees who are arrogant tend to exhibit higher levels of dominance and anger
and lower levels of agreeableness (Johnson et al., 2010). In addition, they are less
likely to engage in organisational citizen behaviours (Johnson et al., 2010). Each of
these traits and behaviours is associated with poorer relationship outcomes (e.g. see
232
Ilies et al., 2007; Kassing, 2000; Russell & Fehr, 1994). Leaders who endorse
relatively high levels of insubordination in their ideal follower, therefore, may place
less value on followers’ attempts to build more positive leader-follower relationships
compared to those who endorse low levels. Consequently, the behavioural tendencies
of followers with high experiential processing styles should be more congruent with
those leaders who endorse lower levels of insubordination in their ideal follower.
The proposed relationship between followers’ experiential processing and
leaders’ LMX may also be moderated by the degree to which leaders’ endorse
commitment in their ideal follower. The items that make up the committed IIFT
factor (see Table 6.1) appear to be aligned with some of the potential behavioural
tendencies of followers with high experiential processing. Specifically, followers
with high experiential processing, especially those who show high levels of
imagination, should show an increased ability to empathise with the needs and goals
of leaders and subsequently adapt their behaviours to address them. From the
perspective of leaders, this is likely to be experienced quite positively (e.g. leader is
understood; work goals are addressed; anticipation of needs). The result is that these
followers are perceived as more competent and supportive. Consequently, the
leaders’ perception of the relationship improves. In support, the receiving of respect
from followers is related to improved leader-follower relationships and
organisational outcomes (Clarke & Mahadi, 2015). Based on the above, it is
hypothesised that:
Hypothesis 2: a significant positive relationship will exist between followers’
experiential processing and leaders’ LMX…
233
Hypothesis 2a and b: ...and this relationship will be stronger for leaders who prefer
low levels of insubordination and high levels of commitment in their followers.
Followers’ Constructive Thinking and Leaders’ LMX
Emotional coping relates to how individuals deal with frustration,
disappointment, rejection, criticism and disapproval (Epstein, 2001). Like leaders,
followers may have to deal with negative situations on a regular basis in the
workplace. For example, they may receive a poor performance review by their leader
(i.e. experience of criticism) or their leader does not listen to their advice (i.e.
frustration). Poor emotional copers would tend to interpret these events via
destructive pathways (e.g. takes the criticism personally; Epstein, 2001). In contrast,
good emotional copers would tend to interpret these situations via constructive
pathways (e.g. do not catastrophise or take criticism personally). Drawing upon
broaden-and-build theory (Fredrickson, 1998; 2001; Garland et al., 2010; Rathunde,
2000), in addition to the findings and argument presented in the previous chapter,
emotional competence in followers should increase their ability to focus their
attention externally (e.g. is criticism valid?) and improve the pool of skills and
resources on which they can draw upon to solve novel issues and problems. In other
words, followers’ increased openness and non-defensiveness should increase
opportunities to develop a more positive working relationship with leaders. In
contrast, low emotional coping in followers should result in an increased use of
psychological and behavioural defence mechanism (e.g. feels their competence is
“attacked” resulting in psychological and/or behavioural defences being marshalled
directly or indirectly towards the leader).
234
In support, research demonstrates that good emotional copers tend to develop
more favourable relationships (Epstein & Meier, 1989). Furthermore, high levels of
negative affect, of which emotional coping has strong negative relationships
(Epstein, 1987b as cited in Epstein, 2003), is associated with a tendency for
individuals to focus on the negative aspects of others and themselves (Levin &
Stokes, 1989; Watson & Clark, 1984), poorer social skills and lower trust (Raja,
Johns, & Ntalianis, 2004). Each of these is likely to undermine the development of a
positive and supportive leader-follower relationship (e.g. see Bernerth, Armenakis,
Field, Giles, & Walker, 2007). Based on the above, we should expect a positive
relationship between followers’ emotional coping and leaders’ LMX ratings.
This relationship may be moderated by the degree to which leaders endorse
conformity in their ideal followers. The openness and non-defensiveness that
followers with high levels of emotional coping may exhibit should help increase
opportunities for finding new ways of doing things. For example, it may facilitate
more creative approaches to issues and/or an increased openness to learning (e.g. see
Amabile, Barsade, Mueller, & Staw, 2005; Bledow, Rosing, & Frese, 2013; Naquin
& Holton, 2002). In support, positive affect is positivity related to employee self-
directed, proactive behaviour while negative affect is negatively related (Ashforth,
Sluss, & Saks, 2007; Fritz & Sonnentag, 2007). For leaders who seek conformity in
their ideal follower, a follower who exhibits these behavioural tendencies may be
less likely to be seen in a positive manner. In contrast, these behavioural tendencies
should be less congruent with those leaders who endorse high levels of conformity.
Based upon the above, it is hypothesised that:
235
Hypothesis 3: a significant positive relationship will exist between followers’
emotional coping and leaders’ LMX…
Hypothesis 3a: ...and this relationship will be stronger for leaders who prefer low
levels of conformity in their followers.
High behavioural copers tend to be more proactive when confronted with
everyday issues and seek to achieve realistic solutions through careful planning and
doing their best (Epstein, 1992; 1998; 2001). Kelley (1988) states that one of the key
dimensions of effective followers is the ability to take the initiative in decision-
making and be active in helping leaders achieve their goals. Therefore, leaders are
likely to show favourable views toward followers who are proactive (e.g. see Green,
1988 as cited by Epstein, 2003; Nadeau, 1994 as cited by Epstein, 2003). In support,
followers’ proactivity is positively associated with increased performance as rated by
managers (Rank et al., 2007; Thompson, 2005).
The tendency for high behavioural copers to be optimistic and positive
(Epstein, 2001) is also likely to help followers’ address the more socio-emotional
needs of leaders. Positivity is a key factor in the development and maintenance of
social relationships as it helps individuals focus on the positive aspects of themselves
and others (Naquin & Holton, 2002). Indeed, positivity and optimism have been
shown to help support individuals communicate more effectively and empathise with
others, which are key elements in the development of relationships (Dulebohn et al.,
2012; Prati et al., 2003; Scott & Bruce, 1994). Consequently, we would expect
followers with high levels of behavioural coping to support the development of a
positive working relationship with their leaders.
236
The generally positive and proactive nature that followers with high
behavioural coping would exhibit is likely to be particularly valued by leaders who
endorse high levels of enthusiasm (see items in Table 6.1). In contrast, leaders who
endorse low levels of enthusiasm may see such behavioural tendencies in followers
somewhat more negatively. In addition, the relationship may be moderated by the
degree leaders endorse commitment. The items that make up the committed IIFT
dimension appear to be congruent with the behavioural tendencies of followers with
high behavioural coping (e.g. goes above and beyond, productive, reliable). Based
upon the above, it is hypothesised that:
Hypothesis 4: a significant positive relationship will exist between followers’
behavioural coping and leaders’ LMX…
Hypothesis 4a and b: ...and this relationship will be stronger for leaders who prefer
enthusiasm and commitment in their followers.
Categorical thinkers tend to think that their way is the only way and that
others are in error for taking alternative perspectives (Epstein, 1998; 2001; Epstein &
Meier, 1989). When their beliefs are challenged, they are also prone to experience
frustration, annoyance and anger (Epstein, 1992; 2001). In addition, they tend to be
more critical, callous and less trusting in their social interactions (Epstein, 1992;
2001). Such tendencies are unlikely to support the ability to develop positive and
supportive working relationships with leaders and other employees (e.g. see Chan &
Treacy, 1996; Katz, 1963; May et al., 2004; Nembhard & Edmondson, 2006;
237
Rogers, 1959; Walumbwa & Schaubroeck, 2009; Yukl, 1998). For example, trust is
an important factor for interpersonal cooperation and the resolution of conflict
(Deutsch, 1973; Kramer & Tyler, 1996; Lewicki et al., 2006; McAllister, 1995).
Accordingly, less trusting followers are unlikely to develop positive relationships
with leaders.
For the same reasons, they are unlikely to develop supportive and effective
working relationships with other team and organisational members. The ability to
cooperate with others though is an important antecedent in positive organisational
outcomes (e.g. see Chakraborti, Boonyasai, Wright, & Kern, 2008; Hoegl,
Parboteeah, & Gemuenden, 2003; Morgeson, Reider, & Campion, 2005). Bringing
about and maintaining a high level of team functioning, therefore, should on average
be a key focus for most leaders (Zaccaro, Rittman, & Marks, 2001). Consequently,
categorical thinking in followers is likely to undermine both directly (e.g. more
insensitive follower-leader interactions) and indirectly (e.g. complaints from other
team members to the leader) leaders’ relationships with them.
The items that make up the insubordinate IIFT dimension (see Table 6.1)
appears to be somewhat congruent with the behavioural tendencies of high
categorical thinkers as described within CEST (Epstein, 2001). In other words,
leaders who endorse insubordination in their ideal follower may value the more
callous and critical nature of followers with high categorical thinking. The proposed
negative relationship between followers’ categorical thinking and leaders’
perceptions of the relationship, therefore, may be moderated by the degree to which
leaders endorse insubordination. Specifically, leaders who endorse high levels of
insubordination should perceive relationships with such followers more positively.
In contrast, leaders who endorse low levels are likely to take a particularly negative
238
view of these followers. Consequently, a poorer leader-follower relationship would
be perceived. Based upon the above, it is hypothesised that:
Hypothesis 5: a significant negative relationship will exist between followers’
categorical thinking and leaders’ LMX…
Hypothesis 5a: ...and this relationship will be stronger for leaders who prefer low
levels of insubordination in their followers.
Method
Participants
The same participants from the previous study were used in the current study
(i.e. small boutique legal firm, a local branch of a national union, in addition to direct
participation). Data was collected from employees and their line managers.
However, in some cases, leaders had not rated the follower who had specifically
participated. Therefore, there was a reduction in the number of dyadic relationships
over the previous study. The final sample consisted of 77 leader-follower dyads
which consisted of 77 followers (68.8% female, 31.2% male) and 18 direct-line
leaders (61.1% male, 38.9% female). Unfortunately, due to time and logistical
constraints, more participants could not be recruited. Average follower age was
41.16 (SD = 10.09; range 20-62). Average leader age was 41.83 (SD = 9.20; range
26-67). The average length of time a follower and leader had been working together
was 42.30 months (SD = 48.28; range 1-300).
239
Measures - Follower
Rational-Experiential Multimodel Inventory (REIm; Norris & Epstein,
2011). The REIm (Norris & Epstein, 2011) is a 42 item measure which asks
individuals to respond on a 5-point Likert scale (1 = Definitely False; 5 = Definitely
True) the degree to which they use a rational (12 items) or experiential processing
style (30 items). The experiential scale consists of 3 sub-scales (10 items each) to
measure its different aspects: (1) intuitive; (2) emotionality and; (3) imagination. The
intuitive facet is designed to assess the ability and engagement of an individual’s use
of intuitive judgements (e.g. “I often go by my instincts when deciding on a course of
action”.) The emotionality sub-scale assesses the intensity, frequency, duration, and
favourable attitude towards strong affect (e.g. “My emotions don’t make much
difference in my life”; reversed scored). The imagination sub-scales assesses the
engagement in, and appreciation of, imagination, aesthetic productions, and imagery
(e.g. “Sometimes I like to just sit back and watch things happen.”) Higher scores
indicate a higher preference for processing on that particular dimension.
Constructive Thinking Inventory (CTI; Epstein, 2001). The constructive
and destructive components of the experiential system were assessed by the 108-item
constructive thinking inventory (CTI, Epstein, 2001). The constructive components
of the experiential system include global constructive thinking, emotional coping,
and behavioural coping. The destructive components are personal superstitious
thinking, categorical thinking, esoteric thinking, and naïve-optimism (Epstein, 2001).
The reader is referred to Table 1.3 presented in Chapter 1 for main-scale and sub-
scale definitions. People respond on a 5-point Likert scale how much a statement is
true of them (1 = Definitely False; 5 = Definitely True). Raw scores are entered into
240
an electronic scoring programme (Epstein & PAR Staff, 2008) and converted to T-
Scores based upon gender norms.
Measures - Leader
Implicit Followership Theory Questionnaire (Sy, 2010) – adapted. The
current study used the Implicit Followership Theory Questionnaire (Sy, 2010)
utilised in the previous studies. Individuals were asked to rate on a 10-point Likert
scale (1 = Not at all Characteristic; 10 = Extremely Characteristic) the degree to
which the traits were characteristic of “their ideal follower in a business setting”.
However, for the reasons described in the introduction, only the 14 items that loaded
on the five ILLTs dimensions (insubordination, committed, incompetence,
conformity and enthusiasm; see Table 6.1) were used. Higher scores are assumed to
indicate a preference for a particular follower dimension.
Multi-dimensional Leader-Member Exchange Questionnaire (LMX-
MDM; Liden & Maslyn, 1998). The Multidimensional Leader-Member Exchange
Questionnaire (LMX-MDM; Liden & Maslyn, 1998) is an extensively used measure
of LMX. Leaders are asked to respond on a 7-point Likert scale (1 = Strongly
Disagree; 7 = Strongly Agree) how much they agree a statement reflects their
perception of four separate, but interrelated, dimensions in the exchange relationship
with a particular follower: (1) affect; (2) loyalty; (3) contribution and; (4)
professional respect. Affect refers to the affection felt for the follower based
primarily upon interpersonal attraction as opposed to professional values. An
example item is “I like this follower very much as a person”. Loyalty refers to the
241
belief that the follower of the dyad will express public support for the leader. An
example item is “This follower would come to my defence if I were attacked by
others”. Contribution refers to the perceptions held by leaders of the support and
involvement they give to the relationship beyond their contractual obligations. An
example item is “I do work for this follower that goes beyond what is specified in my
job description”. Professional respect refers to the perceived reputation of the
follower. An example item is “I admire this follower’s professional skills”. All sub-
scales contain 3 items. An overall LMX score was calculated by taking the average
of all 12 items, which was used as the dependent variable in the current study.
Higher scores indicate higher quality relationships. It should be noted that the name
of a leader’s actual follower was presented as opposed to the generic word of
“follower”.
Procedure & Analytical Approach
The same two-stage procedure described in the previous chapters was used.
In stage one, followers completed the REIm (Norris & Epstein, 2011) while leaders
completed the adapted Implicit Followership Theory Questionnaire (Sy, 2010). Both
also completed some demographic items. Leaders and followers who completed
stage one were invited two days later to complete stage two via email. In stage two,
followers completed the CTI (Epstein, 2001) while leaders completed the LMX
questionnaire (Liden & Maslyn, 1998). The average time for followers between the
completion of stage one and stage two was 4.10 days (SD = 2.90, range 2-14). The
average time for leaders between the completion of stage one and stage two was 5.11
days (SD = 3.79, range 2-14). For the reasons described in the previous chapter, a
242
similar multi-level analytical approach using HLM 7 (Raudenbush et al., 2011) was
used. The reader is directed to the previous chapter for more information.
Results
For each manager in this sample, there was an average of 4.23 followers who
reported to them. Any data point that was more than 1.5 times the Interquartile range
(IQR) above or below the first or above the third quartile was classified as an outlier
(see Tuckey, 1977). Due to the small sample size, outliers were truncated to the
nearest highest/lowest non-outlier score, which allows the relative order of scores
and statistical power to remain but reduces distribution issues (see Osborne &
Overbay, 2004). A total of 10 outliers across all factors used in the current study
were recoded (M = 1.11, SD = 1.97 items per factor). Inspection of the histograms
and Q-Q plots indicated some minor normality issues but the inferential statistics
used are generally quite robust against this (Tabachnick & Fidell, 2007). Table 6.2
presents means, standard deviations, and scale reliabilities for all manager and
follower variables. Except for enthusiasm IIFT scale, all main scales showed
acceptable to excellent levels of reliability (range .73 to .96; Kline, 2000). As can be
seen, though there were some reliability issues with the enthusiasm component (<.6;
George & Mallery, 2003; Kline, 2000). However, it did meet the minimum level of
.5 suggested for research purposes (Kline, 2000) and was similar to the reliability
observed in Chapter 4. Furthermore, Cronbach’s alpha provides the lower-bound
estimate of reliability (Tavakol & Dennick, 2011) Therefore, the enthusiasm
component was retained to explore the hypothesised relationships. However, it is
important to recognise that some authors recommend at least .7 as significant
243
random error exists when reliability is low, which may result in interpretation issues
(e.g. see Bland & Altman, 1997; Tavakol & Dennick, 2011).
Hypothesis Testing
The first step in the analysis was to determine if there were systematic
between-cluster variances in the dependent variable (i.e., leader LMX) by calculating
an inter-class correlation (ICC; Luke, 2004; Woltman, Feldstain, MacKay, &
Rocchi, 2012). The ICC indicated that 37.9% of the variance in leader LMX was
between managers. Given the observed ICC was above the recommended threshold
of .1 (Aguinis et al., 2013), a decision was made to use multi-level modelling for
analyses.
A similar process for analysing the data to that outlined in the previous
chapter was undertaken (see also Harris et al., 2011). Specifically, three statistical
steps were applied to assess each hypothesis. In the first step, the predictor was
entered to test main effect hypotheses (i.e. hypotheses 1 to 5). The proposed
moderator was entered in the second step to assess whether leaders’ LMX ratings are
influenced by their IIFTs (e.g. see g. Bass & Avolio, 1989; Eden & Leviatan, 1975;
2005; Lord et al., 1984; Rush et al., 1977; Schyns & Meindl, 2005; Uhl-Bien et al.,
2014). In the third step, a cross-level interaction variable was formed by multiplying
the predictor and proposed moderator variables. Significant findings at this step
would support the presence of a moderating effect (i.e. hypotheses 1a to 5a). Grand
centring was undertaken on all independent and moderator variables (see Aiken &
West, 1991; Geiser et al., 2013; Hofmann & Gavin, 1998).
244
Table 6.2.
Descriptive statistics, inter-correlations and scale reliabilities for all manager (n = 18) and follower (n = 77) main variables.
Mean SD 1 2 3 4 5 6 7 8 9 10
Follower Variables
REIm
1. Rational 3.88 .57 (.90) -.01 -.10 .33** -.10 .22^ .12 -.05 .14 .25*
2. Experiential 3.40 .36 (.76) .06 .27* -.01 -.02 -0.10 -.10 .23* .24*
Constructive Thinking
3. Emotional Coping 50.97 9.98
(N/A&) .25* -.68** -.09 .10 .07 .03 .32**
4. Behavioural Coping 50.30 6.10
(N/A&) -.11 .13 .09 -.25* -.05 .38**
5. Categorical Thinking 41.44 8.91
(N/A&) .15 .04 -.15 -.24* -.36**
Leader Variables
IIFTs
6. Committed 8.72 1.08
(.79) .18 .02 -.12 .22^
7. Enthusiasm 6.42 1.58
(.55) .16 -.19^ .13
8. Conformity 3.19 1.58
(.73) .26* -.23*
9. Insubordination 1.48 .99
(.96) .21^
LMX
10. Overall 5.68 .74
(.92)
Notes: Cronbach’s alpha reliabilities are shown on the diagonal in brackets.
*p < .05; **p < .01; ^p < .10 &Reliabilities cannot be calculated due to the CTI (Epstein, 2001) being a controlled test. It is therefore assumed that these scales are reliable.
245
As described in the previous chapter, R2 in multi-level modelling is
interpreted as the proportional, or percentage change, in the variance in the residuals
between the tested and comparison model (Luke, 2004). Simple slopes using single-
level analyses (as recommended by Muthen, 2015) were plotted at both one standard
deviation above and below the mean of the proposed moderator and independent
variable to investigate whether a significant moderation was in the expected
direction (see Aiken & West, 1991; Schubert & Jacoby, 2004; Stone & Hollenbeck,
1989).
Hypotheses Related to Follower Information Processing Dimensions
The results of the HLM analysis investigating hypotheses 1 (main effect of
follower rational processing) and 1a (moderating effect of leader conformity IIFT)
are presented in Table 6.3. As can be seen in step one, followers’ rational processing
was positively related to leaders’ LMX and was significant. Consequently,
hypothesis 1 was supported. Step two demonstrates that leaders’ conformity IIFT
was not significantly related to their LMX rating. Step three also demonstrates no
significant interaction between followers’ rational processing and leaders’
conformity. Therefore, hypothesis 1a was not supported.
246
Table 6.3.
Hypotheses 1 and 1a - Hierarchical linear modelling results predicting leaders’
LMX with followers’ rational processing and leaders’ conformity IIFT as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Rational (A) .262** .263** .220^
Moderator
Conformity (B) -.038 -.107
Interaction Term
A×B .017
Level 1 variance of residuals .293 .276 .276 .277
Variance change N/A .017 .000 -.001
ΔR2 N/A .058 .000 -.004
^p<.10; *p<.05; **p<.01
The results of the HLM analysis investigating hypotheses 2 (main effect of
follower experiential processing) and 2a (moderating effect of leader insubordination
IIFT) are presented in Table 6.4. As can be seen, in step one, while followers’
experiential processing was positively related to leaders’ LMX it only approached
significance (p = .08). Consequently, hypothesis 2 was not supported. Step two
demonstrates that leaders’ insubordination IIFT was positively related to their LMX
rating and was significant. However, step three demonstrates no significant
interaction between followers’ rational processing and leaders’ conformity.
Therefore, hypothesis 1a was not supported.
Table 6.4.
Hypotheses 2 and 2a - Hierarchical linear modelling results predicting leaders’
LMX with followers’ experiential processing and leaders’ insubordination IIFT as
moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Experiential (A) .394^ .361 .263
Moderator
Insubordination (B) .191** -.041
Interaction Term
A×B .063
Level 1 variance of residuals .293 .279 .277 .277
Variance change N/A .014 .002 .000
ΔR2 N/A .048 .007 .000
^p<.10; *p<.05; **p<.01
247
The results of the HLM analysis investigating hypothesis 2b (moderating
effect of leader enthusiasm IIFT) are presented in Table 6.5. Specifically, step three
demonstrates no significant interaction between followers’ experiential processing
and leaders’ enthusiasm. Therefore, hypothesis 2b was not supported.
Table 6.5.
Hypothesis 2b - Hierarchical linear modelling results predicting leaders’ LMX with
followers’ experiential processing and leaders’ committed IIFT as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Experiential (A) .394^ .400^ -1.292
Moderator
Committed (B) .132 -.504
Interaction Term
A×B .187
Level 1 variance of residuals .293 .279 .279 .272
Variance change N/A .014 .000 .007
ΔR2 N/A .048 .000 .025
^p<.10; *p<.05; **p<.01
The lack of significance using the experiential main scale was similar to that
observed in the study in Chapter 5. On the basis that imagination is an important
factor in empathy (Katz, 1963; Yukl, 1998) which, in turn, is key to developing
effective relationships (Kellett et al., 2002; Gaesser, 2013; Rudebeck, 2002) a
decision was made to investigate the experiential sub-scale of imagination (M =
3.50, SD = .60, reliability = .758) on leaders’ LMX in addition to whether the two
proposed moderators of the experiential main scale (insubordination and committed)
would instead moderate this sub-scale relationship.
As can be seen in Table 6.6, imagination emerged as a significant positive
predictor of leaders’ LMX in step one. Therefore, some partial support was found for
hypothesis 2. Step two demonstrates that leaders’ committed IIFT was not
significantly related to their LMX rating. Step three demonstrates no significant
interaction between followers’ imagination and leaders’ insubordination. Table 6.7
248
though demonstrates in step three that leaders’ committed IIFT significantly
moderated the relationship between followers’ imagination and leaders’ LMX. To
explore this, simple slopes analysis was undertaken and is presented in Figure 6.1. It
was observed that the slope at 1SD beneath the mean of the moderator only
approached significance (t = 1.68, p =.097). The slope at 1SD above the mean
though was significant (t = 2.602, p = .011). This provides partial support for
hypothesis 3b.
Table 6.6.
Hypothesis 2b - Hierarchical linear modelling results predicting leaders’ LMX with
followers’ imagination and leaders’ insubordination IIFT as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Imagination (A) .362** .347** .121
Moderator
Insubordination (B) .182 -.476
Interaction Term
A×B .166
Level 1 variance of residuals .293 .255 .253 .248
Variance change N/A .038 .002 .005
ΔR2 N/A .130 .008 .020
^p<.10; *p<.05; **p<.01
Table 6.7
Hypothesis 2b - Hierarchical linear modelling results predicting leaders’ LMX with
followers’ imagination and leaders’ committed IIFT as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Imagination (A) 362** .371** -1.633^
Moderator
Committed (B) .151 -.666^
Interaction Term
A×B .223**
Level 1 variance of residuals .293 .255 .254 .238
Variance change N/A .038 .001 .016
ΔR2 N/A .130 .004 .063
^p<.10; *p<.05; **p<.01
249
Figure 6.1. Moderating effect of leader committed IIFT (1±SD) on the relationship
between follower imagination (1±SD) and leader LMX.
Hypotheses Related to Follower Constructive Thinking Dimensions
The results of the HLM analysis investigating hypotheses 3 (main effect of
follower emotional coping) and 3a (moderating effect of leader conformity IIFT) are
presented in Table 6.8. Step one demonstrates that followers’ emotional coping was
positively related to leaders’ LMX and was significant. Consequently, hypothesis 3
was supported. Step two demonstrates that leaders’ conformity IIFT was not
significantly related to their LMX rating. However, step three demonstrates a
significant interaction between followers’ emotional coping and leaders’ conformity.
To explore this, simple slopes analysis was undertaken and is presented in Figure
6.2. The slope 1SD beneath the mean of the moderator was not significant (t = 1.328,
p = .188). The slope 1SD above the mean though was significant (t = 3.214, p =
.002). This provides support for hypothesis 3a.
1
2
3
4
5
6
7
Low High
Lea
der
LM
X
Follower Imagination
Low Committed High Committed
250
Table 6.8.
Hypotheses 3 and 3a - Hierarchical linear modelling results predicting leaders’
LMX with followers’ emotional coping and leaders’ conformity IIFT as moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Emotional Coping (A) .020* .020* -.002
Moderator
Conformity (B) -.041 -.049*
Interaction Term
A×B .009*
Level 1 variance of residuals .293 .273 .274 .251
Variance change N/A .020 -.001 .023
ΔR2 N/A .068 -.004 .084
^p<.10; *p<.05; **p<.01
Figure 6.2. Moderating effect of leader conformity IIFT (1±SD) on the relationship
between follower emotional coping (1±SD) and leader LMX.
The results of the HLM analysis investigating hypotheses 4 (main effect of
follower behavioural coping) and 4a (moderating effect of leader enthusiasm IIFT) is
presented in Table 6.9. As can be seen, in step one followers’ behavioural coping
was positively related to leaders’ LMX and was significant. Consequently,
1
2
3
4
5
6
7
Low High
Lea
der
LM
X
Follower Emotional Coping
Low Conformity High Conformity
251
hypothesis 4 was supported. Step two demonstrates that leaders’ enthusiasm IIFT
was not significantly related to their LMX rating. Step three demonstrates no
significant interaction between followers’ behavioural coping and leaders’
enthusiasm. Therefore, hypothesis 4a was not supported.
Table 6.9.
Hypotheses 4 and 4a - Hierarchical linear modelling results predicting leader LMX
with Follower Behavioural Coping and Leader Enthusiasm IIFT as Moderator
Baseline Step 1 Step 2 Step 3
Independent Variable
Behavioural Coping (A) .031** .031** .055
Moderator
Enthusiasm (B) .039 .220
Interaction Term
A×B -.003
Level 1 variance of residuals .293 .283 .281 .279
Variance change N/A .010 .002 .002
ΔR2 N/A .034 .007 .007
^p<.10; *p<.05; **p<.01
The results of the HLM analysis investigating hypothesis 4b (moderating
effect of leader committed IIFT) are presented in Table 6.10. Specifically, step three
demonstrates no significant interaction between followers’ behavioural coping and
leaders’ committed IIFT. Therefore, hypothesis 4b was not supported.
Table 6.10.
Hypothesis 4b - Hierarchical linear modelling results predicting leader LMX with
Follower Behavioural Coping and Leader Committed IIFT as Moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Behavioural Coping (A) .031** .031** -.086
Moderator
Committed (B) .121 -.528
Interaction Term
A×B .013
Level 1 variance of residuals .293 .283 .281 .282
Variance change N/A .010 .002 -.001
ΔR2 N/A .034 .007 -.004
^p<.10; *p<.05; **p<.01
252
The results of the HLM analysis investigating hypotheses 5 (main effect of
follower categorical thinking) and 5a (moderating effect of leader insubordination
IIFT) is presented in Table 6.11. As can be seen, in step one followers’ categorical
thinking was negatively related to leaders’ LMX and was significant. Consequently,
hypothesis 5 was supported. Step two demonstrates that leaders’ insubordination
IIFT was significantly related to their LMX rating. Step three demonstrates a
significant interaction between followers’ categorical thinking and leaders’
insubordination. To explore this, simple slopes analysis was undertaken and is
presented in Figure 6.3. It was observed that the slope at 1SD beneath the mean of
the moderator was not significant (t = 1.328, p = .188). The slope at 1SD above the
mean though was significant (t = 3.214, p = .002). Surprisingly, therefore, the
interaction was not in the expected direction. Thus, hypothesis 5a was not supported.
Table 6.11.
Hypotheses 5 and 5a - Hierarchical linear modelling results predicting leaders’
LMX with followers’ categorical thinking and leaders’ insubordination IIFT as
moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Categorical Thinking (A) -.032** -.031** -.008
Moderator
Insubordination (B) .170** .739**
Interaction Term
A×B -.016**
Level 1 variance of residuals .293 .239 .237 .228
Variance change N/A .054 .002 .009
ΔR2 N/A .184 .008 .038
^p<.10; *p<.05; **p<.01
253
Figure 6.3. Moderating effect of leader insubordination IIFT (1±SD) on the
relationship between follower categorical thinking (1±SD) and leader LMX.
Discussion
As leaders and followers do not usually experience their relationship in the
same way (see Gerstner & Day, 1997; Sin et al., 2009; van Gils et al., 2010; Wilson
et al., 2010), the previous chapter presented an argument which integrated CEST
(Epstein, 1998, 2001; 2003) with implicit leadership and followership theory (Lord
& Maher, 1993; Sy, 2010) to explain the variations observed between leaders’ and
followers’ LMX ratings (Gerstner & Day, 1997; Sin et al., 2009). Two general
models were developed based on this argument (see Figures 5.1a and 5.1b in Chapter
5). In these models, both leaders’ and followers’ information processing style (Norris
& Epstein, 2011) and constructive thinking (Epstein, 2003) were theorised as
antecedents to followers’ and leaders’ LMX rating respectively. This was based on
the premise that these personality factors, in particular, those linked to the
experiential system would influence a wide range of human behaviours across
1
2
3
4
5
6
7
Low High
Lea
der
LM
X
Follower Categorical Thinking
Low Insubordination High Insubordination
254
contexts and situations (Epstein, 1994; 2001; 2014) including those occurring in
leader-follower relationships. Drawing upon ILT and IFT theories (see Lord &
Maher, 1993; Sy, 2010), these two models also explicitly recognise that the
behaviours of the dyadic partner are likely to be interpreted and evaluated in
different ways. Specifically, followers and leaders would draw upon their IILT or
IIFT respectively when making judgements and evaluating the dyadic partner. This
is because these would take account of some of the observer’s personal preferences
and tendencies as opposed to a more general ILT or IFT (see Chapters 3 and 4).
Evidence in Chapter 5 supports the notion that information processing and
constructive thinking in leaders were, in general, related to followers’ LMX ratings.
Furthermore, these relationships were moderated by followers’ IILTs. This chapter,
therefore, tested whether these personality dimensions in followers would relate to
leaders’ LMX ratings and whether they would also be moderated by leaders’ IIFTs.
In general, the results support the idea that these personality dimensions in followers
predicted leaders’ LMX ratings. However, while there was some evidence to
demonstrate these relationships as moderated by leaders’ IIFTs, it was not as
conclusive as that observed in Chapter 5. Theoretical and practical implications are
discussed below.
Followers’ Information Processing and Leaders’ LMX
As hypothesised, rational processing in followers was positively related to
leaders’ LMX ratings. This finding can be contrasted with the results presented in
Chapter 5, which found no significant relationship between leaders’ rational
processing and followers’ LMX. One potential reason for this difference relates to
255
what leaders and followers perceive as valued resources from their dyadic partner
(see Wilson et al., 2010). For followers, their inherent lower power in the
relationship is likely to increase the value they place on how leaders use that power
(i.e. coercively vs. participatory). Specifically, they may place a higher value
(compared to leaders) on leaders’ social behaviours (e.g. inclusion in decision
making, support, caring, empathy). In support, consideration of followers’ needs and
facilitating their participation in decision-making are amongst the most important
antecedents of follower satisfaction with leaders (Brown, 1996; Yukl, 1998). In
contrast, leaders with their increased power and focus on managing and directing
action towards individual, team and organisational outcomes are likely to place a
higher value on their followers’ competence in achieving these (e.g. see Hollander &
Offermann, 1990; Sy, 2010).
The rational and experiential systems appear to be differentially related to
contributing to resources that are valued by leaders and followers. Specifically
relative to the experiential system, rational processing is not as strongly related to the
development of positive and supportive relationships (Epstein et al., 1996).
Consequently, while rationality in leaders may help them personally think through
issues and problems in the workplace and correct initial judgments in error (e.g. see
Cerni et al., 2008; King, 2012; Norris & Esptein, 2011), it does not appear to be
perceived or valued in the same way by followers (Norris & Epstein, 2011; Simoens,
2014). Rational processors though tend to be action focused (Epstein et al., 1996)
and use rational influencing tactics (Cerni et al., 2012) both of which are likely to
help increase the opportunities for followers to behave in ways which are valued by
most leaders (Oc & Bashshur, 2013; Rank et al., 2007; Thompson, 2005). Rational
processing in followers, therefore, appears to support behaviours, which are more
256
valued by leaders. In contrast, the experiential system appears to support those
behaviours in leaders that are more valued by followers. This position is aligned with
wider leadership theories which generally see leaders as focused on instilling a
positive vision of the future and effectively managing relationships with followers in
order to align their actions with this vision (e.g. Baker, 2007; Chaleff, 2003; Kotter,
1990; Wieseke, Ahearne, Lam, & Dick, 2009). In contrast, followers enact this
vision and, therefore, need to be more practical focused or ‘hands-on’.
The relationship between followers’ rational processing and leaders’ LMX
though was not moderated by leader conformity. It was implicitly assumed that a
high rational processing style would serve to act as a barrier to influence attempts by
others such as the leader. However, research suggests that this assumption may not
be correct. Specifically, there is limited and conflicting evidence to suggest a linear
relationship between influenceability and cognitive intelligence (see Rhodes &
Wood, 1992), of which, rational processing shows the strongest correlations (see
Epstein et al., 1996). From the perspective of leaders, therefore, while high levels of
rational processing may help followers to be effective in their role (e.g. think through
problems, proactive), it does not necessarily mean that such followers would be hard
to influence. Consequently, the congruence argument may not apply in a simple
linear process as originally believed.
Contrary to the hypothesis, followers’ experiential processing was not
positively related to leaders’ LMX, although it was in the expected direction and did
approach significance. This is not totally surprising as a similar non-significant
finding between leaders’ experiential processing and followers’ LMX was observed
in Chapter 5. However, similar to Chapter 5, exploratory analyses did demonstrate
followers’ imagination, a sub-factor of the experiential system (Norris & Epstein,
257
2011), was positively related to leaders’ rating of the relationship. Leaders who
endorsed high levels of commitment in their ideal follower particularly valued this
type of processing in followers’. Similar to the explanation for leaders’, it appears
that imagination, due to its integral links to empathy (see Gaesser, 2013), helps
followers to take the perspective of their leaders so they can understand and predict
what they may need, and how they may be affected by, the followers’ decisions and
actions. Furthermore, it helps support followers’ performance (e.g. Salanova, Agut,
& Perió, 2005; Borman, Penner, Allen, & Motowidlo, 2001) and improves team
cohesiveness (Rapisarda, 2002), which is likely to help further address leaders’
needs. Consequently, imagination in followers appears to support the development
of positive and supportive working relationships with leaders.
Contrary to expectations, though, a leader insubordination IIFT did not have
the same moderating effect as their committed IIFT on the relationship between
followers’ imagination and leaders’ LMX. It may be that leaders who seek relatively
higher levels of insubordination in their ideal follower may not actually experience
such follower behaviours in a congruent way with their basic needs (see Epstein,
2003). According to ILT and IFT theory (Lord & Maher, 1990; 1993; Lord et al.,
1984: Phillips & Lord, 1982; Sy, 2010), individuals demonstrate an increased
propensity to look for information, which confirms or disconfirms whether an
individual matches their ILT or IFT and, thus, can be classified as either a leader or
follower. Individuals, though, who show an increased propensity to see anger and
hostility in others, of which leaders who endorse high levels of follower
insubordination would likely do, can result in themselves being more angry and
aggressive (Penton-Voak et al., 2013). Being angry and aggressive can be considered
as a negative affective state. This is unlikely to address the basic need of minimising
258
pain and maximising pleasure (see Epstein, 2003). Furthermore, such behaviours in
both leaders and followers are unlikely to promote a supportive and positive working
relationship. Yet the need to relate to others is also a basic need specified in CEST
(Epstein, 2003). Consequently, while some leaders may state they seek
insubordination in their ideal follower the actual experience of such followers may
actually be experienced quite negatively relative to those who do not.
Followers’ Constructive Thinking and Leaders’ LMX
It appears that high levels of emotional coping in followers help them to
develop positive working relationships with their leaders. This finding aligns with
wider research demonstrating negative links between neuroticism and the
development of positive and supportive relationships across broad domains (Aeron
& Pathak, 2012; Botwin, Buss, & Shakelford, 1997; Bouchard, Lussier, & Saborin,
1999; Caughlin, Huston, & Houts, 2000; Neyer & Voigt, 2004). According to
broaden-and-build theory, negative affect results in attention turning inward towards
the self which then result in rigid, defensive behaviours being marshalled (Garland et
al., 2010). In the context of the leader-follower relationship, therefore, a high level of
follower emotional coping appears to facilitate a cognitive and emotional openness
to the needs of leaders1. Specifically, when faced with criticism, rejection or
frustration, high levels of emotional coping appears to act as a protective mechanism
against turning their attention inward and marshalling psychological and behavioural
defence mechanisms. This, in turn, should facilitate increased opportunities, albeit
via indirect pathways, to attend to the needs of leaders (e.g. open to feedback, more
1 The same should also occur with other members of the follower’s team and other co-workers (e.g.
see Aeron & Pathak, 2012) which should indirectly support the meeting of the leader’s needs (e.g.
improved team cohesiveness, more positive working climate).
259
professional behaviour, increased teamwork). From the perspective of leaders,
therefore, high levels of emotional coping in their followers is likely to be
experienced positively via direct (i.e. face-to-face interactions) and indirect (e.g. via
other employees, improved team cohesiveness) pathways.
This is supported by research which demonstrates that employees who
experience low levels of negative affect and/or increased emotional stability tend to
show improved responses to leader feedback (Ilies, de Pater, & Judge, 2007),
increased performance (Barrick & Mount, 1991; Tett, Jackson, & Rothstein, 1991;
Judge, Erz, & Bono, 1998), and improved teamwork (Aeron & Pathak, 2012).
Consequently, there should, in general, be a corresponding increase in the receipt of
valued resources from followers (i.e. address leaders’ needs) who demonstrate high
levels of emotional coping. This, in turn, would result in the improvement, which is
observed in leaders’ perceptions of their relationships.
Unsurprisingly, the proactive and positive tendencies of followers with high
levels of behavioural coping (Epstein, 2001) have positive impacts on leaders’
perceptions of the relationship. A similar explanation to that presented in Chapter 5
which draws upon broaden-and-build theory (Fredrickson, 1998; 2001; 2003) can be
used to explain this. Specifically, it appears that the increased tendency for followers
with high levels of behavioural coping to experience positive affect helps expand
their awareness, thinking and behaviours (see Fredrickson, 2003). Due to this
increased experience, openness, and proactivity followers develop an increased pool
of personal, professional, and social skills and knowledge. This increased variety of
resources (relative to followers with lower behavioural coping) theoretically should
enable them to better adapt to new situations and problems in the workplace. Indeed,
employees who demonstrate higher levels of behavioural coping tend to voluntarily
260
take on increased workloads while also showing less stress and increased happiness
(Green 1988 as cited in Epstein, 2003). Furthermore, individuals who show
increased positive affect are more open to critical feedback and information (Estrada
et al., 1997; Raghunathan & Trope, 2002) in addition to showing increased empathy
(Waugh & Fredrickson, 2002). From the perspective of leaders, such behavioural
tendencies in followers are likely to be highly valued which, in turn, should improve
their perception of their followers’ competence and loyalty. The result, therefore, is
an improved working relationship.
The observation that neither leaders’ committed nor enthusiasm IIFTs
significantly moderated the relationship between followers’ behavioural coping and
leaders’ LMX was quite surprising. Theoretically, the congruence between the
proactive and positive minded tendencies of followers with high behavioural coping
and leaders who seek high levels of commitment and enthusiasm in their ideal
follower should have moderated this relationship. For the enthusiasm factor, this
may be due to the low level of reliability coupled with the small sample size, which
undermined the ability to detect significant differences. For the committed IIFT
though, this may have been due to the observation that the majority of leaders seek
followers who are reliable, productive and work hard (i.e. high levels of follower
commitment). This is obviously unsurprising considering organisations exist, at their
most basic level, to perform in some way (e.g. financial performance, customer
satisfaction). This lack of variability in what leaders seek from followers, therefore,
may not be enough to detect any moderation effects.
As expected, the cognitive and behavioural tendencies of followers with high
levels of categorical thinking negatively impacted on leaders’ perceptions of the
relationship. The fact that categorical thinkers tend to show an increased likelihood
261
to see others who voice antithetical positions as in error, be less trusting of others, be
more pejorative, and are quick to anger and experience annoyance (Epstein, 1998;
2001), is unlikely to support the development of positive and supportive working
relationships with leaders. For example, followers who regularly take a different
perspective to leaders on the best course of action, due to differences in frames of
reference or how they tend to see problems and situations, may see such leaders as
typically in error. They may voice such a position in a particularly judgmental
manner both directly (i.e. face-to-face with leader) and indirectly (e.g. tell other team
members that the ‘boss is an idiot’ which then gets feedback to the leader by these
other team members). From the perspective of leaders, such behaviours are unlikely
to reinforce the perception of a positive and supportive working relationship. Indeed,
while effective followers show the courage to voice antithetical positions to their
leaders when required, they need to do this in a positive and respectful way that
promotes a positive working environment and demonstrates support for their leaders
(Chaleff, 1995; 2003; 2008; 2009; Kelly, 1988; 1992; 2008). It appears, though, that
categorical thinking in followers does not help achieve this.
While the negative impact of followers’ categorical thinking on leaders’
LMX is unsurprising, the observation that this relationship was stronger when
leaders endorsed high, rather than low, levels of insubordination was not anticipated.
As noted previously, ILT and IFT theory state that implicit leadership and
followership theories serve as an important basis for what perceivers attend to, what
they encode, the information they retrieve when recalling leader or follower related
information, and guiding people’s interpretations of the leaders’ or followers’
behaviours (Lord & Maher, 1990; 1993; Lord et al., 1984; Phillips & Lord, 1982).
Consequently, there is an increased propensity to look for information, which
262
confirms of disconfirms the notion that the observed is either a leader or follower
and, thus, how they act and feel towards them. Individuals who show an increased
propensity to see anger and hostility in others results in themselves being angrier and
more aggressive (Penton-Voak et al., 2013), which are negative affective states.
As noted previously, a negative affective state is not aligned with the basic
need for maximising positive affect and minimising negative affect (Epstein, 2003;
2014). Consequently, leaders who have a propensity to seek insubordination in their
followers may actually result in them experiencing negative affect if they do indeed
observe such behaviours in their follower. This negative leader affective state results
in a corresponding decrease in these leaders’ perceptions of such followers resulting
in a more negative relationship rating. Categorical thinking in followers, therefore,
appears to be experienced particularly negatively by leaders’ who endorse relatively
higher levels of insubordination in their ideal follower. This is due to their increased
tendency to seek out (consciously and unconsciously) negative behaviours in
followers, which results in an increased propensity to react negatively to such
behaviours. This is still aligned with what may be predicted by person-environment
theory (e.g. Kristof-Brown, 1996; Kristof-Brown et al., 2005) but in this case, the
originally hypothesised congruence target was incorrect. Specifically, rather than
IIFT schema congruence, it was the basic need congruence in the experiential system
that proved to be key. From a practical standpoint, this suggests that leadership
training should seek to reduce the degree to which leaders seek insubordinate
followers. This should improve relationships with more callous and less trusting
employees given that managers do not react as negatively to them (i.e. a reinforcing
cycle of negative interactions is reduced or mitigated).
263
Concluding Remarks and Practical Implications
The current study demonstrates support that follower information processing
style and constructive thinking impacts on how leaders’ experience the leader-
follower relationship. However, support for the proposition that these relationships
would be moderated by what leaders’ seek in their ideal followers was not as
convincing as the results observed in Chapter 5. Two possible reasons are offered as
to why this may be the case. First, there may not be enough variability in some of the
IIFT factors to detect moderation effects. Indeed, given that most organisations exist
to produce some form of outcome, it is not surprising that most leaders’ seek their
employees to be competent, committed, hardworking and high performing (see Sy,
2010). Indeed, Lord and Engle (1997) state that “supervisors rely on implicit
performance theories to form impressions of subordinates” (p. 992). Such theories
can be considered more normative in terms of the expectations of how followers
should act. The IIFTs that the current sample of leaders held, therefore, could reflect
more socially normative follower expectations as opposed to specific individual
preferences for a particular type of follower. For example, it is a socially normative
expectation that followers be hard working and competent; friendship with followers
therefore become less important for leaders. Consequently, this socially normative,
but limited IIFT variability undermines the ability to detect any moderation effects
potentially becomes minimised. The explicit measurement approach may have also
compounded this issue (i.e. explicit attitudes of follower).
The second potential reason is due to the small sample size, in particular, the
limited number of leaders. The variation in the number of followers per leader
suggests that the responses from a leader with a higher than average number of
followers (e.g. some leaders had up to 10 followers) could have a particularly high
264
risk of biassing statistical analyses and undermining the ability to uncover the true
relationships within the general population. It is not possible to tell if this has
occurred in the current data set. However, there is some indirect support.
Specifically, the average conformity IIFT in the current study was somewhat lower
than that observed in Chapter 4.
One of the key practical implications from the current study is that leader-
follower relationships could be theoretically improved via follower interventions,
which seek to develop and improve their information processing style and
constructive thinking. Research demonstrates that both of these personality elements
can be improved through training (see Cerni et al., 2010; Epstein, 1998; 2014).
Therefore, follower training which aims to improve their rational processing,
emotional coping, behavioural coping, while reducing their categorical thinking
should improve leaders’ perceptions of followers’ competence over time as these
aspects of personality develop. While they did not use CEST as their framework,
Proudfoot, Corr, Guest and Dunn (2009) provide evidence to support that such
follower training programmes would have positive outcomes. Specifically,
employees who went through a 13-week training programme based upon cognitive-
behavioural therapy principles (e.g. automatic thoughts, goal setting, attributions,
action planning, Socratic questioning, self-observation, and weekly assignments)
showed improvements in relation to productivity, job satisfaction, attribution styles,
self-esteem, employee turnover, and psychological strain.
265
Limitations
This study has limitations that are similar to those evidenced in Chapter 5.
First, as discussed above, the low sample size may have undermined the ability to
detect some relationships. Second, as leaders’ IIFTs are related to their information
processing styles and constructive thinking (see Chapter 4), it is quite possible that it
is a congruence between leaders’ and followers’ information processing styles and
constructive thinking that is contributing to leaders’ LMX rating. Indeed, there are
significant bodies of research demonstrating that congruence between leader and
follower personality factors contributes to higher LMX rating (see Schaubroeck &
Lam, 2002; Strauss et al., 2001; Zhang, et al., 2012). Third, the IFT questionnaire is
an explicit measure used to measure an implicit construct. Therefore, there is a
misalignment in the methodology that can cause interpretation issues (Uhlmann et
al., 2012). However, as noted by Epitropaki and colleagues (2013), the use of
explicit and implicit measures is still considered useful for extending theory and
research. Future research though may wish to use an implicit measure to further
investigate whether similar relationships emerge.
266
CHAPTER 7
Review of Studies and Conclusions
The current research sought to address three gaps in leadership research and
theory. The first gap studies 1 and 2 (Chapters 3 and 4 respectively) addressed was
the need to extend theory around, and identify the antecedents to, both implicit
leadership and followership theories (see Keller, 2003; Schyns & Meindl, 2005).
Extending research in these areas carries important implications, given these
constructs influence how individuals make sense of leaders and followers (Lord &
Maher, 1990; Poole et al., 1989; Sy, 2010; Weick, 1995) and, thus, influence group
and organisational outcomes (see Hansbrough, 2005; Lord et al., 1984; Nye, 2005;
Nye & Forsyth, 1991). The second gap, which studies 3 and 4 addressed, was the
need to extend the research looking at the antecedents of leader-member exchange
(Mahsud et al., 2010). The third gap, which was also addressed by studies 3 and 4
(Chapters 5 and 6 respectively), was to explain why leaders and followers do not
usually see ‘eye-to-eye’ with regards to their existing relationship (see Gerstner &
Day, 1997; Sin et al., 2009). This is despite existing LMX theory which suggests that
both leaders and followers will experience their relationship similarly (Graen & Uhl-
Bien, 1995; van Gils et al., 2010). It was proposed that these gaps could be addressed
through an integrative approach.
An integrative approach to understanding leadership is important as
traditionally the focus of leadership research and practice has been on leaders at the
expense of followers (Day, 2001; Goffee & Jones, 2001; Lord & Brown, 2001; Lord
& Maher, 1993; Shamir, 2007). However, many theorists recognise that followers
play highly active roles in the leadership process (e.g. Chaleff, 1995; 2003; 2008;
2009; Kelley, 1988; 1992; 2008; Lord & Brown, 2001; Shamir, 2007; Uhl-Bien et
267
al., 2014). Consequently, there have been increasingly stronger calls in the literature
to take a more holistic approach to leadership (e.g. Hosking, 2002; Shamir, 2007;
Uhl-Bien et al., 2014; Schyns et al., 2011; 2012). Recognising this need for theory
integration, the research in the preceding chapters tested various components of a
generalised leadership model, which is reproduced in Figure 7.1.
Figure 7.1. General research model tested in studies 1 to 4.
Leadership in this model was conceptualised as a relationship between
leaders and followers’ and that the quality of this relationship is influenced by the
interacting internal worlds (i.e. cognitive-emotional) and role expectations of both
dyadic partners (see Bornstein, 2003; Lord & Brown, 2001; Stech, 2007). Based
upon this conceptualisation, this model integrated Cognitive-Experiential Self
268
Theory (CEST; Epstein, 1998; 2003; 2014), implicit leadership theory (Lord &
Maher, 1990; 1993), implicit followership theory (Sy, 2010), and leader-member
exchange (Dansereau et al., 1975; Graen et al., 1973; Graen & Uhl-Bien, 1995) to
help: (1) explain why leaders and followers vary in their preferences for specific
follower and leader traits respectively (see Offerman, Kennedy, & Wirtz, 1994); and
(2) understand how these preferences may impact on the value perceived by both
leaders and followers with regards to what is provided by the dyadic partner in their
relationship and thus how they each perceive the quality of the relationship (see van
Gils et al., 2010; Wilson, Sin, & Conlon, 2010).
Studies 1 and 2 – General Findings and Conclusions
Studies 1 and 2 extend theory on how and why IILT and IIFT ‘schemas’ are
created and develop over time. Specifically, studies 1 and 2 explored how leaders’
and followers’ information processing style and constructive thinking as described in
CEST (Epstein, 1998; 2001; 2003; Norris & Epstein, 2011) influenced their
preferences for particular traits and behaviours in followers and leaders respectively.
Across both studies, it was theorised that followers and leaders seek to think, behave
and communicate in ways that are aligned with their information processing style
and constructive thinking (see Berne, 1961). When leaders or followers
communicate in ways that are aligned with their preferred style, they will experience
positive vibes and emotions (see Berne, 1961; McCann & Higgins, 1988). When
communication is incongruent, these will be less positive. Given that emotions are
fundamental to how the experiential system learns, these affective experiences act as
learning opportunities for leaders and followers to associate particular follower and
269
leader traits respectively which have contributed to interactions and working
environments which are congruent or incongruent with how they tend to processes
information. The research in studies 1 and 2 generally support this theoretical
framework. Indeed, there was general evidence to support both information
processing styles and constructive thinking independently contributed to explaining
both leaders and follower IIFTs and IILTs respectively to varying degrees.
Therefore, these aspects of the initial research model were generally supported.
It is important to emphasise that ILT and IFT targets investigated were that of
an ideal leader and follower (i.e. IILT and IIFT) rather than a more general ILT or
IFT target. This point is pertinent as the original ILT was initially developed based
upon responses to the targets of leaders, effective leaders, and supervisors (see
Offerman et al., 1994). Similarly, the IFT measure by Sy (2010) was developed
based upon responses to the targets of a follower, effective follower, ineffective
follower and subordinate. None of these targets specifically references an ideal
leader or ideal follower. Perhaps the closest targets are effective leaders and effective
followers. However, effective and ideal are not necessarily synonymous in their
meaning. For example, the word effective is defined as a success is achieving some
desired outcome (Effective, n.d.). Ideal though is defined as satisfying one’s idea of
what is perfect or most appropriate (Ideal, n.d.). In the former, the definition is more
about achieving some external objective or the extent to which some particular
problem is overcome. In the latter, though, the emphasis is more on addressing one’s
preferences and perceptions. These differences are likely to influence how
respondents conceptualise and respond to these targets (see Offerman et al., 1994;
Sy, 2010).
270
This observation is important for both theoretical and methodological
reasons. Offerman and colleagues (1994) measure is one of the most popular in the
ILT literature. However, the ways in which the ILT and IFT measures were
developed suggests that it may not fully capture followers’ conceptualisations of an
ideal leader; but rather only the schemas of more general or effective leader targets.
It is important, therefore, to recognise that previous research which has stated that
their focus is on individuals’ ideal leader schemas (e.g. Epitropaki & Martin, 2004;
Keller, 1999; Koommoo-Welch, 2008) appears to have used these popular measures
without adequate adaption on the assumption they will be appropriate for assessing
participants’ ideal leader schemas.
This methodological issue, therefore, may partly explain why the reported
relationships between personality characteristics and ILTs in past research have been
somewhat weak (e.g. Keller, 1999). Specifically, the antecedents to an ideal and
general ILT and IFT would likely be different. As touched upon in the discussion of
studies 1 and 2, the use of an ideal leader or ideal follower schema should be more
cognitively primed than a general schema in the mind of followers or leaders
respectively during leader-follower interactions in the general workforce. This is
because these schemas would take into account their needs, values and preferences.
It is, therefore, important for future ILT and IFT research to clearly distinguish the
leader or follower target when designing their research.
Studies 3 and 4 – General Findings and Conclusions
Studies 3 and 4 held two core aims. The first of these aims was to explore
how leaders’ and followers’ information processing style and degree of constructive
271
thinking impact on their working relationships as perceived by their dyadic partner.
The general theoretical approach adopted was that both leaders’ and followers’
information processing and constructive thinking would strongly influence the
general behaviours that manifest in the leader-follower relationship (e.g. see Cerni et
al., 2012; Curtis & Lee, 2013; Epstein, 2001; Norris & Epstein, 2011).
Consequently, these cognitive and behavioural differences should influence how
leaders’ and followers’ perceive the nature of their leader-follower relationship.
Studies 3 and 4 both generally found that leaders’ and followers’ information
processing style and various elements of their constructive thinking influenced their
followers’ and leaders’ perceptions of the leader-follower relationship respectively.
In particular, the research suggested that these cognitive factors contributed to traits
and behaviours that are valued (or not) by the dyadic partner. However, there was
also evidence to suggest that rational thinking impacts on the leader-follower
relationship differently depending on whether the individual is a leader or a follower.
Specifically, leaders’ rational processing did not influence the quality of the leader-
follower relationship as perceived by followers. However, in study 4, followers’
rational processing did influence their leaders’ perceptions of the relationship.
Coupled with the other research findings discussed elsewhere (see Cerni et al., 2008;
King, 2012; Simoens, 2014), these differences suggest that rational processing
contributes differently to the leader-follower relationship depending on what a leader
or follower values in an individual who occupies a follower or leader role
respectively. Explicitly, it appears that rational processing in followers contributes to
behaviours which are valued by leaders. In contrast, rational processing in leaders
does not appear to contribute to behaviours that are valued by followers.
272
The second core research aim of studies 3 and 4, that is perhaps the most
relevant for extending LMX theory and research, was to explore if and how
followers’ IILT and leaders’ IIFT influenced their interpretations and evaluations of
their dyadic partner’s behaviours. Drawing upon existent ILT and IFT theory (Lord
& Maher, 1990; 1993; Sy, 2010), it was hypothesized that leaders and followers
draw upon their IIFT and IILT respectively during leader-follower interactions to
interpret and evaluate their dyadic counterpart’s behaviours (see also van Gils et al.,
2010; Wilson et al., 2010). As noted above, an ideal schema is most likely to be
prevalent during these interactions because these would, at least in part, account for
the observer’s cognitive-emotional tendencies (as suggested by the findings of
studies 1 and 2). Consequently, this influences how leaders and followers evaluate
the quality of their relationship.
Study 3 found strong evidence that followers’ IILTs influence how they
interpret and evaluate the behaviours of leaders. Study 4 though did not provide
strong support for the same process occurring from the perspective of leaders. This is
probably due to the small sample utilised. However, it is important to recognise that,
even though the small sample size likely undermined the ability to detect significant
moderation effects, the fact that some did emerge provides initial evidence that
leaders also draw upon their IIFTs when interpreting and evaluating the actions of
their followers.
These findings extend LMX theory and research by providing insight into
how and why followers and leaders do not necessarily perceive their relationship in
the same way (see Gerstner & Day, 1997; Sin et al., 2009). Specifically, all things
being equal, variation in followers’ LMX is not solely dependent on the cognitive-
emotional tendencies of the leader, but also on how followers’ interpret these. These
273
findings sustain the proposition in LMX that not all followers perceive and evaluate
leaders in the same way (Dansereau et al., 1975; Graen, et al., 1973). They also
provide some insight into what follower characteristics influence these differences.
There is also some limited support that a similar process is occurring from the
perspective of leaders. Specifically, that variation in leaders’ LMX is not only
dependent on the cognitive-emotional tendencies of followers, but also on how
leaders’ interpret these tendencies in their day-to-day interactions.
These results have important theoretical implications for more mainstream
leadership theories with regards to where leadership resides. Mainstream leadership
theories (e.g. transformational leadership) specify a predefined model which
describes what an effective leader is (see Antonakis et al., 2003; Avolio & Bass,
1991; Bass & Avolio, 1990; 1993; 1997; House, 1977). As noted by Day (2001),
such models are used in leader development interventions with the aim to change
leaders’ cognitions and/or behaviours to fall in line with these models. While there is
evidence to demonstrate that these leadership models positively influence the
performance, motivation and satisfaction of the average follower (see DeGroot,
Kiker, & Cross, 2000; Judge & Piccolo, 2004), the current research strongly suggests
that such models will not fit, or be effective for, all followers’ conceptualisations of
what they believe an ideal leader should be (see also Dvir & Shamir, 2003; Moss &
Ngu, 2006; Wofford et al., 2001). Therefore, there is an inherent danger in typical
leader development interventions that aim to align a leader’s cognitions and
behaviours to some abstract model without taking consideration the beliefs of a
leader’s followers.
To highlight the danger of assuming that one type of leadership model will be
effective on all followers, assume, for a moment, that a normal bell curve represents
274
the distribution of followers for whom a particular leadership model is effective for.
This would suggest that any pre-specified leadership model would not be “effective”
for about 31 % of followers. Obviously, this is an overly simplistic view of the
problem, but it helps highlight the theoretical and practical issues of specifying one
model of leading for all. Instead, conceptualising leadership as a relationship
between leaders and followers appears a more promising approach to gaining better
knowledge and insight (see also Shamir, 2007; Uhl-Bien et al., 2014). Studies 3 and
4 provide evidence that seeing leadership as a relationship can be fruitful for
extending leadership and followership theory and opens up new lines of research.
Specifically, the current research provides evidence that leadership can be
conceptualised as a complex interplay between the cognitive-emotional tendencies of
leaders and followers and their respective beliefs about what and how an ideal leader
or follower should think and behave.
One way of understanding the complex interplay between the cognitive-
emotional tendencies of leaders and followers and their respective beliefs is via an
integration of connectionist ILT and IFT models with an individual’s overall implicit
theory of reality as described in CEST (Epstein, 1994; 2003). An implicit theory of
reality is a hierarchical cognitive structure of interrelated schemas. At the top of this
hierarchy are the more general, abstract schemas (e.g. the world is safe, the self is
worthy). At the bottom are situation specific schemas (Epstein, 2003). Connectionist
models of ILTs and IFTs (see Brown & Lord, 2001; Hanges et al., 2000; Lord et al.,
2001), at their most basic, understand ILTs and IFTs as existing within the pattern of
knowledge unit activations and inhibitions. This pattern depends on contextual
factors including the organisational culture, the leader, the follower, and the current
task (Lord et al., 2001). This description is compatible to the situation specific
275
schemas described in CEST (Epstein, 2003). Specifically, ILTs and IFTs can be
understood as instances of situation specific schemas that exist in the wider implicit
theory of reality. Furthermore, a recognition that characteristics of followers and
leaders influence the activation pattern in the connectionist models appears
theoretically congruent with the more abstract and broader schemas of followers’
and leaders’ implicit theory of reality. Therefore, these theoretical paradigms appear
to be mutually compatible.
Other Theoretical Implications
There are other theoretical implications of the current research. First, the way
in which leadership is conceptualised has potential implications for understanding
the consequences of high and low-quality leader-follower relationships on
organisational outcomes. From the perspective of the current research, leaders’ and
followers’ perceptions of their relationships are dependent on the interplay between
their experiences during leader-follower interactions, their cognitive-emotional
preferences and tendencies, and their implicit beliefs. As leaders and followers look
for dyadic partners who are congruent with their cognitive-emotional tendencies and
implicit beliefs, the implication is that subjectively reported “high-quality
relationships” or “low-quality relationships” may, from an objective perspective,
look very different across each dyad (e.g. different combinations of congruent or
incongruent leader-follower cognitive-emotional tendencies and beliefs).
As a consequence, it cannot be assumed that a subjectively reported high or
low-quality relationship will necessarily result in positive or negative organisational
outcomes implicitly assumed in LMX research (e.g. Dulebohn et al., 2012; Dunegan
276
et al., 2002; Erdogan & Liden, 2002; Graen & Uhl-Bien, 1995; Judge & Ferris,
1993; Liden et al., 1993). Instead, there may be some combinations of leader-
follower cognitive-emotional tendencies and beliefs, which may result in both
leaders and followers subjectively reporting a high-quality relationship that actually
objectively results in poorer organisational outcomes. Alternatively, there may be
situations where perceived low-quality leader-follower relationships may actually
contribute to positive team and organisational performance. For example, study 3
demonstrates that followers who seek highly dedicated leaders tend to report higher
levels of relationship quality when their leader is proactive and optimistic. In
contrast, followers who do not want highly dedicated leaders are more likely to
report having a poor relationship with these types of leaders. Instead, they tend to
report having better relationships with less proactive and optimistic leaders. From the
perspective of these dyadic combinations, while the former relationship may be
subjectively experienced negatively from the perspective of the follower, from an
organisational perspective, these relationships may be more beneficial (e.g. leaders
are proactive in ensuring followers complete their tasks). In contrast, while the latter
relationship may be subjectively experienced positively from the followers’
perspective, these relationships may actually be less beneficial for the organisation
(e.g. leaders are more laissez-faire towards ensuring followers complete tasks).
Consequently, future research may benefit by distinguishing relationship quality in
terms of a subjective perception or some objective outcome.
The second wider theoretical implication is the potential application of this
general theoretical framework to other relationship contexts (i.e. beyond leadership).
Specifically, it may be useful to integrate CEST (Epstein, 1998; 2001; 2003; 2014)
with other implicit role based theories to understand how these may be influencing a
277
relationship in a particular context. For example, perceptions of marriage quality
may be, in part, dependent on how each believes their partner should act in their role.
Specifically, similar to how IFTs and ILTs influence the observers’ perceptions of
followers and leaders respectively (Lord & Maher, 1990; Sy, 2010), each partner in
the relationship is likely hold an ideal image for their dyadic counterpart (e.g. see
DeHart, Pelham, & Murray, 2004; Knee, Nanayakkara, Vietor, Neighbors, &
Patrick, 2001). These differences would correspondingly influence what they attend
to, encode, and how they judge and evaluate the behaviours of their dyadic
counterpart. From this perspective, new insights into relationships across different
contexts could be generated.
General Research Strengths and Limitations
One of the key methodological strengths across studies 1 to 4 was the
adoption of a two-stage research design. These two stages were separated by at least
two days. Furthermore, the surveys in each stage had response scales that differed
from each other (i.e. 5-point and 10-point in stage 1 and 5-point and 7-point in stage
2). These design factors should reduce issues associated with common method bias
(see Podsakoff et al., 2003 for a detailed discussion). Common method bias was also
investigated using statistical analyses in studies 1 and 2. In these cases, limited
evidence of common-method bias was found. The relationships observed across
these studies, therefore, should be more likely to reflect the actual relationships in
the population rather than an artefact of how the data was collected.
A consideration of the strengths and limitations in adopting CEST (Epstein,
1973; 1994; 2003) instead of other dual-processing models (see Kahnman, 2003;
278
Smith & DeCoster’s, 2000; Strack & Deutsch, 2004) in the theorising of the current
research also needs to be made. As noted in the introduction, all dual-processing
theories essentially propose humans have two information processing systems
(Evans, 2008; Evans & Over, 1996; Gawronski & Creighton, 2013; Kahneman 2011;
Kahneman & Frederick 2002). The first system is typically seen high capacity and
processes information implicitly or automatically at the unconscious level. In
contrast, the second system is lower in its capacity and is engaged in more explicit or
controlled processes. In the case of CEST (Epstein, 1973; 1994; 2003), it is clear that
experiential system aligns with the first system described, and the rational system
aligns with the second system (see also Evans, 2008). Where CEST strengths lie
compared to these models is that the dual-processing aspect is placed in the wider
context of a global theory of personality (Epstein, 2003; Pacini & Epstein, 1999). In
addition, the core strength in relation to the current research was the fact that it was
explicitly compatible with psychodynamic theories of leadership and followership,
but within a scientific defensible framework (Epstein, 1994; Frankish & Evans,
2009). The importance of emotions in the experiential system is also a feature that is
not typical in other dual processing theories (Evans, 2008). Yet, emotions have been
increasingly recognised as important in the workplace (e.g. Bono et al., 2007; Brief
& Weiss, 2002).
While CEST clearly has strengths, especially in relation to the current
research, it does have specific weaknesses for interpreting some aspects of the
current research. In particular, unlike more recent dual-processing theories (e.g. see
Kahnman, 2003; Smith & DeCoster’s, 2000; Strack & Deutsch, 2004), CEST does
not specifically outline how the experiential operates at the cognitive level. For
example, RIM (Strack & Deutsch, 2004) draws upon recent conceptualisations of
279
associative-network models (e.g. Smith, 1998), to describe how the IS operates.
Specifically, within the RIM, the links between knowledge elements have different
strengths, which are based upon prior learning (i.e. elements that occur in close
temporal or spatial proximity with have links that are strengthened). Once an
element is activated via some environmental input, other elements that are linked
also become activated. The pattern of this activation depends on the nature and
strengths of these links resulting in cognitive structures (i.e. schemas) within RIM’s
IS. CEST only provides a more simple description of how the experiential system
(e.g. see Epstein, 2003), although this is somewhat compatible with how RIM states
the IS system operates.
One of the other limitations of CEST is the focus primarily on the
experiential system (Epstein, 2003). Consequently, there is subsequently less focus
on the rational system. At the outset of this research, this was an intentional
imbalance as, from the perspective of CEST (Epstein, 2001; 2003), research
assessing the relationships between the rational components of personality and ILT’s
have already undertaken in the literature (e.g. Keller, 1999).
In addition to the small sample sizes utilised in studies 3 and 4, one of the
main research limitations was the use of explicit measures to assess implicit
constructs (see Uhlmann et al., 2012 for a detailed discussion). Specifically, there is
a divergence between the construct of interest and how it is measured. It is important
to recognise though that Epitropaki and colleagues (2013) state that both explicit and
implicit measures are important for furthering our understanding of how these beliefs
influence leader-follower relationships. Indeed, according to CEST, while both the
experiential and rational systems are seen to be independent, they do operate in
parallel and interact with each other to influence the processing that occurs (Epstein,
280
1998; Epstein et al., 1996). Consequently, there is a reason to believe that, despite
the methodological mismatch, what individuals explicitly report as their ideal
follower or leader will bear a resemblance to their implicit schemas. However, it is
suggested that future research should seek to use implicit measures. It may be
expected that similar relationships would emerge and may be even stronger given
that implicit leadership and followership theories would exist in the preconscious
experiential system. Consequently, the experiential aspects of personality should be
more proximal antecedents.
Finally, in general, all the studies presented adopted analysed one dependent
variable at a time. In contrast, a pattern approach would consider the internal
coherent and consistent pattern of ILT variables that exist within followers and
provide a more holistic understanding of follower ILTs (e.g. see Foti & Hauenstein,
2007). Consequently, future research may wish to adopt a pattern approach. It is
important to note that the pattern approach is considered as a complementary
approach to the variable approach, not as a substitute. Indeed, Foti and Hauenstein
(2007) state that “the pattern approach can complement the variable approach” (p.
353). Consequently, results presented across the four presented studies are still
important for extending leadership research.
Practical Implications
There are a number of practical implications that emerge from the current
research for employee selection, team composition, and for how leadership
development should be approached. From an employee selection perspective, the
research demonstrates that both information processing style and constructive
281
thinking will, on average, influence any candidates’ ability for building and
maintaining high-quality leader-follower relationships. This is an important outcome
as high-quality leader-follower relationships contribute to lower employee turnover,
increased organisational commitment, job performance and satisfaction, and
organisational citizenship (Dulebohn et al., 2012; Ilies et al., 2007) which ultimately
impact the bottom line of organisations (see Cascio, 1991; Koys, 2001). However,
there are no guarantees that the candidates’ traits and beliefs will be congruent with
that of existing employees (i.e. who the candidate will lead or follow depending on
the role they are applying for). Consequently, selection processes could also consider
the match of candidates’ traits, behaviours, and beliefs to the existing employees
where the candidate will be placed. In other words, the best outcomes for the leader-
follower relationship will be where a consideration of the congruence between
candidates’ and the existing leader’s or follower’s information style, constructive
thinking, and implicit beliefs are made.
There are similar implications for team composition decisions. Team
composition refers to how individual characteristics in a team are arranged to bring
about increased effectiveness on team processes and outcomes (Bell, 2007). As
stated by Marks, Mathieu, and Zaccaro (2001), members of a team must engage in
“interdependent acts that convert inputs to outcomes through cognitive, verbal, and
behavioral activities directed toward organizing taskwork to achieve collective
goals” (p. 357). Similar to employee selection, the current research provides insight
into leader and follower characteristics that will facilitate group cohesiveness,
processes and outcomes. Specifically, team members can be selected based on how
congruent their cognitive-emotional tendencies and beliefs about leaders and
followers are to other team members. For example, the research would suggest that
282
the most congruent followers for a team which has a leader with high levels of
imagination, and emotional and behavioural coping, would be most effective with
those followers who endorse high levels of leader sensitivity and dedication.
The nature of congruence as described above for employee selection and
team composition decisions, therefore, extends the person-supervisor fit research
(see Kristof-Brown et al., 2005). Specifically, it provides insight into which leader
and follower traits and beliefs may be most congruent for influencing leader-
follower relationships. However, coupled with a relationship based perspective of
leadership (see also Lord & Brown, 2001; Shamir, 2007; Uhl-Bien et al., 2014), the
use of CEST (Epstein, 1998; 2001; 2003) as a framework for understanding how
these aspects of personality and beliefs may interact together has particularly
important implications for how leadership development is approached.
There have been many practitioners who have written about the limitations of
current mainstream approaches to leader development (e.g. Day, 2001; Hosking,
2002; Schyns et al., 2011; 2012; Velsor & McCauley, 2010). Specifically, they
generally argue that what are typically referred to, or marketed as, leadership
training programmes are actually leader training programmes (Day, 2001). Such
programmes seek to increase the knowledge, skills and abilities of individuals who
hold a leader role (Day, 2001; Velsor & McCauley, 2010). However, from a
relationship perspective, such an approach to improving leadership is problematic.
Specifically, they fail to understand the active and dynamic impacts that followers
have on the leadership process. The findings in the current research support the
notion that current leader development programmes should evolve to actual
leadership development programmes before they become truly effective (see e.g.
Day, 2001; Hosking, 2002; Schyns et al., 2011; 2012; Velsor & McCauley, 2010).
283
Similar to psychodynamic approaches to improving leadership effectiveness
(e.g. see Bornstein, 2003; Kroeger et al., 2002; Stech, 2007), these leadership
programmes would incorporate both leaders and followers. Specifically, they would
provide feedback and insight into both leaders’ and followers’ personalities, in
particular, the experiential aspects due to its significant influence over everyday
cognitions and behaviours (see Epstein, 1998; 2003) and how these may impact on
their working relationship and, subsequently, individual, team, and organisational
outcomes. These training programmes though should seek to go further than simply
understanding the impact of each other’s information processing style and
constructive thinking. Specifically, they should also seek to increase leaders’ and
followers’ understanding of how these elements of personality influence what they
subsequently want in leaders and followers (i.e. their IIFTs and IILTs) to ultimately
influence their working relationship. Indeed, Schyns and colleagues (2011) describe
a technique for bringing leaders’ attention to their implicit leadership theories by
getting them to draw images they associate with leaders and then discussing these
images to help gain a better understanding of how their beliefs influence behaviours
and, subsequently, the leadership process. Such techniques could be integrated into a
holistic leadership development programme in which both leaders’ and followers’
personalities and implicit beliefs are explored.
Importantly for these leadership development programs is the fact that
followers’ and leaders’ information processing styles and constructive thinking can
be specifically targeted for change (see Cerni et al., 2010; Epstein, 1998; 2003).
Such changes are likely to change the nature of individuals’ IILTs and IIFTs.
Specifically, leader and follower interventions which seek to change leaders’ and
follower’ information processing style and constructive thinking, therefore, should,
284
over time, come to influence not only the ways in which they process information
generally but also what they look for in their ideal leader or follower. As noted in
previous chapters, Kruse and Sy (2011) have found that increasing positive affect
can correspondingly increase the expression of prototypical followership traits. In
contrast, negative affect increases the endorsement of anti-prototypical traits. The
transient nature of affect though suggests that such changes are likely to revert to
their original IFT.
Affect in CEST though is conceptualised as an outcome of the initial
cognitive interpretative processes in the experiential system (see Figure 1.1 in
Chapter 1; Epstein, 1998; 2003). If leadership development programmes therefore
target such processes for change, it suggests that both leaders and followers can be
developed. Specifically, cognitive processes can be changed resulting in less intense
and frequent negative affective experiences in addition to more intense and frequent
positive affective experiences. Such changes should result in longer-lasting changes
to both leaders’ and followers’ IIFTs and IILTs respectively. Consequently, these
changes should improve the leader-follower relationship. In support, research
demonstrates the Pygmalion effects of leaders who hold positive expectations of
their followers on actual follower performance (Eden, 1990; Whiteley et al., 2012).
Concluding Remarks
Leadership is clearly a highly complex phenomenon. The current research
suggests that such complexity can be better understood by adopting a more
integrative and holistic approach to studying leadership, which can provide greater
insights and understanding. Specifically, by taking such an approach, three
285
limitations in the leadership literature were overcome. Studies 1 and 2 demonstrate
that the cognitive-emotional tendencies, especially those occurring unconsciously, of
leaders and followers, influence the beliefs that they hold about what an ideal
follower or leader should be respectively. These images of the ideal leader and
follower generally align with individuals’ cognitive-emotional tendencies. As
demonstrated in studies 3 and 4, both the cognitive-emotional tendencies of leaders
and followers along with their implicit beliefs subsequently influence how they come
to perceive their leader-follower relationships. Such a perspective has significant
theoretical and practical implications for how we approach leadership, followership,
leaders and followers.
286
References
Abdi, H., & Williams, L. J. (2010). Principal component analysis. Wiley
Interdisciplinary Reviews: Computational Statistics, 2(4), 433-459.
doi:10.1002/wics.101
Ackerman, P. L., & Heggestad, E. D. (1997). Intelligence, personality, and interests:
Evidence for overlapping traits. Psychological Bulletin, 121(2), 219-245.
doi:10.1037/0033-2909.121.2.219
Adam, H., & Galinsky, A. D. (2012). Enclothed cognition. Journal of Experimental
Social Psychology, 48(4), 918-925. doi:10.1016/j.jesp.2012.02.008
Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, N. (1950). The
Authoritarian Personality. New York: Harper and Brothers.
Aeron, S., & Pathak, S. (2013). Relationship between team member personality and
team cohesion: An exploratory study in IT industry. Management and
Labour Studies, 37(3), 267-282. doi:10.1177/0258042x13484839
Aguinis, H., Gottfredson, R. K., & Culpepper, S. A. (2013). Best-practice
recommendations for estimating cross-level interaction effects using
multilevel modeling. Journal of Management, 39(6), 1490-1528.
doi:10.1177/0149206313478188
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. Thousand Oaks, CA: SAGE.
Alcorn, D. S. (1992). Dynamic followership: Empowerment at work. Management
Quarterly, 33(1), 9-13.
Allen, P. J., & Bennett, K. (2010). PASW statistics by SPSS: A practical guide:
Version 18.0. Melbourne, VIC: Cengage Learning.
Amabile, T. M. (1998). How to kill creativity. Harvard Business Review, 76(5), 76-
287
87.
Amabile, T. M., Barsade, S. G., Mueller, J. S., & Staw, B. M. (2005). Affect and
creativity at work. Administrative Science Quarterly, 50(3), 369-403.
doi:10.2189/asqu.2005.50.3.367
Andersson, L. M., & Bateman.T.S. (1997). Cynicism in the workplace: Some causes
and effects. Journal of Organizational Behavior, 18(5), 449-469.
doi:10.1002/(SICI)1099-1379(199709)18:5<449::AID-JOB808>3.0.CO;2-O
Anseel, F., & Lievens, F. (2007). The long-term impact of the feedback environment
on job satisfaction: A field study in a Belgian context. Applied Psychology:
An International Review, 56(2), 254-266. doi:10.1111/j.1464-
0597.2006.00253.x
Antonakis, J., Avolio, B. J., & Sivasubramaniam, N. (2003). Context and leadership:
An examination of the nine-factor full-range leadership theory using the
Multifactor Leadership Questionnaire. The Leadership Quarterly, 14(3), 261-
295. doi:10.1016/S1048-9843(03)00030-4
Antonioni, D. (1998). Relationship between the big five personality factors and
conflict management styles. International Journal of Conflict Management,
9(4), 336-355. doi:10.1108/eb022814
Argyris, C., & Schön, D. (1978). Organizational learning: A theory of action
perspective. Massachusetts: Addison-Wesley Publishing Company.
Ashforth, B. E., Sluss, D. M., & Saks, A. M. (2007). Socialization tactics, proactive
behavior, and newcomer learning: Integrating socialization models. Journal
of Vocational Behavior, 70(3), 447-462. doi:10.1016/j.jvb.2007.02.001
Atwater, L. E., & Yammarino, F. J. (1993). Personal attributes as predictors of
superiors' and subordinates' perceptions of military academy leadership.
288
Human Relations, 46(5), 645-668. doi:10.1177/001872679304600504
Aulakh, P. S., & Gencturk, E. F. (2000). International principal–agent relationships:
Control, governance and performance. Industrial Marketing Management,
29(6), 521-538. doi:10.1016/S0019-8501(00)00126-7
Avolio, B. J. (1999). Full leadership development: Building the vital forces in
organizations. Thousand Oaks, CA: SAGE.
Avolio, B. J., & Bass, B. M. (1991). The full range leadership development
programs: Basic and advanced manuals. Binghamton, NY: Bass, Avolio &
Associates.
Avolio, B. J., & Reichard, R. J. (2008). The rise of Authentic Followership. In E. R.
Ronald, I. Chaleff, & J. Lipman-Blumen (Eds.), The Art of Followership:
How Great Followers Creat Great Leaders and Organizations (pp. 325-337).
San Francisco, CA: Jossey-Bass.
Avolio, B. J., Walumbwa, F. O., & Weber, T. J. (2009). Leadership: Current
theories, research, and future directions. Annual Review of Psychology, 60,
421-449. doi:10.1146/annurev.psych.60.110707.163621
Awamleh, R., & Gardner, W. L. (1999). Perceptions of leader charisma and
effectiveness: The effects of vision content, delivery, and organizational
performance. The Leadership Quarterly, 10(3), 345-373. doi:10.1016/S1048-
9843(99)00022-3
Ayman, R. (1993). Leadership perception: The role of gender and culture. In M. C.
Martin & R. Ayman (Eds.), Leadership theory and research: Perspectives
and directions (pp. 137-166). San Diego, CA: Academic Press.
Ayman-Nolley, S., & Ayman, R. (2005). Children's implicit theory of leadership. In
B. Schyns & J. R. Meindl (Eds.), Implicit leadership theories: Essays and
289
explorations (pp. 189–233). Greenwich, CT: Information Age Publishing.
Badin, I. J. (1974). Some moderator influences on relationships between
consideration, initiating structure and organizational criteria. Journal of
Applied Psychology, 59(3), 380-382. doi:10.1037/h0036768
Baer, M., & Frese, M. (2003). Innovation is not enough: Climates for initiative and
psychological safety, process innovations, and firm performance. Journal of
Organizational Behavior, 24(1), 45-68. doi:10.1002/job.179
Baker, S. D. (2007). Followership: The theoretical foundation of a contemporary
construct. Journal of Leadership & Organizational Studies, 14(1), 50-60.
doi:10.1177/0002831207304343
Baker, S. D., Anthony, E. L., & Stites-Doe, S. A. (2015). Most admired leader/most
admired follower. Organization Management Journal, 12(1), 23-33.
doi:10.1080/15416518.2014.969366
Bargh, J. A. (1982). Attention and automaticity in the processing of self-relevant
information. Journal of Personality and Social Psychology, 43(3), 425-436.
doi:10.1037/0022-3514.43.3.425
Bargh, J. A. (2006). What have we been priming all these years? On the
development, mechanisms, and ecology of nonconscious social behaviour.
European journal of social psychology, 36(2). doi:10.1002/ejsp.336
Barrett, L. F., & Pietromonaco, P. R. (1997). Accuracy of the five-factor model in
predicting perceptions of daily social interactions. Personality and Social
Psychology Bulletin, 23(11), 1173-1187. doi:10.1177/01461672972311005
Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and
job performance: A meta-analysis. Personnel Psychology, 44(1), 1-26.
Bass, B. M. (1985). Leadership and performance beyond expectations. London:
290
Collier Macmillan.
Bass, B. M. (1990). Bass & Stogdill's handbook of leadership: Theory, research &
managerial applications (3rd ed.). New York: Free Press.
Bass, B. M., & Avolio, B. J. (1989). Potential biases in leadership measures: How
prototypes, leniency, and general satisfaction relate to ratings and rankings of
transformational and transactional leadership constructs. Educational and
Psychological Measurement, 49(3), 509-527.
doi:10.1177/001316448904900302
Bass, B. M., & Avolio, B. J. (1990). Transformational leadership development:
Manual for the multifactor leadership questionnaire. Palo Alto, CA:
Consulting Psychologists Press.
Bass, B. M., & Avolio, B. J. (1993). Transformational leadership: A response to
critiques. In M. M. Chemers & R. Ayman (Eds.), Leadership theory and
research: Perspectives and directions (pp. 49-80). San Diego, CA: Academic
Press.
Bass, B. M., & Avolio, B. J. (1994). Transformational leadership and organizational
culture. International Journal of Public Administration, 17(3-4), 541-554.
doi:10.1080/01900699408524907
Bass, B. M., & Avolio, B. J. (1997). Full range leadership development: Manual for
the Multifactor Leadership Questionnaire. Palo Alto, CA: Mindgarden.
Basu, R., & Green, C. (1997). Leader‐member exchange and transformational
leadership: An empirical examination of innovative behaviors in
leader‐member dyads. Journal of Applied Social Psychology, 27(6), 477-499.
doi:10.1111/j.1559-1816.1997.tb00643.x
Bauer, D. J., Preacher, K. J., & Gil, K. M. (2006). Conceptualizing and testing
291
random indirect effects and moderated mediation in multilevel models: New
procedures and recommendations. Psychological Methods, 11(2), 142-163.
doi:10.1037/1082-989X.11.2.142
Bauer, T.N., & Erdogan, B. (2016). Leader-Member Exchange (LMX) Theory: An
introduction and Overview. In T.N. Bauer & B. Erdogan (Eds.). The Oxford
Handbook of Leader-Member Exchange, (pp. 3-8). New York: Oxford
University Press.
Bauer, T. N., & Green, S. G. (1996). Development of leader-member exchange: A
longitudinal test. The Academy of Management Journal, 39(6), 1538-1567.
doi:10.2307/257068
Bell, S. T. (2007). Deep-level composition variables as predictors of team
performance: A meta-analysis. Journal of Applied Psychology, 92(3), 595-
615. doi:10.1037/0021-9010.92.3.595
Belschak, F. D., Hartog, D. N., & Fay, D. (2010). Exploring positive, negative and
context-dependent aspects of proactive behaviours at work. Journal of
Occupational and Organizational Psychology, 83(2), 267-273.
doi:10.1348/096317910x501143
Bennis, W. (2008). Introduction. In E. R. Ronald, I. Chaleff, & J. Lipman-Blumen
(Eds.), The art of followership: How great followers create great leaders and
organizations (pp. xxiii-xxvii). San Francisco, CA: Jossey-Bass.
Bentler, P. M., & Chou, C. P. (1987). Practical issues in structural modeling.
Sociological Methods & Research, 16(1), 78-117.
doi:10.1177/0049124187016001004
Berg, D. N. (1998). Resurrecting the muse: Followership in organizations. In E. B.
Klein, F. Gabelnick, & P. Herr (Eds.), The psychodynamics of leadership (pp.
292
27-52). Madison, CT: Psychosocial Press.
Berne, E. (1961). Transactional analysis in psychotherapy: A systematic individual
and social psychiatry. New York: Grove Press.
Bernerth, J. B., Armenakis, A. A., Feild, H. S., Giles, W. F., & Walker, H. J. (2007).
Leader–member social exchange (LMSX): development and validation of a
scale. Journal of Organizational Behavior, 28(8), 979-1003.
doi:10.1002/job.443
Biddle, B. (1979). Role theory: Expectations, identities and behaviors. New York:
Academic Press
Bjugstad, K., Thach, E., Thompson, K., & Morris, A. (2006). A fresh look at
followership: A model for matching followership and leadership styles.
Journal of Behavioral and Applied Management, 7(3), 304-311.
Blanchard, K. H. (1985). The situational leader. New York: Warner Books.
Blanchard, K. H., Zigarmi, P., Zigarmi, D., & Dowdy, K. D. (1985). Leadership and
the one minute manager. London: Fontana/Collins.
Bland, J.M., & Altman, D.G. (1997). Statistics notes: Cronbach's alpha. BMJ. British
Medical Journal (Clinical research ed.), 314(7080), 572. Retrieved from
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2126061/pdf/9055718.pdf
Blass, T. (2009). From New Haven to Santa Clara: A historical perspective on the
Milgram obedience experiments. American Psychologist, 64(1), 37-45.
doi:10.1037/a0014434
Blau, P. M. (1964). Exchange and power in social life. New Brunswick, NJ:
Transaction Publishers.
Bledow, R., Rosing, K., & Frese, M. (2013). A dynamic perspective on affect and
creativity. Academy of Management Journal, 56(2), 432-450.
293
doi:10.5465/amj.2010.0894
Boies, K., & Howell, J. M. (2006). Leader–member exchange in teams: An
examination of the interaction between relationship differentiation and mean
LMX in explaining team-level outcomes. The Leadership Quarterly, 17(3),
246-257. doi:10.1016/j.leaqua.2006.02.004
Bold. (n.d.). Roget's 21st Century Thesaurus, Third Edition. Retrieved May 20,
2015, from Thesaurus.com website: http://www.thesaurus.com/browse/bold
Bolino, M. C. (1999). Citizenship and impression management: Good soldiers or
good actors? The Academy of Management Review, 24(1), 82-98.
doi:10.5465/AMR.1999.1580442
Bolino, M. C., & Turnley, W. H. (2003). Counternormative impression management,
likeability, and performance ratings: the use of intimidation in an
organizational setting. Journal of Organizational Behavior, 24(2), 237-250.
doi:10.1002/job.185
Bono, J.E., Foldes, H.J., Vinson, G., Muros, J.P. (2007). Workplace emotions: The
role of supervision and leadership. Journal of Applied Psychology, 92(5),
1357-1367. doi:10.1037/0021-9010.92.5.1357
Boora, S., & Shanti. (2010). Personality and conflict resolution styles. Paper
presented at the Violence Peace and Conflict Resolution, Chandigarh, India.
http://globalvisionpub.com/globaljournalmanager/pdf/1392359698.pdf
Borman, W. C., Penner, L. A., Allen, T. D., & Motowidlo, S. J. (2001). Personality
predictors of citizenship performance. International Journal of Selection and
Assessment, 9(1-2), 52-69. doi:10.1111/1468-2389.00163
Bornstein, R. F. (2003). Psychodynamic models of personality. In T. Millon & M. J.
Lerner (Eds.), Handbook of Psychology (Vol. 5, pp. 117-134). Hoboken, NJ:
294
John Wiley & Sons.
Botero, I. C., & Dyne, L. V. (2009). Employee voice behavior interactive effects of
LMX and power distance in the United States and Colombia. Management
Communication Quarterly, 23(1), 84-104. doi:10.1177/0893318909335415
Botwin, M. D., Buss, D. M., & Shackelford, T. K. (1997). Personality and mate
preferences: Five factors in mate selection and marital satisfaction. Journal of
Personality, 65(1), 107-136. doi:10.1111/j.1467-6494.1997.tb00531.x
Bouchard, G., Lussier, Y., & Sabourin, S. (1999). Personality and marital
adjustment: Utility of the five-factor model of personality. Journal of
Marriage and the Family, 61(3), 651-660. doi:10.2307/353567
Bowlby, J. (1969). Attachment. Attachment and loss: Vol. 1. Loss. New York: Basic
Books.
Bowlby, J. (1980). Loss: Sadness & depression. London: Hogarth Press.
Bradley, J. H., & Hebert, F. J. (1997). The effect of personality type on team
performance. The Journal of Management Development, 16(5), 337-353.
Brief, A.P., & Weiss, H.M. (2002). Organizational Behavior: Affect in the
Workplace. Annual Review of Psychology, 52(1), 279-307.
doi:10.1146/annurev.psych.53.100901.135156
Brown, D. (1991). Human universals. Boston, MA: McGraw-Hill.
Brown, S. P. (1996). A meta-analysis and review of organizational research on job
involvement. Psychological Bulletin, 120(2), 235-255. doi:10.1037/0033-
2909.120.2.235
Bryant, F. B., & Yarnold, P. R. (1995). Principal-components analysis and
exploratory and confirmatory factor analysis. In L. G. Grimm & P. R.
Yarnold (Eds.), Reading and understanding multivariate statistics (pp. 99-
295
136). Washington, DC: American Psychological Association.
Bucci, W. (1997). Psychoanalysis and cognitive science: A multiple code theory.
New York: Guildford Press.
Burns, J. M. (1978). Leadership. New York: Harper & Row.
Burris, E. R. (2012). The risks and rewards of speaking up: Managerial responses to
employee voice. Academy of Management Journal, 55(4), 851-875.
doi:10.5465/amj.2010.0562
Bushman, B. J., & Baumeister, R. F. (1998). Threatened egotism, narcissism, self-
esteem, and direct and displaced aggression: Does self-love or self-hate lead
to violence? Journal of Personality & Social Psychology, 75(1), 219-229.
doi:10.1037/0022-3514.75.1.219
Buunk, B. P., Doosje, B. J., Jans, L. G. J. M., & Hopstaken, L. E. M. (1993).
Perceived reciprocity, social support, and stress at work: The role of
exchange and communal orientation. Journal of Personality & Social
Psychology, 65(4), 801-811. doi:10.1037/0022-3514.65.4.801
Calder, B. J. (1977). An attribution theory of leadership. In M. S. Barry & G. R.
Salancik (Eds.), New directions in organizational behavior (pp. 179-204).
Oxford: St. Clair.
Cameron, K. S. (2008). Paradox in positive organizational change. The Journal of
Applied Behavioral Science, 44(1), 7-24. doi:10.1177/0021886308314703
Caplan, R. D. (1983). Person-environment fit: Past, present, and future. In C. Cooper
(Ed.), Stress research: New directions for the 1980s (Vol. 35-78). London:
Wiley.
Caplan, R. D. (1987). Person-environment fit theory and organizations:
Commensurate dimensions, time perspectives, and mechanisms. Journal of
296
Vocational Behavior, 31(3), 248-267. doi:10.1016/0001-8791(87)90042-X
Carsten, M. K., & Uhl-Bien, M. (2012). Follower beliefs in the co-production of
leadership. Zeitschrift für Psychologie, 220(4), 210-220. doi:10.1027/2151-
2604/a000115
Carsten, M. K., Uhl-Bien, M., West, B. J., Patera, J. L., & McGregor, R. (2010).
Exploring social constructions of followership: A qualitative study. The
Leadership Quarterly, 21(3), 543-562. doi:10.1016/j.leaqua.2010.03.015
Carver, C. S., Antoni, M. H., & Scheier, M. F. (1985). Self-consciousness and self-
assessment. Journal of Personality & Social Psychology, 48(1), 117-124.
doi:10.1037/0022-3514.48.1.117
Cascio, W. F. (1991). Costing human resources (3rd ed.). Boston: PWS-KENT
Publishing Company.
Caughlin, J. P., Huston, T. L., & Houts, R. M. (2000). How does personality matter
in marriage? An examination of trait anxiety, interpersonal negativity, and
marital satisfaction. Journal of Personality & Social Psychology, 78(2), 326-
336. doi:10.1037/0022-3514.78.2.326
Cerni, T., Curtis, G. J., & Colmar, S. H. (2008). Information processing and
leadership styles: Constructive thinking and transformational leadership.
Journal of Leadership Studies, 2(1), 60-73. doi:10.1002/jls.20049
Cerni, T., Curtis, G. J., & Colmar, S. H. (2010). Executive coaching can enhance
transformational leadership. International Coaching Psychology Review,
5(1), 81-85.
Cerni, T., Curtis, G. J., & Colmar, S. H. (2012). Cognitive‐experiential self theory
and conflict‐handling styles: Rational and constructive experiential systems
are related to the integrating and compromising conflict‐handling styles.
297
International Journal of Conflict Management, 23(4), 362-381.
doi:10.1108/10444061211267263
Chaiken, S., & Trope, Y. (1999). Dual-process theories in social psychology. New
York: Guilford Press.
Chakraborti, C., Boonyasai, R. T., Wright, S. M., & Kern, D. E. (2008). A
systematic review of teamwork training interventions in medical student and
resident education. Journal of General Internal Medicine, 23(6), 846-853.
doi:10.1007/s11606-008-0600-6
Chaleff, I. (1995). The courageous follower: Standing up to and for our leaders. San
Francisco: Berrett-Koehler.
Chaleff, I. (2003). The courageous follower (2nd ed.). San Francisco: Berrett-
Koehler.
Chaleff, I. (2008). Creating new ways of following. In R. E. Riggio, I. Chaleff, & J.
Lipman-Blumen (Eds.), The art of followership: How great followers create
great leaders and organizations (pp. 67-88). San Francisco, CA: Jossey-
Bass.
Chaleff, I. (2009). The courageous follower. San Francisco, CA: Berrett-Koehler
Publishers.
Chan, C. S., & Treacy, M. J. (1996). Resistance in multicultural courses: Student,
faculty, and classroom dynamics. The American Behavioral Scientist, 40(2),
212-221.
Cheek, J. M., & Melchior, L. A. (1990). Shyness, self-esteem, and self-
consciousness. In H. Leitenberg (Ed.), Handbook of social and evaluation
anxiety (pp. 78-82). New York: Springer.
Chemers, M. M. (1997). An integrative theory of leadership. Mahwah, NJ: Erlbaum.
298
Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., & Henrich, J. (2013). Two
ways to the top: Evidence that dominance and prestige are distinct yet viable
avenues to social rank and influence. Journal of Personality & Social
Psychology, 104(1), 103-125. doi:10.1037/a0030398
Cheung, M.F.Y. & Wong, C.S. (2011). Transformational leadership, leader support,
and employee creativity. Leadership & Organization Development Journal,
32(7), 656-672. doi:10.1108/01437731111169988
Chiaburi, D.S., Smith, T.A., Wang, J., & Zimmerman, R.D. (2014). Relative
importance of leader influences for subordinates' proactive behaviors,
prosocial behaviors, and task performance. Journal of Personnel Psychology,
13(2) 70-86. doi:10.1027/1866-5888/a000105
Chiricos, T., Barrick, K., Bales, W., & Bontrager, S. (2007). The labeling of
convicted felons and its consequences for recidivism. Criminology, 45(3),
547-581. doi:10.1111/j.1745-9125.2007.00089.x
Chong, E., & Wolf, H. (2009). Factors influencing followers’ perception of
organisational leaders. Leadership & Organization Development Journal,
31(5), 402-419. doi:10.1108/01437731011056434
Christner, C. A., & Hemphill, J. K. (1955). Leader behavior of B-29 commanders
and changes in crew members' attitudes toward the crew. Sociometry, 18(1),
82-87. doi:10.2307/2785831
Clarke, N., & Mahadi, N. (2015). Mutual recognition respect between leaders and
followers: Its relationship to follower job performance and well-being.
Journal of Business Ethics, 1-16. doi:10.1007/s10551-015-2724-z
Collinson, D. (2012). Prozac leadership and the limits of positive thinking.
Leadership, 8(2), 87-107. doi:10.1177/1742715011434738
299
Colquitt, J. A., Scott, B. A., & LePine, J. A. (2007). Trust, trustworthiness, and trust
propensity: A meta-analytic test of their unique relationships with risk taking
and job performance. Journal of Applied Psychology, 92(4), 909-927.
doi:10.1037/0021-9010.92.4.909
Comrey, A. L., & Lee, H. B. (1992). A first course in factor analysis. Hillsdale, NJ:
Lawrence Erlbaum.
Conger, J. A., & Kanungo, R. N. (1987). Toward a behavioral theory of charismatic
leadership in organizational settings. The Academy of Management Review,
12(4), 637-647. doi:10.5465/AMR.1987.4306715
Conger, J. A., & Kanungo, R. N. (1998). Charismatic leadership in organizations.
Thousand Oaks, CA: SAGE.
Costa, P. T., & McCrae, R. R. (1989). Reinterpreting the Myers-Briggs Type
Indicator from the perspective of the five-factor model of personality.
Journal of Personality, 57(1), 17-40. doi:10.1111/j.1467-
6494.1989.tb00759.x
Costa, P. T., & McCrae, R. R. (1992). Normal personality assessment in clinical
practice: The NEO Personality Inventory. Psychological Assessment, 4(1), 5-
13. doi:10.1037/1040-3590.4.1.5
Costa, P. T., & McCrae, R. R. (1997). Stability and change in personality
assessment: The revised NEO Personality Inventory in the year 2000.
Journal of Personality Assessment, 68(1), 86-94.
doi:10.1207/s15327752jpa6801_7
Crant, J. M. (2000). Proactive behavior in organizations. Journal of Management,
26(3), 435-462. doi:10.1177/014920630002600304
Cummings, G. G., MacGregor, T., Davey, M., Lee, H., Wong, C. A., Lo, E., Muisea,
300
M., & Stafford, E. (2010). Leadership styles and outcome patterns for the
nursing workforce and work environment: A systematic review. International
Journal of Nursing Studies, 47(3), 363-385.
doi:10.1016/j.ijnurstu.2009.08.006
Curtis, G. J., & Lee, M. W. H. (2013). Connecting cognitive-experiential self-
theory's information-processing styles with organisational-influencing tactics:
Rational thinkers are rational persuaders. Australian and New Zealand
Journal of Organisational Psychology, 6(e2), 1-11. doi:10.1017/orp.2013.2
Daly, J. A., & Stafford, L. (1984). Correlates and consequences of social-
communicative anxiety. In John. A. Daly & J. C. McCroskey (Eds.),
Avoiding communication: Shyness, reticence, and communication
apprehension (pp. 125–144). Beverly Hills, CA: SAGE.
Dansereau, F., Graen, G., & Haga, W. J. (1975). A vertical dyad linkage approach to
leadership within formal organizations: A longitudinal investigation of the
role making process. Organizational Behavior and Human Performance,
13(1), 46-78. doi:10.1016/0030-5073(75)90005-7
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a
multidimensional approach. Journal of Personality & Social Psychology,
44(1), 113-125. doi:10.1037/0022-3514.44.1.113
Day, D. R., & Stogdill, R. M. (1972). Leader behavior of male and female
supervisors: A comparative study. Personnel Psychology, 25(2), 353-360.
doi:10.1111/j.1744-6570.1972.tb01110.x
Day, D.V. (2001). Leadership development: A review in context. The Leadership
Quarterly, 11(4), 581-613. doi:10.1016/S1048-9843(00)00061-8
Day, D.V., & Miscenki, D. (2016). Leader-Member Exchange (LMX): Construct
301
evolution, contributions, and future prospects for advancing leadership
theory. In T.N. Bauer & B. Erdogan (Eds.). The Oxford Handbook of Leader-
Member Exchange, (pp. 3-8). New York: Oxford University Press.
Deal, J. J., & Stevenson, M. A. (1998). Perceptions of female and male managers in
the 1990s: Plus ça change... . Sex Roles, 38(3), 287-300.
doi:10.1023/A:1018741318216
DeBono, A., Shmueli, D., & Muraven, M. (2011). Rude and inappropriate: the role
of self-control in following social norms. Personality and Social Psychology
Bulletin, 37(1), 136-146. doi:10.1177/0146167210391478
DeGroot, T., Kiker, D. S., & Cross, T. C. (2000). A meta‐analysis to review
organizational outcomes related to charismatic leadership. Canadian Journal
of Administrative Sciences / Revue Canadienne des Sciences de
l'Administration, 17(4), 356-372. doi:10.1111/j.1936-4490.2000.tb00234.x
DeHart, T., Pelham, B., & Murray, S. L. (2004). Implicit dependency regulation:
Self-esteem, relationship closeness, and implicit evaluations of close others.
Social Cognition, 22(1), 126-146. doi:10.1521/soco.22.1.126.30986
Deluga, R. J. (1998). Leader-member exchange quality and effectiveness ratings: the
role of subordinate-supervisor conscientiousness similarity. Group &
Organization Management, 23(2), 189-216. doi:10.1177/1059601198232006
Denes-Raj, V., & Epstein, S. (1994). Conflict between intuitive and rational
processing: when people behave against their better judgment. Journal of
Personality & Social Psychology, 66(5), 819-829. doi:10.1037/0022-
3514.66.5.819
Denes-Raj, V., Epstein, S., & Cole, J. (1995). The generality of the ratio-bias
phenomenon. Personality & Social Psychology Bulletin, 21(10), 1083-1092.
302
doi:10.1177/01461672952110009
Derler, A., & Weibler, J. (2014). The ideal employee: Context and leaders’ implicit
follower theories. Leadership & Organization Development Journal, 35(5),
386-409. doi:10.1108/lodj-12-2012-0158
Derue, D. S., Nahrgang, J. D., Wellman, N., & Humphrey, S. E. (2011). Trait and
behavioral theories of leadership: An integration and meta‐analytic test of
their relative validity. Personnel Psychology, 64(1), 7-52.
doi:10.1111/j.1744-6570.2010.01201.x
Dessler, G., & Valenzi, E. R. (1977). Initiation of structure and subordinate
satisfaction: A path analysis test of path-goal theory. Academy of
Management Journal, 20(2), 251-259. doi:10.2307/255398
Deutsch, M. (1973). The resolution of conflict: Constructive and destructive
processes. New Haven, CT: Yale University Press.
Devine, P.G. (1989). Stereotypes and prejudice: Their automatic and controlled
components. Journal of Personality and Social Psychology, 56(1), 5-18.
doi:10.1037/0022-3514.56.1.5
Dienesch, R. M., & Liden, R. C. (1986). Leader-member exchange model of
leadership: A critique and further development. The Academy of Management
Review, 11(3), 618-634. doi:10.5465/AMR.1986.4306242
Dionne, S. D., Yammarino, F. J., Atwater, L. E., & Spangler, W. D. (2004).
Transformational leadership and team performance. Journal of
Organizational Change Management, 17(2), 177-193.
doi:10.1108/09534810410530601
Downey, H. K., Sheridan, J. E., & Slocum, J. W. (1975). Analysis of relationships
among leader behavior, subordinate job performance and satisfaction: A
303
path-goal approach. The Academy of Management Journal, 18(2), 253-262.
doi:10.2307/255528
Downey, H. K., Sheridan, J. E., & Slocum, J. W. (1976). The path-goal theory of
leadership: A longitudinal analysis. Organizational Behavior and Human
Performance, 16(1), 156-176. doi:10.1016/0030-5073(76)90011-8
Driskell, J. E., Goodwin, G. F., Salas, E., & O'Shea, P. G. (2006). What makes a
good team player? Personality and team effectiveness. Group Dynamics:
Theory, Research, and Practice, 10(4), 249-271. doi:10.1037/1089-
2699.10.4.249
Duchon, D., Dunegan, K. J., & Uhl-Bien, M. (1992). Examining the link between
leader member exchange and subordinate performance: The role of task
analyzability and variety as moderators. Journal of Management, 18(1), 59-
76. doi:10.1177/014920639201800105
Duchon, D., Green, S. G., & Taber, T. B. (1986). Vertical dyad linkage: A
longitudinal assessment of antecedents, measures, and consequences. Journal
of Applied Psychology, 71(1), 56-60. doi:10.1037/0021-9010.71.1.56
Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. R. (2012).
A meta-analysis of antecedents and consequences of Leader-Member
Exchange: Integrating the past with an eye toward the future. Journal of
Management, 38(6), 1715-1759. doi:10.1177/0149206311415280
Dunegan, K. J., Uhl-Bien, M., & Duchon, D. (2002). LMX and subordinate
performance: The moderating effects of task characteristics. Journal of
Business and Psychology, 17(2), 275-285. doi:10.1023/A:1019641700724
Duong, J. (2011). Leaders' conceptions and evaluations of followers as antecedents
of leadership style, leader-member exchange and employee outcomes.
304
Dissertation Abstracts International Section A: Humanities and Social
Sciences, 72(8-A), 179.
Dutton, J. E., & Ashford, S. J. (1993). Selling issues to top management. The
Academy of Management Review, 18(3), 397-428.
doi:10.5465/AMR.1993.9309035145
Dvir, T., & Shamir, B. (2003). Follower developmental characteristics as predicting
transformational leadership: A longitudinal field study. The Leadership
Quarterly, 14(3), 327-344. doi:10.1016/s1048-9843(03)00018-3
Dweck, C. S. (1975). The role of expectations and attributions in the alleviation of
learned helplessness. Journal of Personality & Social Psychology, 31(4),
674-685. doi:10.1037/h0077149
Eagly, A. H., & Carli, L. L. (2007). Women and the labyrinth of leadership. Harvard
Business Review, 85(9), 63-71.
Eagly, A. H., & Wood, W. (1991). Explaining sex differences in social behavior: A
meta-analytic perspective. Personality and Social Psychology Bulletin, 17(3),
306-315. doi:10.1177/0146167291173011
Eden, D. (1990). Pygmalion in management: Productivity as a self-fulfilling
prophecy. Lexington, MA: Lexington Books.
Eden, D. (1993). Leadership and expectations: Pygmalion effects and other self-
fulfilling prophecies in organizations. The Leadership Quarterly, 3(4), 271-
305. doi:10.1016/1048-9843(92)90018-B
Eden, D., & Leviatan, U. (1975). Implicit leadership theory as a determinant of the
factor structure underlying supervisory behavior scales. Journal of Applied
Psychology, 60(6), 736-741. doi:10.1037/0021-9010.60.6.736
Eden, D., & Leviatan, U. (2005). From implicit personality theory to implicit
305
leadership theory: A side-trip on the way to implicit organization theory. In
B. Schyns & J. R. Meindl (Eds.), Implicit leadership theories: Essays and
explorations. Greenwich, Conn: Information Age Publishing.
Ehrhart, M. G., & Klein, K. J. (2001). Predicting followers' preferences for
charismatic leadership: The influence of follower values and personality. The
Leadership Quarterly, 12(2), 153-179. doi:10.1016/S1048-9843(01)00074-1
Ellis, J. B., & Dyne, L. V. (2009). Voice and silence as observer reactions to
defensive voice: Predictions based on communication competence theory. In
Jerald Greenberg & M. S. Edwards (Eds.), Voice and Silence in
Organisations (pp. 37-61). Bingley, UK: Emerald Group Publishing.
Epitropaki, O. (2000). From ideal leaders to actual managers: A longitudinal study
of implicit leadership theories, leader-member exchanges, transformational
leadership and employee outcomes (Doctorate Doctoral), Cardiff University,
Unpublished Ph.D. Retrieved from
http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.271000
(uk.bl.ethos.271000 )
Epitropaki, O., & Martin, R. (2004). Implicit leadership theories in applied settings:
Factor structure, generalizability, and stability over time. Journal of Applied
Psychology, 89(2), 293-310. doi:10.1037/0021-9010.89.2.293
Epitropaki, O., & Martin, R. (2005). From ideal to real: A longitudinal study of the
role of implicit leadership theories on leader-member exchanges and
employee outcomes. Journal of Applied Psychology, 90(4), 659-676.
doi:10.1037/0021-9010.90.4.659
Epitropaki, O., Sy, T., Martin, R., Tram-Quon, S., & Topakas, A. (2013). Implicit
Leadership and Followership Theories “in the wild”: Taking stock of
306
information-processing approaches to leadership and followership in
organizational settings. The Leadership Quarterly, 24(6), 858-881.
doi:10.1016/j.leaqua.2013.10.005
Epstein, S. (1983). Aggregation and beyond: Some basic issues on the prediction of
behavior. Journal of Personality, 51(3), 360-392. doi:10.1111/j.1467-
6494.1983.tb00338.x
Epstein, S. (1987). Implications of cognitive self-theory for psychopathology and
psychotherapy. In N. Cheshire & H. Thomae (Eds.), Self, symptoms and
psychotherapy: Wiley series on methods in psychotherapy (pp. 43-58).
Oxford: John Wiley & Sons.
Epstein, S. (1992). Constructive thinking and mental and physical well-being. In L.
Montada, S.-H. Filipp, & M. J. Lerner (Eds.), Life crises and experiences of
loss in adulthood. (pp. 385-409). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Epstein, S. (1994). Integration of the cognitive and the psychodynamic unconscious.
American Psychologist, 49(8), 709-724. doi:10.1037/0003-066X.49.8.709
Epstein, S. (1998). Constructive thinking: The key to emotional intelligence.
Westport, CT: Praeger
Epstein, S. (2001). Constructive Thinking Inventory: Professional manual. (n.p):
Psychological Assessment Resources (PAR).
Epstein, S. (2003). Cognitive‐experiential self‐theory of personality. In I. B. Weiner
(Ed.), Handbook of psychology: Personality and social psychology (Vol. 5,
pp. 159-184). Hoboken, NJ: John Wiley & Sons.
Epstein, S. (2014). Cognitive-experiential theory: An integrative theory of
personality. Oxford: Oxford University Press.
307
Epstein, S., & Katz, L. (1992). Coping ability, stress, productive load, and
symptoms. Journal of Personality & Social Psychology, 62(5), 813-825.
doi:10.1037/0022-3514.62.5.813
Epstein, S., & Meier, P. (1989). Constructive thinking: A broad coping variable with
specific components. Journal of Personality & Social Psychology, 57(2),
332-350. doi:10.1037/0022-3514.57.2.332
Epstein, S., Pacini, R., Denes-Raj, V., & Heier, H. (1996). Individual differences in
intuitive–experiential and analytical–rational thinking styles. Journal of
Personality & Social Psychology, 71(2), 390-405. doi:10.1037/0022-
3514.71.2.390
Epstein, S., & Psychological Assessment Resources [PAR] Staff. (2008). CTI
scoring program [computer software]. Lutz, Florida: PAR.
Erdogan, B., & Bauer, T. N. (2005). Enhancing career benefits of employee
proactive personality: The role of fit with jobs and organizations. Personnel
Psychology, 58(4), 859-891. doi:10.1111/j.1744-6570.2005.00772.x
Erdogan, B., & Liden, R. C. (2002). Social exchanges in the workplace: A review of
recent developments and future research directions in leader-member
exchange theory. In L. L. Neider & C. A. Schriesheim (Eds.), Leadership
(pp. 65-114). Greenwich, CT: Information Age Publishing.
Erdogan, B., Liden, R. C., & Kraimer, M. L. (2006). Justice and leader-member
exchange: The moderating role of organizational culture. The Academy of
Management Journal, 49(2), 395-406. doi:10.5465/AMJ.2006.20786086
Estrada, C. A., Isen, A. M., & Young, M. J. (1997). Positive affect facilitates
integration of information and decreases anchoring in reasoning among
physicians. Organizational Behavior and Human Decision Processes, 72(1),
308
117-135. doi:10.1006/obhd.1997.2734
Evans, J.S.B.T. (2007). Hypothetical thinking: Dual processes in reasoning and
judgement. Hove, UK: Psychology Press
Evans, J.S.B.T. (2008). Dual-processing accounts of reasoning, judgment, and social
cognition. Annual Review of Psychology, 59(1), 255-278.
doi:10.1146/annurev.psych.59.103006.093629
Evans, J.S.B.T., & Over, D.E. (1996). Rationality and Reasoning. Hove, UK:
Psychology Press.
Eysenck, H. J., & Eysench, S. B. G. (1975). Manual of the Eysenck Personality
Questionnaire. Kent, UK: Hodder & Stoughton.
Eysenck, M. W., MacLeod, C., & Mathews, A. (1987). Cognitive functioning and
anxiety. Psychological Research, 49(2), 189-195.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999).
Evaluating the use of exploratory factor analysis in psychological research.
Psychological Methods, 4(3), 272-299. doi:10.1037/1082-989X.4.3.272
Fairhurst, G. T. (1993). Echoes of the vision when the rest of the organization talks
total quality. Management Communication Quarterly, 6(4), 331-371.
doi:10.1177/0893318993006004001
Fairhurst, G. T., & Chandler, T. A. (1989). Social structure in leader‐member
interaction. Communications Monographs, 56(3), 215-239.
doi:10.1080/03637758909390261
Fairhurst, G. T., & Uhl-Bien, M. (2012). Organizational discourse analysis (ODA):
Examining leadership as a relational process. The Leadership Quarterly,
23(6), 1043-1062. doi:10.1016/j.leaqua.2012.10.005
Fazio, R.H. (1990). Multiple processes by which attitudes guide behavior: The
309
MODE model as an integrative framework. Advances in Experimental Social
Psychology, 23(1), 75-109. doi:10.1016/S0065-2601(08)60318-4
Felfe, J. (2005). Personality and romance of leadership. In B. Schyns & J. R. Meindl
(Eds.), Implicit leadership theories: Essays and explorations (pp. 199-225).
Greenwich, CT: Information Age Publishing.
Ferguson, E., & Cox, T. (1993). Exploratory factor analysis: A users' guide.
International Journal of Selection and Assessment, 1(2), 84-94.
doi:10.1111/j.1468-2389.1993.tb00092.x
Fernandez, C. F., & Vecchio, R. P. (1997). Situational leadership theory revisited: A
test of an across-jobs perspective. The Leadership Quarterly, 8(1), 67-84.
doi:10.1016/S1048-9843(97)90031-X
Fiedler, F. E. (1964). A contingency model of leadership effectiveness. In L.
Berkowitz (Ed.), Advances in experimental social psychology (Vol. 1, pp.
150-191). New York: Academic Press.
Fiedler, F. E. (1967). A theory of leadership effectiveness. New York: McGraw-Hill.
Fiedler, F. E. (1986). The contribution of cognitive resources and leader behavior to
organizational performance. Journal of Applied Social Psychology, 16(6),
532-548. doi:10.1111/j.1559-1816.1986.tb01157.x
Fiedler, F. E., & Chemers, M. M. (1984). Improving leadership effectiveness: The
leader match concept (2nd ed.). New York: John Wiley & Sons.
Fiedler, F. E., & Garcia, J. E. (1987). New approaches to effective leadership:
Cognitive resources and organizational performance. Oxford: John Wiley &
Sons.
Fiedler, F. E., & Leister, A. F. (1977). Leader intelligence and task performance: A
test of a multiple screen model. Organizational Behavior and Human
310
Performance, 20(1), 1-14. doi:10.1016/0030-5073(77)90040-X
Fiedler, F. E., Potter, E. H., Zais, M. M., & Knowlton, W. A. (1978). Organizational
stress and the use and misuse of managerial intelligence and experience.
Journal of Applied Psychology, 64(6), 635-647. doi:10.1037/0021-
9010.64.6.635
Filley, A. C., & House, R. J. (1969). Managerial process and organizational
behavior. Glenview, Ill: Scott Foresman.
Finman, B. G. (1973). An investigation of the relationships among supervisory
attitudes, behaviors, and outputs: An examination of McGregors's Theory Y.
Personnel Psychology, 26(1), 95-105. doi:10.1111/j.1744-
6570.1973.tb01121.x
Fleishman, E. A. (1953). The description of supervisory behavior. Journal of
Applied Psychology, 37(1), 1-6. doi:10.1037/h0056314
Fleishman, E. A. (1957). A leader behavior description for industry. In R. M.
Stogdill & A. E. Coons (Eds.), Leader behavior: Its description and
measurement. Columbus, OH: Bureau of Buisness Research, Ohio State
University.
Fleishman, E. A., & Harris, E. F. (1962). Patterns of leadership behavior related to
employee grievances and turnover. Personnel Psychology, 15(1), 43-56.
doi:10.1111/j.1744-6570.1962.tb01845.x
Fleishman, E. A., Harris, E. F., & Burtt, H. E. (1955). Leadership and supervision in
industry: An evaluation of a supervisory training program. Bureau of
Educational Research Monograph, 33, 110.
Fleishman, E. A., & Peters, D. R. (1962). Interpersonal values, leadership attitudes,
and managerial “success”. Personnel Psychology, 15, 217-143.
311
doi:10.1111/j.1744-6570.1962.tb01855.x
Foa, E. B., & Foa, U. G. (1974). Societal structures of the mind. Oxford: Charles C
Thomas
Fogarty, L. A., Curbow, B. A., Wingard, J. R., McDonnell, K., & Somerfield, M. R.
(1999). Can 40 seconds of compassion reduce patient anxiety? Journal of
Clinical Oncology, 17(1), 371-379.
Foti, R.J. & Hauenstein, N. (2007). Pattern and variable approaches in leadership
emergence and effectiveness. Journal of Applied Psychology, 92(2), 347-355.
doi:10.1037/0021-9010.92.2.347
Foti, R. J., & Lord, R. G. (1987). Prototypes and scripts: The effects of alternative
methods of processing information on rating accuracy Organizational
Behavior and Human Decision Processes, 39(3), 318-340.
Frankish, K., & Evans, J.S.B.T. (2009). The duality of mind: An historical
perspective. In. J. Evans and K. Frankish (Eds.). In Two Minds: Dual
Processes and Beyond (pp.1-29). New York: Oxford University Press.
Fredrickson, B. L. (1998). What good are positive emotions? Review of General
Psychology, 2(3), 300-319. doi:10.1037/1089-2680.2.3.300
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The
broaden-and-build theory of positive emotions. American Psychologist,
56(3), 218-226. doi:10.1037/0003-066X.56.3.218
Fredrickson, B. L. (2003). Positive emotions and upward spirals in organizations. In
K. S. Cameron, J. E. Dutton, & R. E. Quinn (Eds.), Positive organizational
scholarship (pp. 163–175). San Francisco: Berrett-Kohler.
Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden the scope of
attention and thought-action repertoires. Cognition & Emotion, 19(3), 313-
312
332. doi:10.1080/02699930441000238
Frese, M., & Fay, D. (2001). Personal initiative: An active performance concept for
work in the 21st century. Research in Organizational Behavior, 23, 133-187.
doi:10.1016/S0191-3085(01)23005-6
Friedman, R. A., Tidd, S. T., Currall, S. C., & Tsai, J. C. (2000). What goes around
comes around: The impact of personal conflict style on work conflict and
stress. International Journal of Conflict Management, 11(1), 32-55.
doi:10.1108/eb022834
Fritz, C., & Sonnentag, S. (2007). Antecedents of day-level proactive behavior: A
look at job stressors and positive affect during the workday. Journal of
Management, 35(1), 94-111. doi:10.1177/0149206307308911
Fulk, J., & Wendler, E. R. (1982). Dimensionality of leader—subordinate
interactions: A path—goal investigation. Organizational Behavior and
Human Performance, 30(2), 241-264. doi:10.1016/0030-5073(82)90220-3
Fuller, B., & Marler, L. E. (2009). Change driven by nature: A meta-analytic review
of the proactive personality literature. Journal of Vocational Behavior, 75(3),
329-345. doi:10.1016/j.jvb.2009.05.008
Furnham, A., Forde, L., & Cotter, T. (1998). Personality and intelligence.
Personality and Individual Differences, 24(2), 187-192. doi:10.1016/S0191-
8869(97)00169-4
Gaesser, B. (2013). Constructing memory, imagination, and empathy: A cognitive
neuroscience perspective. Frontiers in Psychology, 3, 1-6.
doi:10.3389/fpsyg.2012.00576
Gajendran, R. S., & Joshi, A. (2012). Innovation in globally distributed teams: The
role of LMX, communication frequency, and member influence on team
313
decisions. Journal of Applied Psychology, 97(6), 1252-1261.
doi:10.1037/a0028958
Gao, L., Janssen, O., & Shi, K. (2011). Leader trust and employee voice: The
moderating role of empowering leader behaviors. The Leadership Quarterly,
22(4), 787-798. doi:10.1016/j.leaqua.2011.05.015
Gardner, W. L., & Avolio, B. J. (1998). The charismatic relationship: A
dramaturgical perspective. The Academy of Management Review, 23(1), 32-
58. doi:10.5465/AMR.1998.192958
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 Psychology Review, 30(7), 849-864.
doi:10.1016/j.cpr.2010.03.002
Garon, M. (2012). Speaking up, being heard: registered nurses' perceptions of
workplace communication. Journal of Nursing Management, 20(3), 361-371.
doi:10.1111/j.1365-2834.2011.01296.x
Garon, M., & Ringl, K. (2004). Job satisfaction of hospital-based registered nurses.
Online Journal of Clinical Innovations, 7, 1-45.
Gawronski, B., & Creighton, L.A. (2013). Dual Process Theories. In D.E. Carlston
(ed.). The Oxford Handbook of Social Cognition, (pp.282-312). New York,
NY: Oxford University Press.
Geiser, C., Bishop, J., Lockhart, G., Shiffman, S., & Grenard, J. L. (2013).
Analyzing latent state-trait and multiple-indicator latent growth curve models
as multilevel structural equation models. Frontiers in Psychology, 4, 1-23.
314
doi:10.3389/fpsyg.2013.00975
George, D., & Mallery, M. (2003). Using SPSS for Windows step by step: a simple
guide and reference. Boston, MA: Allyn and Bacon.
Gerstner, C. R., & Day, D. V. (1997). Meta-analytic review of leader–member
exchange theory: Correlates and construct issues. Journal of Applied
Psychology, 82(6), 827-844. doi:10.1037/0021-9010.82.6.827
Gibb, C. A. (1947). The principles and traits of leadership. The Journal of Abnormal
and Social Psychology, 42(3), 267-284. doi:10.1037/h0056420
Goffee, R., & Jones, G. (2001). Followership: It's personal too. Harvard Business
Review, 79(11), 148.
Goldberg, L. R. (1990). An alternative" description of personality": The big-five
factor structure. Journal of Personality & Social Psychology, 59(6), 1216-
1229. doi:10.1037/0022-3514.59.6.1216
Goodwin, V. L., Bowler, W. M., & Whittington, J. L. (2008). A social network
perspective on LMX relationships: Accounting for the instrumental value of
leader and follower networks. Journal of Management, 35(4), 954-980.
doi:10.1177/0149206308321555
Goodwin, V. L., Wofford, J. C., & Boyd, N. G. (2000). A laboratory experiment
testing the antecedents of leader cognitions. Journal of Organizational
Behavior, 21(7), 769-788. doi:10.1002/1099-1379(200011)21:7<769::AID-
JOB53>3.0.CO;2-J
Graeff, C. L. (1983). The situational leadership theory: A critical view. The Academy
of Management Review, 8(2), 285-291. doi:10.5465/AMR.1983.4284738
Graen, G., Dansereau, F., Minami, T., & Cashman, J. (1973). Leadership behaviors
as cues to performance evaluation. The Academy of Management Journal,
315
16(4), 611-623. doi:10.2307/254694
Graen, G. B., & Scandura, T. A. (1987). Toward a psychology of dyadic organizing.
In L. L. Cummings & B. M. Staw (Eds.), Research in organizational
behavior (Vol. 9, pp. 175-208). Greenwich, CT: JAI.
Graen, G. B., & Uhl-Bien, M. (1991). The transformation of professionals into self-
managing and partially self-designing contributors: Toward a theory of
leadership-making. Journal of Management Systems, 3(3), 25-39. Retrieved
from
http://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1015&context=m
anagementfacpub&sei-
redir=1&referer=http%3A%2F%2Fscholar.google.com.au%2Fscholar%3Fhl
%3Den%26as_SDt%3D0%2C5%26q%3Dgraen%2Buhl-
bien%2B1991#search=%22graen%20uhl-bien%201991%22
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. doi:10.1016/1048-9843(95)90036-5
Grant, A. M., & Ashford, S. J. (2008). The dynamics of proactivity at work.
Research in Organizational Behavior, 28, 3-34.
doi:10.1016/j.riob.2008.04.002
Grant, A. M., Parker, S., & Collins, C. (2009). Getting credit for proactive behavior:
Supervisor reactions depend on what you value and how you feel. Personnel
Psychology, 62(1), 31-55. doi:10.1111/j.1744-6570.2008.01128.x
Graziano, W. G., Habashi, M. M., Sheese, B. E., & Tobin, R. M. (2007).
Agreeableness, empathy, and helping: A person x situation perspective.
316
Journal of Personality and Social Psychology, 93(4), 583-599.
doi:10.1037/0022-3514.93.4.583
Green, C. (2008). Leader member exchange and the use of moderating conflict
management styles: Impact on relationship quality. International Journal of
Conflict Management, 19(2), 92-111. doi:10.1108/10444060810856058
Griffin, R. W. (1980). Relationships among individual, task design, and leader
behavior variables. The Academy of Management Journal, 23(4), 665-683.
doi:10.2307/255555
Griffin, R. W. (1981). A longitudinal investigation of task characteristics
relationships. The Academy of Management Journal, 24(1), 99-113.
doi:10.2307/255826
Grojean, M. W., Resick, C. J., Dickson, M. W., & Smith, D. B. (2004). Leaders,
values, and organizational climate: Examining leadership strategies for
establishing an organizational climate regarding ethics. Journal of Business
Ethics, 55(3), 223-241. doi:10.1007/s10551-004-1275-5
Grossberg. (2013). Adaptive resonance theory: How a brain learns to consciously
attend, learn, and recognize a changing world. Neural Networks, 37, 1-47.
doi:10.1016/j.neunet.2012.09.017
Gwartney-Gibbs, P. A., & Lach, D. H. (1994). Gender differences in clerical
workers' disputes over tasks, interpersonal treatment, and emotion. Human
Relations, 47(6), 611-639. doi:10.1177/001872679404700603
Haddock, G., Maio, G. R., Arnold, K., & Huskinson, T. (2008). Should persuasion
be affective or cognitive? The moderating effects of need for affect and need
for cognition. Personality and Social Psychology Bulletin, 34(6), 769-778.
doi:10.1177/0146167208314871
317
Halpin, A. W., & Winer, B. J. (1957). A factorial study of the leader behavior
descriptions. In R. M. Stogdill & A. E. Coons (Eds.), Leader behavior: Its
description and measurement (pp. 39-51). Columbus, OH: Bureau of
Buisness Research, Ohio State University.
Hammond, K. R. 1996. Human judgment and social policy. Oxford: Oxford
University Press.
Hanges, P. J., Lord, R. G., & Dickson, M. (2000). An information‐processing
perspective on leadership and culture: A case for connectionist architecture.
Applied Psychology, 49(1), 133-161. doi:10.1111/1464-0597.00008
Hansbrough, T. (2005). Cognition matters: Leader images and their implications for
organizational life. In B. Schyns & J. R. Meindl (Eds.), Implicit leadership
theories: Essays and explorations (pp. 63-80). Greenwich, CT: Information
Age Publishing.
Harris, K. J., Harvey, P., & Kacmar, K. M. (2011). Abusive supervisory reactions to
coworker relationship conflict. The Leadership Quarterly, 22(5), 1010-1023.
doi:10.1016/j.leaqua.2011.07.020
Hartog, D. N., & Belschak, F. D. (2007). Personal initiative, commitment and affect
at work. Journal of Occupational and Organizational Psychology, 80(4),
601-622. doi:10.1348/096317906X171442
Hatcher, L. (1994). A step-by-step approach to using SAS for factor analysis and
structural equation modeling. Cary, NC: SAS Publishing.
Hautala, T. M. (2005). The effects of subordinates' personality on appraisals of
transformational leadership. Journal of Leadership & Organizational Studies,
11(4), 84-92. doi:10.1177/107179190501100407
Hautala, T. M. (2006). The relationship between personality and transformational
318
leadership. Journal of Management Development, 25(8), 777-794.
doi:10.1108/02621710610684259
Hemphill, J. K., & Coons, A. E. (1957). Development of the leader behavior
description questionnaire. In R. M. Stodgill & A. E. Coons (Eds.), Leader
behavior: Its description and measurement. (pp. 6-38). Columbus, Ohio:
Bureau of Business Research, Ohio State University.
Henry, P. K. (1997). Overcoming resistance to organizational change. American
Dietetic Association. Journal of the American Dietetic Association, suppl.
Best of the DPG Newsletters, 97(10), S145-S147. doi:10.1016/S0002-
8223(97)00751-7
Hersey, P., & Blanchard, K. H. (1969). Management of organizational behavior.
Englewood Cliffs, N.J: Prentice-Hall.
Hersey, P., & Blanchard, K. H. (1977). Management of organizational behavior:
Utilizing human resources (3rd ed.). Englewood Cliffs, N.J Prentice-Hall
Higgins, C. A., Judge, T. A., & Ferris, G. R. (2003). Influence tactics and work
outcomes: A meta-analysis. Journal of Organizational Behavior, 24(1), 89-
106. doi:10.1002/job.181
Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and
salience. In E. Higgins & A. Kruglanski (Eds.), Social psychology: Handbook
of basic principles (pp.133–168). New York: Guilford Press.
Higgins, E. T., King, G. A., & Mavin, G. H. (1982). Individual construct
accessibility and subjective impressions and recall. Journal of Personality
and Social Psychology, 43(1), 35-47. doi:10.1037/0022-3514.43.1.35
Hitt, M. A., Beamish, P. W., Jackson, S. E., & Mathieu, J. E. (2007). Building
theoretical and empirical bridges across levels: Multilevel research in
319
management. The Academy of Management Journal, 50(6), 1385-1399.
doi:10.2307/20159480
Hodges, W. F. (1968). Effects of ego threat and threat of pain on state anxiety.
Journal of Personality and Social Psychology, 8(4), 364-372.
doi:10.1037/h0025491
Hoegl, M., Parboteeah, P.K., & Gemuenden, H. G. (2003). When teamwork really
matters: Task innovativeness as a moderator of the teamwork–performance
relationship in software development projects. Journal of Engineering and
Technology Management, 20(4), 281-302.
doi:10.1016/j.jengtecman.2003.08.001
Hofmann, D. A. (1997). An overview of the logic and rationale of hierarchical linear
models. Journal of Management, 23(6), 723-744.
doi:10.1177/014920639702300602
Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in hierarchical linear
models: Implications for research in organizations. Journal of Management,
24(5), 623-641. doi:10.1177/014920639802400504
Hollander, E. P. (1971). Style, structure, and setting in organizational leadership.
Organizational Leadership, 16(1), 1-9. doi:10.2307/2391280
Hollander, E. P. (1992). The essential interdependence of leadership and
followership. Current Directions in Psychological Science, 1(2), 71-75.
Hollander, E. P. (2012). Inclusive leadership: The essential leader-follower
relationship. New York: Routledge.
Hollander, E. P., & Julian, J. W. (1969). Contemporary trends in the analysis of
leadership processes. Psychological Bulletin, 71(5), 387-397.
doi:10.1037/h0027347
320
Hollander, E. P., & Offermann, L. R. (1990). Power and leadership in organizations:
Relationships in transition. American Psychologist, 45(2), 179-189.
doi:10.1037/0003-066X.45.2.179
Homans, G. C. (1950). The Human Group. New York: Harcourt Brace.
Hoogh, D., Den Hartog, D. N., & Koopman, J. (2005). Linking the big five‐factors
of personality to charismatic and transactional leadership; perceived dynamic
work environment as a moderator. Journal of Organizational Behavior,
26(7), 839-865. doi:10.1002/job.344
Hoption, C., Christie, A., & Barling, J. (2012). Submitting to the Follower Label.
Zeitschrift für Psychologie, 220(4), 221-230. doi:10.1027/2151-
2604/a000116
Horowitz, M. J. (1988). Introduction to psychodynamics: A new synthesis. New
York: Basic Books.
Hosking, D. M. (2002). Leadership processes and leadership development:
Reflections from a social constructionist paradigm. 19. Retrieved from
http://www.researchgate.net/profile/Dm_Hosking/publication/27702098_Lea
dership_processes_and_leadership_development_Reflections_from_a_social
_constructionist_paradigm/links/004635305f8c8ceeb4000000.pdf
House, R. J. (1971). A path goal theory of leader effectiveness. Administrative
Science Quarterly, 16(3), 321-339. doi:10.2307/2391905
House, R. J. (1977). A 1976 theory of charismatic leadership. In J. G. Hunt & L. L.
Larson (Eds.), Leadership: The cutting edge (pp. 189-207). Carbondale:
Southern Illinois University Press.
House, R. J. (1996). Path-goal theory of leadership: Lessons, legacy, and a
reformulated theory. The Leadership Quarterly, 7(3), 323-352.
321
doi:10.1016/S1048-9843(96)90024-7
House, R. J., & Dessler, G. (1974). The path-goal theory of leadership: Some post
hoc and a priori tests. In J. Hunt & L. Larson (Eds.), Contingency approaches
to leadership (pp. 29-55). Carbondale, IL: Southern Illinois University Press.
House, R. J., Filley, A. C., & Kerr, S. (1971). Relation of leader consideration and
initiating structure to R and D subordinates' satisfaction. Administrative
Science Quarterly, 16(1), 19-30. doi:10.2307/2391283
House, R. J., & Mitchell, T. R. (1975). Path-goal theory of leadership (TR-75-67).
Washington University, Department of Psychology, Seattle: Retrieved from:
www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA009513
Howell, J. P., & Méndez, M. J. (2008). Three perspectives on followership. In E. R.
Ronald, I. Chaleff, & J. Lipman-Blumen (Eds.), The art of followership: How
great followers create great leaders and organizations (pp. 25-40). San
Francisco, CA: Jossey-Bass.
Hunt, J. G., Boal, K. B., & Sorenson, R. L. (1990). Top management leadership:
Inside the black box. The Leadership Quarterly, 1(1), 41-65.
doi:10.1016/1048-9843(90)90014-9
Ikenberry, J. (1996). The future of international leadership. Political Science
Quarterly, 111(3), 385-402. doi:10.2307/2151968
Ilgen, D. R., Fisher, C. D., & Taylor, S. M. (1979). Consequences of individual
feedback on behavior in organizations. Journal of Applied Psychology, 64(4),
349-371. doi:10.1037/0021-9010.64.4.349
Ilies, R., de Pater, I. E., & Judge, T. (2007). Differential affective reactions to
negative and positive feedback, and the role of self‐esteem. Journal of
Managerial Psychology, 22(6), 590-609. doi:10.1108/02683940710778459
322
Ilies, R., Nahrgang, J. D., & Morgeson, F. P. (2007). Leader-member exchange and
citizenship behaviors: a meta-analysis. Journal of Applied Psychology, 92(1),
269-277. doi:10.1037/0021-9010.92.1.269
Isen, A. M., & Daubman, K. A. (1984). The influence of affect on categorization.
Journal of Personality and Social Psychology, 47(6), 1206-1217.
doi:10.1037/0022-3514.47.6.1206
Isen, A. M., & Means, B. (1983). The influence of positive affect on decision-
making strategy. Social Cognition, 2(1), 18-31. doi:10.1521/soco.1983.2.1.18
Jackson, E.M., & Johnson, R.E. (2012). When opposites do (and do not) attract:
Interplay of leader and follower self-identities and its consequences for
leader–member exchange. Leadership Quarterly, 23(3), 488-501.
doi:10.1016/j.leaqua.2011.12.003
Jackson, B., & Parry, K. W. (2011). Follower-centric perspectives of leadership. In
B. Jackson & K. Parry (Eds.), A very short, fairly interesting and reasonably
cheap book about studying leadership (2nd ed., pp. 45-67). London: SAGE.
Jenkins, W. O. (1947). A review of leadership studies with particular reference to
military problems. Psychological Bulletin, 44(1), 54-79.
doi:10.1037/h0062329
Johnson, K. K. P., Schofield, N. A., & Yurchisin, J. (2002). Appearance and dress as
a source of information: A qualitative approach to data collection. Clothing
and Textiles Research Journal, 20 (3), 125-137.
doi:10.1177/0887302X0202000301
Johnson, R. E., Silverman, S. B., Shyamsunder, A., Swee, H., Rodopman, O. B.,
Cho, E., & Bauer, J. (2010). Acting Superior But Actually Inferior?:
Correlates and Consequences of Workplace Arrogance. Human Performance,
323
23(5), 403-427. doi:10.1080/08959285.2010.515279
Johnson, S. K., Murphy, S. E., Zewdie, S., & Reichard, R. J. (2008). The strong,
sensitive type: Effects of gender stereotypes and leadership prototypes on the
evaluation of male and female leaders. Organizational Behavior and Human
Decision Processes, 106(1), 39-60. doi:10.1016/j.obhdp.2007.12.002
Judge, T. A., & Bono, J. E. (2001). Relationship of core self-evaluations traits--self-
esteem, generalized self-efficacy, locus of control, and emotional stability--
with job satisfaction and job performance: A meta-analysis. Journal of
Applied Psychology, 86(1), 80-92. doi:10.1037/0021-9010.86.1.80
Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and
leadership: A qualitative and quantitative review. Journal of Applied
Psychology, 87(4), 765-780. doi:10.1037/0021-9010.87.4.765
Judge, T. A., Colbert, A. E., & Ilies, R. (2004). Intelligence and leadership: A
quantitative review and test of theoretical propositions. Journal of Applied
Psychology, 89(3), 542-552. doi: 10.1037/0021-9010.89.3.542
Judge, T. A., Erez, A., & Bono, J. E. (1998). The power of being positive: The
relation between positive self-concept and job performance. Human
Performance, 11(2-3), 167-187. doi:10.1080/08959285.1998.9668030
Judge, T. A., & Ferris, G. R. (1992). The elusive criterion of fit in human resources
staffing decisions. Human Resource Planning, 15(4), 47-67.
Judge, T. A., & Ferris, G. R. (1993). Social context of performance evaluation
decisions. The Academy of Management Journal, 36(1), 80-105.
doi:10.2307/256513
Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership:
A meta-analytic test of their relative validity. Journal of Applied Psychology,
324
89(5), 755-768. doi:10.1037/0021-9010.89.5.755
Judge, T. A., Piccolo, R. F., & Ilies, R. (2004). The forgotten ones? The validity of
consideration and initiating structure in leadership research. Journal of
Applied Psychology, 89(1), 36-51. doi:10.1037/0021-9010.89.1.36
Jung, C. G. (1923). Psychological types: Or the psychology of individuation. (H. G.
Baynes, Trans.). Oxford: Harcourt.
Jung, C. G. (1991). Psyche and Symbol (R. F. C. Hull, Trans. V. S. d. Laszlo Ed.).
Princeton, NJ: Princeton University Press.
Jussim, L., Nelson, T. E., Manis, M., & Soffin, S. (1995). Prejudice, stereotypes, and
labeling effects: Sources of bias in person perception. Journal of Personality
and Social Psychology, 68(2), 228-246. doi:10.1037/0022-3514.68.2.228
Kahn, B. E., & Isen, A. M. (1993). The influence of positive affect on variety
seeking among safe, enjoyable products. Journal of Consumer Research,
20(2), 257-270.
Kahn, R. L. (1956). The prediction of productivity. Journal of Social Issues, 12(2),
41-49. doi:10.1111/j.1540-4560.1956.tb00367.x
Kahneman D. (2002). Nobel prize lecture: Maps of Bounded Rationality: a
perspective on intuitive judgment and choice. In T Frangsmyr (Ed.). Nobel
Prizes 2002: Nobel Prizes, Presentations, Biographies, & Lectures, (pp.
416–99). Stockholm: Almqvist & Wiksell Int. Retrieved from
http://down.cenet.org.cn/upfile/58/2005419105516174.pdf
Kahneman, D. (2003). Maps of bounded rationality: Psychology for behavioral
economics. American Economic Review, 93(5), 1449-1475.
doi:10.1257/000282803322655392
Kahneman, D. (2011). Thinking, fast and slow. New York: Farrar, Straus and Giroux
325
Kahneman, D., & Frederick, S. (2002). Representativeness revisited: Attribute
substitution in intuitive judgment. In. T. Gilovich, D. Griffin and D.
Kahneman (Eds.). Heuristics of Intuitive Judgment: Extensions and
Application (pp. 49-81). Cambridge, UK: Cambridge University Press
Kassing, J. W. (2000). Investigating the relationship between superior‐subordinate
relationship quality and employee dissent. Communication Research Reports,
17(1), 58-69. doi:10.1080/08824090009388751
Katz, D., & Kahn, R. L. (1978). The social psychology of organizations (2nd ed.).
New York: John Wiley & Sons.
Katz, R. L. (1963). Empathy: Its nature and uses. London: Free Press.
Keashly, L., Trott, V., & MacLean, L. M. (1994). Abusive behavior in the
workplace: A preliminary investigation. Violence and Victims, 9(4), 341-357.
Keller, R. T. (1989). A test of the path-goal theory of leadership with need for clarity
as a moderator in research and development organizations. Journal of
Applied Psychology, 74(2), 208-212. doi:10.1037/0021-9010.74.2.208
Keller, T. (1999). Images of the familiar: Individual differences and implicit
leadership theories. The Leadership Quarterly, 10(4), 589-607.
doi:10.1016/S1048-9843(99)00033-8
Keller, T. (2003). Parental images as a guide to leadership sensemaking: An
attachment perspective on implicit leadership theories. The Leadership
Quarterly, 14(2), 141-160. doi:10.1016/S1048-9843(03)00007-9
Kellett, J. B., Humphrey, R. H., & Sleeth, R. G. (2002). Empathy and complex task
performance: Two routes to leadership. The Leadership Quarterly, 13(5),
523-544. doi:10.1016/s1048-9843(02)00142-x
Kelley, R. E. (1988). In praise of followers. Harvard Business Review, 66(6), 142-
326
148.
Kelley, R. E. (1992). The power of followership: How to create leaders people want
to follow, and followers who lead themselves. New York: Doubleday
Currency.
Kelley, R. E. (2008). Rethinking Followership. In R. E. Riggio, I. Chaleff, & J.
Lipman-Blumen (Eds.), The art of followership: How great followers create
great leaders and organizations (pp. 5-16). San Francisco, CA: Jossey-Bass.
Kenney, R. A., Schwartz-Kenney, B. M., & Blascovich, J. (1996). Implicit
leadership theories: Defining leaders described as worthy of influence.
Personality and Social Psychology Bulletin, 22(11), 1128-1143.
doi:10.1177/01461672962211004
Kenny, D. A., & Zaccaro, S. J. (1983). An estimate of variance due to traits in
leadership. Journal of Applied Psychology, 68(4), 678-685.
doi:10.1037/0021-9010.68.4.678
Kets de Vries, M. (2004). Organizations on the couch: A clinical perspective on
organizational dynamics. European Management Journal, 22(2), 183-200.
doi:10.1016/j.emj.2004.01.008
King, G. (2012). A re-examination of the relationship between information
processing and transformational leadership (Unpublished Master Thesis).
Murdoch University, Perth, Australia.
Kirkman, B. L., & Rosen, B. (1999). Beyond self-management: Antecedents and
consequences of team empowerment. The Academy of Management Journal,
42(1), 58-74. doi:10.2307/256874
Kirkpatrick, L. A., & Epstein, S. (1992). Cognitive-experiential self-theory and
subjective probability: Further evidence for two conceptual systems. Journal
327
of Personality and Social Psychology, 63(4), 534-544. doi:10.1037/0022-
3514.63.4.534
Kline, P. (2000). A psychometrics primer. London: Free Association Books.
Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on
performance: A historical review, a meta-analysis, and a preliminary
feedback intervention theory. Psychological Bulletin, 119(2), 254-284.
doi:10.1037/0033-2909.119.2.254
Knee, C. R., Nanayakkara, A., Vietor, N. A., Neighbors, C., & Patrick, H. (2001).
Implicit theories of relationships: Who cares if romantic partners are less than
ideal? Personality and Social Psychology Bulletin, 27(7), 808-819.
doi:10.1177/0146167201277004
Koommoo-Welch, P. (2008). Implicit leadership theories: Perceptions of charisma,
people, and performance. (Unpublished Ph.D Thesis). North Carolina State
University, Raleigh. Retrieved from
http://repository.lib.ncsu.edu/ir/bitstream/1840.16/4009/1/etd.pdf
Korman, A. K. (1966). “Consideration,”“initiating structure,” and organizational
criteria: A review. Personnel Psychology, 19(4), 349-361.
doi:10.1111/j.1744-6570.1966.tb00310.x
Kotter, J. P. (1990). What leaders really do. Harvard Business Review, 68(3), 103-
111.
Koys, D. J. (2001). The effects of employee satisfaction, organizational citizenship
behavior, and turnover on organizational effectiveness: A unit‐level,
longitudinal study. Personnel Psychology, 54(1), 101-114.
doi:10.1111/j.1744-6570.2001.tb00087.x
Kramer, R. M., & Tyler, T. R. (1996). Trust in organizations: Frontiers of theory
328
and research. Thousand Oaks, CA: SAGE.
Krishnan, R., Martin, X., & Noorderhaven, N. G. (2006). When does trust matter to
alliance performance? The Academy of Management Journal, 49(5), 894-917.
doi:10.5465/AMJ.2006.22798171
Kristof, A. L. (1996). Person‐organization fit: An integrative review of its
conceptualizations, measurement, and implications. Personnel Psychology,
49(1), 1-49. doi:10.1111/j.1744-6570.1996.tb01790.x
Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C. (2005). Consequences of
individuals' fit at work: A meta-analysis of person-job, person-organization,
person-group, and person-supervisor fit. Personnel Psychology, 58(2), 281-
342. doi:10.1111/j.1744-6570.2005.00672.x
Kroeger, O., Thuesen, J. M., & Rutledge, H. (2002). Type talk at work (revised):
How the 16 personality types determine your success on the job. n.p: Delta.
Kruse, E. T., & Sy, T. (2011). Manipulating implicit theories by inducing affect.
Academy of Management Proceedings, 1(1), 1-6.
doi:10.5465/AMBPP.2011.65870487
Larson, J. R. (1982). Cognitive mechanisms mediating the impact of implicit
theories of leader behavior on leader behavior ratings. Organizational
Behavior and Human Performance, 29(1), 129-140. doi:10.1016/0030-
5073(82)90245-8
Larson, L. L., Hunt, J. G., & Osborn, R. N. (1974). Correlates of leadership and
demographic variables in three organizational settings. Journal of Business
Research, 2(3), 335-348. doi:10.1016/0148-2963(74)90009-5
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature
review and two-component model. Psychological Bulletin, 107(1), 34-47.
329
doi:10.1037/0033-2909.107.1.34
Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthetic
appreciation and aesthetic judgments. British Journal of Psychology, 95(4),
489-508. doi:10.1348/0007126042369811
LePine, J. A., & van Dyne, L. (2001). Voice and cooperative behavior as contrasting
forms of contextual performance: Evidence of differential relationships with
big five personality characteristics and cognitive ability. Journal of Applied
Psychology, 86(2), 326-336. doi:10.1037/0021-9010.86.2.326
Levin, I. M., & Stokes, J. P. (1989). Dispositional approach to job satisfaction: Role
of negative affectivity. Journal of Applied Psychology, 74(5), 752-758.
doi:10.1037/0021-9010.74.5.752
Lewicki, R. J., Tomlinson, E. C., & Gillespie, N. (2006). Models of interpersonal
trust development: Theoretical approaches, empirical evidence, and future
directions. Journal of Management, 32(6), 991-1022.
doi:10.1177/0149206306294405
Liden, R. C., & Maslyn, J. M. (1998). Multidimensionality of Leader-Member
Exchange: An empirical assessment through scale development. Journal of
Management, 24(1), 43-72. doi:10.1177/014920639802400105
Liden, R. C., Sparrowe, R. T., & Wayne, S. J. (1997). Leader-member exchange
theory: The past and potential for the future. Research in Personnel and
Human Resources Management, 15, 47-119.
Liden, R. C., Wayne, S. J., & Stilwell, D. (1993). A longitudinal study on the early
development of leader-member exchanges. Journal of Applied Psychology,
78(4), 662-674. doi:10.1037/0021-9010.78.4.662
Lopes, P. N., Salovey, P., & Straus, R. (2003). Emotional intelligence, personality,
330
and the perceived quality of social relationships. Personality and Individual
Differences, 35(3), 641-658. doi:10.1016/S0191-8869(02)00242-8
Lord, R. G. (2008). Beyond transactional and transformational leadership: Can
leaders till lead when they don't know what to do? In M. Uhl-Bien & R.
Marion (Eds.), Complexity leadership (pp. 155-184). Charlotte, NC:
Information Age Publishing.
Lord, R. G., & Alliger, G. M. (1985). A comparison of four information processing
models of leadership and social perceptions. Human Relations, 38(1), 47-65.
doi:10.1177/001872678503800103
Lord, R. G., & Brown, D. J. (2001). Leadership, values, and subordinate self-
concepts. The Leadership Quarterly, 12(2), 133-152. doi:10.1016/S1048-
9843(01)00072-8
Lord, R. G., Brown, D. J., Harvey, J. L., & Hall, R. J. (2001). Contextual constraints
on prototype generation and their multilevel consequences for leadership
perceptions. The Leadership Quarterly, 12(3), 311-338. doi:10.1016/S1048-
9843(01)00081-9
Lord, R. G., de Vader, C. L., & Alliger, G. M. (1986). A meta-analysis of the
relation between personality traits and leadership perceptions: An application
of validity generalization procedures. Journal of Applied Psychology, 71(3),
402-410. doi:10.1037/0021-9010.71.3.402
Lord, R. G., & Emrich, C. G. (2001). Thinking outside the box by looking inside the
box: Extending the cognitive revolution in leadership research. The
Leadership Quarterly, 11(4), 551-579. doi:10.1016/S1048-9843(00)00060-6
Lord, R. G., Foti, R. J., & de Vader, C. L. (1984). A test of leadership categorization
theory: Internal structure, information processing, and leadership perceptions.
331
Organizational Behavior and Human Performance, 34(3), 343-378.
doi:10.1016/0030-5073(84)90043-6
Lord, R. G., Foti, R. J., & Phillips, J. S. (1982). A theory of leadership
categorization. In G. H. James, U. Sekaran, & C. A. Schriesheim (Eds.),
Leadership: Beyond establishment views (pp. 104-121). Carbondale, IL:
Southern Illinois University Press.
Lord, R. G., & Maher, K. J. (1990). Alternative information-processing models and
their implications for theory, research, and practice. The Academy of
Management Review, 15(1), 9-28. doi:10.5465/AMR.1990.4308219
Lord, R. G., & Maher, K. J. (1993). Leadership and information processing: Linking
perceptions and performance. New York: Routledge.
Lowe, K. B., Kroeck, K. G., & Sivasubramaniam, N. (1996). Effectiveness correlates
of transformational and transactional leadership: A meta-analytic review of
the mlq literature. The Leadership Quarterly, 7(3), 385-425.
doi:10.1016/s1048-9843(96)90027-2
Loyal. (n.d.). Roget's 21st Century Thesaurus, Third Edition. Retrieved May 21,
2015, from Thesaurus.com website: http://www.thesaurus.com/browse/loyal
Luke, D. A. (2004). Multilevel modeling. Thousand Oaks, CA: SAGE.
Lundin, S. C., & Lancaster, L. C. (1990). Beyond leadership... the importance of
followership. The Futurist, 24(3), 18-22.
Mahsud, R., Yukl, G., & Prussia, G. (2010). Leader empathy, ethical leadership, and
relations‐oriented behaviors as antecedents of leader‐member exchange
quality. Journal of Managerial Psychology, 25(6), 561-577.
doi:10.1108/02683941011056932
Maier, N. R. F., & Verser, G. C. (1982). Psychology in Industrial Organizations (5th
332
ed.). Boston, MA: Houghton-Mifflin.
Mann, R. D. (1959). A review of the relationships between personality and
performance in small groups. Psychological Bulletin, 56(4), 241-270.
doi:10.1037/h0044587
Marks, M. A., Mathieu, J. E., & Zaccaro, S. J. (2001). A temporally based
framework and taxonomy of team processes. The Academy of Management
Review, 26(3), 356-376. doi:10.5465/AMR.2001.4845785
Martin, R. (2009). Pathgoal theory of leadership. In J. M. Levine & M. A. Hogg
(Eds.), Encyclopedia of group processes & intergroup relations (pp. 636-
637). Thousand Oaks, CA: SAGE.
May, D. R., Gilson, R. L., & Harter, L. M. (2004). The psychological conditions of
meaningfulness, safety and availability and the engagement of the human
spirit at work. Journal of Occupational and Organizational Psychology,
77(1), 11-37. doi:10.1348/096317904322915892
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of
organizational trust. Academy of Management Review, 20(3), 709-734.
doi:10.5465/AMR.1995.9508080335
McAllister, D. J. (1995). Affect-and cognition-based trust as foundations for
interpersonal cooperation in organizations. The Academy of Management
Journal, 38(1), 24-59. doi:10.2307/256727
McCann, C. D., & Higgins, E. T. (1988). Motivation and affect in interpersonal
relations: The role of personal orientations and discrepancies. In L.
Donohew, H. E. Sypher, & E. T. Higgins (Eds.), Communication, social
cognition, and affect (pp. 53-79). HillSDale, NJ: Lawrence Erlbaum.
McClelland,J.L., McNaughton, B.L., & O’Reilly, R.C. (1995). Why there are
333
complementary learning systems in the hippocampus and neocortex: Insights
from the successes and failures of connectionist models of learning and
memory. Psychological Review, 102(3), 419-457. doi:10.1037/0033-
295X.102.3.419
McColl-Kennedy, J. R., & Anderson, R. D. (2002). Impact of leadership style and
emotions on subordinate performance. The Leadership Quarterly, 13(5), 545-
559. doi:10.1016/S1048-9843(02)00143-1
McGregor, D. M. (1960). The human side of enterprise. New York McGrawHill.
Meindl, J. R. (1995). The romance of leadership as a follower-centric theory: A
social constructionist approach. The Leadership Quarterly, 6(3), 329-341.
doi:10.1016/1048-9843(95)90012-8
Meindl, J. R. (1998). The romance of leadership as a follower-centric theory: A
social constructionist approach. In F. Dansereau & F. Yammarino (Eds.),
Leadership: The multiple-level approaches (pp. 285-298). Stanford, CT: JAI.
Meindl, J. R., Ehrlich, S. B., & Dukerich, J. M. (1985). The romance of leadership.
Administrative Science Quarterly, 30(1), 78-102. doi:10.2307/2392813
Melchior, L. A., & Cheek, J. M. (1990). Shyness and anxious self-preoccupation
during a social interaction. Journal of Social Behavior and Personality, 5(2),
117-130.
Menon, S. T. (2001). Employee empowerment: An integrative psychological
approach. Applied Psychology, 50(1), 153-180. doi:10.1111/1464-
0597.00052
Meyer, W. U., & Starke, E. (1982). Own ability in relation to self-concept of ability:
A field study of information-seeking. Personality and Social Psychology
Bulletin, 8(3), 501-507. doi:10.1177/0146167282083017
334
Milgram, S. (1965). Some conditions of obedience and disobedience to authority.
Human Relations, 18(1), 57-76. doi:10.1177/001872676501800105
Mitchell, T. R. (1982). People in organizations: An introduction to organizational
behavior. New York: McGraw-Hill.
Morgeson, F. P., Reider, M. H., & Campion, M. A. (2005). Selecting individuals in
team settings: The importance of social skills, personality characteristics, and
teamwork knowledge. Personnel Psychology, 58(3), 583-611.
doi:10.1111/j.1744-6570.2005.655.x
Morrison, E. W., & Milliken, F. J. (2000). Organizational silence: A barrier to
change and development in a pluralistic world. The Academy of Management
Review, 25(4), 706-725. doi:10.5465/AMR.2000.3707697
Moss, S. A., & Ngu, S. (2006). The relationship between personality and leadership
preferences. Current Research in Social Psychology, 11(6), 71-91.
Moskowitz, G.B., Skurnik, I., & Galinsky, A.D. (1999). The history of Dual-Process
Notions, and the Future of Preconscious Control. In S. Chaiken & Y. Trope
(Eds.). Dual-Dual Process Theories in Social Psychology (pp. 12-39).
Guilford Press: New York.
Mumford, M. D., Zaccaro, S. J., Harding, F. D., Jacobs, T. O., & Fleishman, E. A.
(2000). Leadership skills for a changing world: Solving complex social
problems. The Leadership Quarterly, 11(1), 11-35. doi:10.1016/S1048-
9843(99)00041-7
Mumford, M. D., Zaccaro, S. J., Johnson, J. F., Diana, M., Gilbert, J. A., & Threlfall,
K. V. (2000). Patterns of leader characteristics: Implications for performance
and development. Leadership Quarterly, 11(1), 115–133.
doi:10.1016/S1048-9843(99)00045-4
335
Murphy, S. E., & Ensher, E. A. (1999). The effects of leader and subordinate
characteristics in the development of leader–member exchange quality.
Journal of Applied Social Psychology, 29(7), 1371-1394. doi:10.1111/j.1559-
1816.1999.tb00144.x
Muthen, B. O. (2015, August 21). Re: Level 1 moderator approach [Online forum
comment]. Retrieved from
http://www.statmodel.com/discussion/messages/12/21764.html?1440189893
Myers, I., & Myers, P. (2010). Gifts differing: Understanding personality type.
Moutain View, CA: Davies-Black Publishing.
Myers, I. B., & McCaulley, M. H. (1985). Manual, a guide to the development and
use of the Myers-Briggs type indicator. Palo Alto, CA: Consulting
Psychologists Press.
Nachtigall, C., Kroehne, U., Funke, F., & Steyer, R. (2003). (Why) Should We Use
SEM? Pros and Cons of Structural Equation Modeling. Methods of
Psychological Research Online, 8(2), 1-22. Retrieved from
http://www.dgps.de/fachgruppen/methoden/mpr-
online/issue20/art1/mpr127_11.pdf
Naquin, S. S., & Holton, E. F. (2002). The effects of personality, affectivity, and
work commitment on motivation to improve work through learning. Human
Resource Development Quarterly, 13(4), 357-376. doi:10.1002/hrdq.1038
Nembhard, I. M., & Edmondson, A. C. (2006). Making it safe: The effects of leader
inclusiveness and professional status on psychological safety and
improvement efforts in health care teams. Journal of Organizational
Behavior, 27(7), 941-966. doi:10.1002/job.413
Neuliep, J. W. (1987). The influence of Theory X and Theory Y management styles
336
on the selection of compliance-gaining strategies. Communication Research
Reports, 4(1), 14-19.
Neuliep, J. W. (1996). The influence of Theory X and Y management style on the
perception of ethical behavior in organizations. Journal of Social Behavior
and Personality, 11(2), 301-310.
Newman, M. L., Groom, C. J., Handelman, L. D., & Pennebaker, J. W. (2008).
Gender differences in language use: An analysis of 14,000 text samples.
Discourse Processes, 45(3), 211-236. doi:10.1080/01638530802073712
Neyer, F. J., & Voigt, D. (2004). Personality and social network effects on romantic
relationships: A dyadic approach. European Journal of Personality, 18(4),
279-299. doi:10.1002/per.519
Nisbett, R.E., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and systems of
thought: holistic versus analytic cognition. Psychological Review, 108(2),
291-310. doi:10.1037/0033-295X.108.2.291
Norman, S. M., Avolio, B. J., & Luthans, F. (2010). The impact of positivity and
transparency on trust in leaders and their perceived effectiveness. The
Leadership Quarterly, 21(3), 350-364. doi:10.1016/j.leaqua.2010.03.002
Norris, P., & Epstein, S. (2011). An experiential thinking style: Its facets and
relations with objective and subjective criterion measures. Journal of
Personality, 79(5), 1043-1080. doi:10.1111/j.1467-6494.2011.00718.x
Northouse, P. G. (2007). Leadership: Theory and practice (4th ed.). Thousand Oaks:
SAGE.
Nye, J. L. (2002). The eye of the follower: Information processing effects on
attributions regarding leaders of small groups. Small Group Research, 33(3),
337-360. doi:10.1177/10496402033003003
337
Nye, J. L. (2005). Implicit theories and leaderships in the thick of it: The effects of
prototype matching, group setbacks, and group outcomes. In B. Schyns & J.
R. Meindl (Eds.), Implicit leadership theories: Essays and explorations (pp.
39-61). Greenwich, CT: Information Age Publishing.
Nye, J. L., & Forsyth, D. R. (1991). The effects of prototype-based biases on
leadership appraisals: A test of leadership categorization theory. Small Group
Research, 22(3), 360-379. doi:10.1177/1046496491223005
Nystrom, P. C. (1990). Vertical exchanges and organizational commitments of
American business managers. Group & Organization Management, 15(3),
296-312.
Oc, B., & Bashshur, M. R. (2013). Followership, leadership and social influence.
The Leadership Quarterly, 24(6), 919-934. doi:10.1016/j.leaqua.2013.10.006
Offermann, L. R., Kennedy, J. K., & Wirtz, P. W. (1994). Implicit leadership
theories: Content, structure, and generalizability. The Leadership Quarterly,
5(1), 43-58. doi:10.1016/1048-9843(94)90005-1
Oglensky, B. D. (1995). Socio-psychoanalytic perspectives on the subordinate.
Human Relations, 48(9), 1029-1054. doi:10.1177/001872679504800903
Oldham, G. R., & Cummings, A. (1996). Employee creativity: Personal and
contextual factors at work. The Academy of Management Journal, 39(3),
607-634. doi:10.2307/256657
Oren, L., Tziner, A., Sharoni, G., Iafit A., & Pini, A. (2004). Relations between
leader‐subordinate personality similarity and job attitudes. Journal of
Managerial Psychology, 27(5), 479-496. doi: 10.1108/02683941211235391
Osborne, J. W. (2002). Notes on the use of data transformations. Practical
Assessment, Research & Evaluation, 8(6). Retrieved from Practical
338
Assessment, Research & Evaluation website:
http://PAREonline.net/getvn.asp?v=8&n=6
Osborne, J. W., & Overbay, A. (2004). The power of outliers (and why researchers
should always check for them). Practical Assessment, Research &
Evaluation, 9(6). http://PAREonline.net/getvn.asp?v=9&n=6
Pacini, R., & Epstein, S. (1999a). The relation of rational and experiential
information processing styles to personality, basic beliefs, and the ratio-bias
phenomenon. Journal of Personality and Social Psychology, 76(6), 972-987.
doi:10.1037/0022-3514.76.6.972
Pacini, R., & Epstein, S. (1999b). The interaction of three facets of concrete thinking
in a game of chance. Thinking & reasoning, 5(4), 303-325.
doi:10.1080/135467899393959
Parker, S. K., Williams, H. M., & Turner, N. (2006). Modeling the antecedents of
proactive behavior at work. Journal of Applied Psychology, 91(3), 636-652.
doi:10.1037/0021-9010.91.3.636
Parker, T. C. (1963). Relationships among measures of supervisory behavior, group
behavior, and situational characteristics. Personnel Psychology, 16(4), 319-
334. doi:10.1111/j.1744-6570.1963.tb01279.x
Penton-Voak, I. S., Thomas, J., Gage, S. H., McMurran, M., McDonald, S., &
Munafo, M. R. (2013). Increasing recognition of happiness in ambiguous
facial expressions reduces anger and aggressive behavior. Psychological
Science, 24(5), 688-697. doi:10.1177/0956797612459657
Perugini, M., Gallucci, M., Presaghi, F., & Ercolani, A. P. (2003). The personal
norm of reciprocity. European Journal of Personality, 17(4), 251-283.
doi:10.1002/per.474
339
Peters, L. H., Hartke, D. D., & Pohlmann, J. T. (1985). Fiedler's Contingency Theory
of Leadership: An application of the meta-analysis procedures of Schmidt
and Hunter. Psychological Bulletin, 97(2), 274-285. doi:10.1037/0033-
2909.97.2.274
Peters, T. J., & Austin, N. (1985). A passion for excellence. The leadership
difference. New York: Random House.
Petty, R.E., & Cacioppo, J.T. (1986). The elaboration likelihood model of
persuasion. Advances in Experimental Social Psychology, 19(1), 123-181.
doi:10.1016/s0065-2601(08)60214-2
Phillips, J. S., & Lord, R. G. (1982). Schematic information processing and
perceptions of leadership in problem-solving groups. Journal of Applied
Psychology, 67(4), 486-492. doi:10.1037/0021-9010.67.4.486
Pirola-Merlo, A., Härtel, C., Mann, L., & Hirst, G. (2002). How leaders influence the
impact of affective events on team climate and performance in R&D teams.
The Leadership Quarterly, 13(5), 561-581. doi:10.1016/S1048-
9843(02)00144-3
Plant, E. A., Hyde, H. S., Keltner, D., & Devine, P. G. (2000). The gender
stereotyping of emotions. Psychology of Women Quarterly, 24(1), 81-92.
doi:10.1111/j.1471-6402.2000.tb01024.x
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common
method biases in behavioral research: A critical review of the literature and
recommended remedies. Journal of Applied Psychology, 88(5), 879-903.
doi:10.1037/0021-9010.88.5.879
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research:
Problems and prospects. Journal of Management, 12(4), 531-544.
340
doi:10.1177/014920638601200408
Poole, P. P., Gioia, D. A., & Gray, B. (1989). Influence modes, schema change, and
organizational transformation. The Journal of Applied Behavioral Science,
25(3), 271-289. doi:10.1177/0021886389253004
Potter, E. H., & Fiedler, F. E. (1981). The utilization of staff member intelligence
and experience under high and low stress. The Academy of Management
Journal, 24(2), 361-376. doi:10.2307/255847
Prati, L. M., Douglas, C., Ferris, G. R., Ammeter, A. P., & Buckley, M. R. (2003).
Emotional intelligence, leadership effectiveness, and team outcomes.
International Journal of Organizational Analysis, 11(1), 21-40.
doi:10.1108/eb028961
Proudfoot, J. G., Corr, P. J., Guest, D. E., & Dunn, G. (2009). Cognitive-behavioural
training to change attributional style improves employee well-being, job
satisfaction, productivity, and turnover. Personality and Individual
Differences, 46(2), 147-153. doi:10.1016/j.paid.2008.09.018
Pruitt, D. G. (1998). Social conflict. In Daniel T. Gilbert, S. T. Fiske, & L. Fardner
(Eds.), The Handbook of Social Psychology (4th ed., pp. 470-503). Boston
McGraw-Hill
Quińones, M. A., Ford, J. K., & Teachout, M. S. (1995). The relationship between
work experience and job performance: A conceptual and meta‐analytic
review. Personnel Psychology, 48(4), 887-910. doi:10.1111/j.1744-
6570.1995.tb01785.x
Raghunathan, R., & Trope, Y. (2002). Walking the tightrope between feeling good
and being accurate: Mood as a resource in processing persuasive messages.
Journal of Personality and Social Psychology, 83(3), 510-525.
341
doi:10.1037/0022-3514.83.3.510
Rahaman, H. M. S., Mollah, M. S., & Uddin, M. K. (2010). Big five personality
factors and conflict handling styles. Dhaka University Journal of
Management, 2(1), 13-24.
Raja, U., Johns, G., & Ntalianis, F. (2004). The impact of personality on
psychological contracts. The Academy of Management Journal, 47(3), 350-
367. doi:10.2307/20159586
Rank, J., Carsten, J. M., Unger, J. M., & Spector, P. E. (2007). Proactive customer
service performance: Relationships with individual, task, and leadership
variables. Human Performance, 20(4), 363-390.
doi:10.1080/08959280701522056
Rapisarda, B. A. (2002). The impact of emotional intelligence on work team
cohesiveness and performance. International Journal of Organizational
Analysis, 10(4), 363-379. doi:10.1108/eb028958
Rathunde, K. (2000). Broadening and narrowing in the creative process: A
commentary on Fredrickson's “broaden-and-build” model. Prevention &
Treatment, 3(6), 1-6. doi:10.1037/1522-3736.3.1.36c
Raudenbush, S., Bryk, A., Cheong, Y. K., Congdon, R., & Toit, M. (2011). HLM 7:
Linear and nonlinear modeling. Lincolnwood, IL: Scientific Software
International.
Reed, G. E. (2004). Toxic Leadership. Military Review, 84(4), 67-71.
Reicher, S. D., Haslam, S. A., & Smith, J. R. (2012). Working toward the
experimenter: Reconceptualizing obedience within the Milgram paradigm as
identification-based followership. Perspectives on Psychological Science,
7(4), 315-324. doi:10.1177/1745691612448482
342
Reliable. (n.d.). Roget's 21st Century Thesaurus, Third Edition. Retrieved June 15,
2015, from Thesaurus.com website:
http://www.thesaurus.com/browse/reliable
Rhodes, N., & Wood, W. (1992). Self-esteem and intelligence affect
influenceability: The mediating role of message reception. Psychological
Bulletin, 111(1), 156-171. doi:10.1037/0033-2909.111.1.156
Ritzer, G. (2000). Sociological Theory (5 ed.). New York: McGraw Hill.
Robertson, I. T., & Kinder, A. (1993). Personality and job competences: The
criterion‐related validity of some personality variables. Journal of
Occupational and Organizational Psychology, 66(3), 225-244.
doi:10.1111/j.2044-8325.1993.tb00534.x
Rogers, C. R. (1959). A theory of therapy, personality, and interpersonal
relationships, as developed in the client-centered framework. In S. Koch
(Ed.), Psychology: A study of a science (Vol. 3, pp. 184-256). New York:
McGraw-Hill.
Rosch, E. (1978). Principles of categorization. In E. Rosch & B. B. Lloyd (Eds.),
Cognition and categorization (pp. 27–48 ). HillSDale: Lawrence Erlbaum
Associates.
Rosenberg, S. W., Kahn, S., Tran, T., & Le, M. (1991). Creating a political image:
Shaping appearance and manipulating the vote. Political Behavior, 13(4),
345-367. doi:10.1007/BF00992868
Rotemberg, J. J., & Saloner, G. (1993). Leadership style and incentives.
Management Science, 39(11), 1299-1318. doi:10.1287/mnsc.39.11.1299
Roter, D. (2000). The medical visit context of treatment decisionmaking and the
therapeutic relationship. Health Expectations, 3(1), 17-25.
343
doi:10.1046/j.1369-6513.2000.00073.x
Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different
after all: A cross-discipline view of trust. Academy of Management Review,
23(3), 393-404. doi:10.5465/AMR.1998.926617
Rubin, M., & Hewstone, M. (1998). Social identity theory's self-esteem hypothesis:
A review and some suggestions for clarification. Personality and Social
Psychology Review, 2(1), 40-62. doi:10.1207/s15327957pspr0201_3
Rudebeck, C. E. (2002). Imagination and empathy in the consultation. British
Journal of General Practice, 52(479), 450-453.
Rush, M. C., Thomas, J. C., & Lord, R. G. (1977). Implicit leadership theory: A
potential threat to the internal validity of leader behavior questionnaires.
Organizational Behavior and Human Performance, 20(1), 93-110.
doi:10.1016/0030-5073(77)90046-0
Russell, J. A., & Fehr, B. (1994). Fuzzy concepts in a fuzzy hierarchy: Varieties of
anger. Journal of Personality and Social Psychology, 67(2), 186-205.
doi:10.1037/0022-3514.67.2.186
Sachs, M. L. (1982). Exercise and running: Effects on anxiety, depression, and
psychology. Journal of Humanistic Education and Development, 21(2), 51-
57. doi:10.1002/j.2164-4683.1982.tb00214.x
Sadler-Smith, E., & Shefy, E. (2004). The intuitive executive: Understanding and
applying “gut feel” in decision-making. The Academy of Management
Executive, 18(4), 76-91. doi:10.5465/AME.2004.15268692
Şahin, F. (2012). The mediating effect of leader–member exchange on the
relationship between Theory X and Y management styles and affective
commitment: A multilevel analysis. Journal of Management & Organization,
344
18(02), 159-174. doi:10.1017/S1833367200000936
Salanova, M., Agut, S., & Peiró, J. M. (2005). Linking organizational resources and
work engagement to employee performance and customer loyalty: The
mediation of service climate. Journal of Applied Psychology, 90(6), 1217-
1227. doi:10.1037/0021-9010.90.6.1217
Schaubroeck, J., & Lam, S. S. K. (2002). How similarity to peers and supervisor
influences organizational advancement in different cultures. The Academy of
Management Journal, 45(6), 1120-1136. doi:10.2307/3069428
Schneider, W., & Shiffrin, R.M. (1977). Controlled and automatic human
information processing: I. Detection, search, and attention. Psychological
Review, 84(1), 1-66. doi:10.1037/0033-295X.84.1.1
Schoel, C., Bluemke, M., Mueller, P., & Stahlberg, D. (2011). When autocratic
leaders become an option—Uncertainty and self-esteem predict implicit
leadership preferences. Journal of Personality and Social Psychology,
101(3), 521-540. doi:10.1037/a0023393
Scholl, A., Ellemers, N., Sassenberg, K., & Scheepers, D. (2015). Understanding
power in social context: How power relates to language and communication
in line with responsibilities or opportunities. In R. Schulze & H. Poshwa
(Eds.), The exercise of power in communication: Devices, reception and
reaction (pp. 312-334). Basingstoke, Hampshire: Palgrave Macmillan.
Schoon, H. J. (2008). Person-supervisor fit: Implications for organizational stress,
organizational commitment, and job satisfaction. (Unpublished Masters
Thesis), Clemson University, Clemson, SC. Retrieved from
http://tigerprints.clemson.edu/cgi/viewcontent.cgi?article=1391&context=all
_theses&sei-
345
redir=1&referer=http%3A%2F%2Fscholar.google.com.au%2Fscholar%3Fq
%3Dschoon%2B2008%2Bleadership%26btnG%3D%26hl%3Den%26as_SD
t%3D0%252C5#search=%22schoon%202008%20leadership%22
Schriesheim, C. A., Castro, S. L., & Cogliser, C. C. (1999). Leader-member
exchange (LMX) research: A comprehensive review of theory, measurement,
and data-analytic practices. The Leadership Quarterly, 10(1), 63-113.
doi:10.1016/S1048-9843(99)80009-5
Schriesheim, C. A., & DeNisi, A. S. (1981). Task dimensions as moderators of the
effects of instrumental leadership: A two-sample replicated test of path–goal
leadership theory. Journal of Applied Psychology, 66(5), 589-597.
doi:10.1037/0021-9010.66.5.589
Schriesheim, J. F., & Schriesheim, C. A. (1980). A test of the path-goal theory of
leadership and some suggested directions for future research. Personnel
Psychology, 33(2), 349-370. doi:10.1111/j.1744-6570.1980.tb02356.x
Schubert, T., & Jacoby, J. (2004). Simple slope syntax for test of moderation and
simple slopes for one dichotomous or continuous moderator candidate of One
centered IV in SPSS. Retrieved from
http://www.johannjacoby.de/stattools/SiSSy1.12.5.html.
Schyns, B., & Bligh, M. C. (2007). Introduction to the special issue on the romance
of leadership—In memory of James R. Meindl. Applied Psychology, 56(4),
501-504. doi:10.1111/j.1464-0597.2007.00301.x
Schyns, B., Kiefer, T., Kerschreiter, R., & Tymon, A. (2011). Teaching implicit
leadership theories to develop leaders and leadership: How and why it can
make a difference. Academy of Management Learning & Education, 10(3),
397-408. doi:10.5465/amle.2010.0015
346
Schyns, B., & Meindl, J. R. (2005). An overview of implicit leadership theories and
their application in organization practice. In B. Schyns & J. R. Meindl (Eds.),
Implicit leadership theories: Essays and explorations (pp. 15-36).
Greenwich, CT: Information Age Publishing.
Schyns, B., Paul, T., Mohr, G., & Blank, H. (2005). Comparing antecedents and
consequences of leader–member exchange in a German working context to
findings in the US. European Journal of Work and Organizational
Psychology, 14(1), 1-22. doi:10.1080/13594320444000191
Schyns, B., & Schilling, J. (2013). How bad are the effects of bad leaders? A meta-
analysis of destructive leadership and its outcomes. The Leadership
Quarterly, 24(1), 138-158. doi:10.1016/j.leaqua.2012.09.001
Schyns, B., Tymon, A., Kiefer, T., & Kerschreiter, R. (2012). New ways to
leadership development: A picture paints a thousand words. Management
Learning, 44(1), 11-24. doi:10.1177/1350507612456499
Scott, S. G., & Bruce, R. A. (1994). Determinants of innovative behavior: A path
model of individual innovation in the workplace. Academy of Management
Journal, 37(3), 580-607. doi:10.2307/256701
Sears, G.J., & Holmvall, C.M. (2010). The joint influence of supervisor and
subordinate emotional intelligence on leader–member exchange. Journal of
Business and Psychology, 25(4), 593-605. doi:10.1007/s10869-009-9152-y
Seers, A., & Graen, G. (1984). The dual attachment concept: A longitudinal
investigation of the combination of task characteristics and leader—member
exchange. Organizational Behavior and Human Performance, 33(3), 283-
306. doi:10.1016/0030-5073(84)90025-4
Segal, D. (1985). Management, leadership and the future battlefield. In J. G. Hunt &
347
J. D. Blair (Eds.), Leadership on the future battlefield. Washington, DC:
Pergamon-Brassey.
Seibert, S. E., Crant, J. M., & Kraimer, M. L. (1999). Proactive personality and
career success. Journal of Applied Psychology, 84(3), 416-427.
doi:10.1037/0021-9010.84.3.416
Seibert, S. E., Kraimer, M. L., & Crant, J. M. (2001). A longitudinal model linking
proactive personality and career success. Personnel Psychology, 54(4), 845-
875.
Seligman, M. E., & Maier, S. F. (1967). Failure to escape traumatic shock. Journal
of Experimental Psychology, 74(1), 1-9. doi:10.1037/h0024514
Seyranian, V. (2009). Contingency theories of leadership. In J. M. Levine & M. A.
Hogg (Eds.), Encyclopedia of group processes & intergroup relations (pp.
152-156). Thousand Oaks, CA: SAGE.
Shalit, A., Popper, M., & Zakay, D. (2010). Followers' attachment styles and their
preference for social or for personal charismatic leaders. Leadership &
Organization Development Journal, 31(5), 458-472.
doi:10.1108/01437731011056461
Shamay-Tsoory, S. G., Aharon-Peretz, A., & Perry, D. (2009). Two systems for
empathy: A double dissociation between emotional and cognitive empathy in
inferior frontal gyrus versus ventromedial prefrontal lesions. Brain, 132(3),
617-627. doi:10.1093/brain/awn279
Shamir, B. (1991). Meaning, self and motivation in organizations. Organization
Studies, 12(3), 405-424. doi:10.1177/017084069101200304
Shamir, B. (2007). From passive recipients to active co-producers: Followers' roles
in the leadership process Follower-centered perspectives on leadership (pp.
348
ix-xxxix). Greenwich, CT: Information Age Publishing.
Shamir, B., House, R. J., & Arthur, M. B. (1993). The motivational effects of
charismatic leadership: A self-concept based theory. Organization Science,
4(4), 577-594. doi:10.1287/orsc.4.4.577
Shin, J. S., & Zhou, J. (2003). Transformational leadership, conservation, and
creativity: Evidence from Korea. The Academy of Management Journal,
46(6), 703-714. doi:10.2307/30040662
Shondrick, S. J., Dinh, J. E., & Lord, R. G. (2010). Developments in implicit
leadership theory and cognitive science: Applications to improving
measurement and understanding alternatives to hierarchical leadership. The
Leadership Quarterly, 21(6), 959-978. doi:10.1016/j.leaqua.2010.10.004
Shondrick, S. J., & Lord, R. G. (2010). Implicit leadership and followership theories:
Dynamic structures for leadership perceptions, memory, leader-follower
processes. In G. P. Hodgkinson & J. K. Ford (Eds.), International Review of
Industrial and Organizational Psychology (Vol. 25, pp. 1-34). Chichester,
West Sussex: John Wiley & Sons
Simoens, D. (2014). Rational thinking and behavioural coping: Agreement in
leaders' and subordinates' perceptions of transformational leadership.
(Unpublished Masters Thesis), Murdoch University, Perth, Australia.
Sin, H. P., Nahrgang, J. D., & Morgeson, F. P. (2009). Understanding why they don't
see eye to eye: An examination of leader-member exchange (LMX)
agreement. Journal of Applied Psychology, 94(4), 1048-1057.
doi:10.1037/a0014827
Slepian, M. L., Ferber, S. N., Gold, J. M., & Rutchick, A. M. (2015). The cognitive
consequences of formal clothing. Social Psychological and Personality
349
Science, 6(6), 661-668. doi:10.1177/1948550615579462
Smyth, T. (1996). Reinstating the person in the professional: Reflections on empathy
and aesthetic experience. Journal of Advanced Nursing, 24(5), 932-937.
doi:10.1111/j.1365-2648.1996.tb02928.x
Sloman, S.A. (1996). The empirical case for two systems of reasoning.
Psychological Bulletin, 119(1), 3-22. doi:10.1037/0033-2909.119.1.3
Smith, E.R., & DeCoster, J. (2000). Dual-process models in social and cognitive
psychology: Conceptual integration and links to underlying memory systems.
Personality and Social Psychology Review, 4(2), 108-131.
doi:10.1207/S15327957PSPR0402_01
Sparrowe, R. T., & Liden, R. C. (1997). Process and structure in leader-member
exchange. The Academy of Management Review, 22(2), 522-552.
doi:10.5465/AMR.1997.9707154068
Stacciarini, J. R., & Tróccoli, B. T. (2004). Occupational stress and constructive
thinking: Health and job satisfaction. Journal of Advanced Nursing, 46(5),
480-487. doi:10.1111/j.1365-2648.2004.03022.x
Stanovich, K.E., & West, R.F. (2000). Individual differences in reasoning:
Implications for the rationality debate?. Behavioral and Brain Sciences,
23(5), 645-726. Retrieved from:
http://www.keithstanovich.com/Site/Research_on_Reasoning_files/bbs2000_
1.pdf
Stech, E. L. (2007). Psychodynamic approach. In P. G. Northouse (Ed.), Leadership:
Theory and practice (4th ed., pp. 237–264). Thousand Oaks, CA: SAGE.
Stein, D. J. (1997). Introduction: Cognitive science and the unconscious. In D. J.
Stein (Ed.), Cognitive science and the unconscious. Washington, DC:
350
American Psychiatric Press.
Stogdill, R. M. (1948). Personal factors associated with leadership: A survey of the
literature. The Journal of Psychology, 25(1), 35-71.
doi:0.1080/00223980.1948.9917362
Stogdill, R. M. (1974). Handbook of leadership: A survey of the literature. New
York: Free Press.
Stone, E. F., & Hollenbeck, J. R. (1989). Clarifying some controversial issues
surrounding statistical procedures for detecting moderator variables:
Empirical evidence and related matters. Journal of Applied Psychology,
74(1), 3-10. doi:10.1037/0021-9010.74.1.3
Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social
behavior. Personality and Social Psychology Review, 8(3), 220-247.
doi:10.1207/s15327957pspr0803_1
Strand, E. B. (2006). Enhanced communication by developing a non-anxious
presence: A key attribute for the successful veterinarian. Journal of
Veterinary Medical Education, 42(3), 65-70. doi:10.3138/jvme.33.1.65
Strong. (n.d.). Roget's 21st Century Thesaurus, Third Edition. Retrieved May 20,
2015, from Thesaurus.com website: http://www.thesaurus.com/browse/strong
Strauss, J.P., Barrick, M.R., & Connerley, M.L. (2001). An investigation of
personality similarity effects (relational and perceived) on peer and
supervisor ratings and the role of familiarity and liking. Journal of
Occupational and Organizational Psychology, 74(5), 637-657.
Doi:10.1348/096317901167569
Suls, J., Martin, R., & David, J. P. (1998). Person-environment fit and its limits:
Agreeableness, neuroticism, and emotional reactivity to interpersonal
351
conflict. Personality and Social Psychology Bulletin, 24(1), 88-98.
doi:10.1177/0146167298241007
Swann, W. B., & Read, S. J. (1981). Self-verification processes: How we sustain our
self-conceptions. Journal of Experimental Social Psychology, 17(4), 351-
372. doi:10.1016/0022-1031(81)90043-3
Sy, T. (2010). What do you think of followers? Examining the content, structure, and
consequences of implicit followership theories. Organizational Behavior and
Human Decision Processes, 113(2), 73-84. doi:10.1016/j.obhdp.2010.06.001
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.).
Boston Pearson/Allyn & Bacon.
Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach's alpha.
International Journal of Medical Education, 2(1), 53-55.
doi:10.5116/ijme.4dfb.8dfd
Tett, R. P., Jackson, D. N., & Rothstein, M. (1991). Personality measures as
predictors of job performance: A meta-analytic review. Personnel
Psychology, 44(4), 703-742. doi:10.1111/j.1744-6570.1991.tb00696.x
Thacker, R. A., & Wayne, S. J. (1995). An examination of the relationship between
upward influence tactics and assessments of promotability. Journal of
Management, 21(4), 739-756. doi:10.1177/014920639502100408
Thompson, J. A. (2005). Proactive personality and job performance: A social capital
perspective. Journal of Applied Psychology, 90(5), 1011-1017.
doi:10.1037/0021-9010.90.5.1011
Thompson, M. L. (1978). Selection of variables in multiple regression: Part I. A
review and evaluation. International Statistical Review / Revue
Internationale de Statistique, 46(1), 1-19. doi:10.2307/1402505
352
Tierney, P. (2016). LMX and Creativity. In T.N. Bauer & B. Erdogan (Eds.). The
Oxford Handbook of Leader-Member Exchange, (pp. 175-188). New York:
Oxford University Press.
Tierney, P., & Farmer, S. M. (2002). Creative self-efficacy: Its potential antecedents
and relationship to creative performance. The Academy of Management
Journal, 45(6), 1137-1148. doi:10.2307/3069429
Toegel, G., Liu, F., Coughlan, S., & Perrinjaquet, M. (2011). Leadership dyads:
Playing to your Strengths. Insights@IMD, 4, 1-4. Retrieved from
International Institute for Management Development. website:
http://www.imd.org/research/publications/upload/6-Leadership-Dyads-w-no-
05-11-12-2.pdf
Tost, L. P., Gino, F., & Larrick, R. (2013). When power makes others speechless:
The negative impact of leader power on team performance. The Academy of
Management Journal, 56(5), 1465-1486. doi:10.5465/amj.2011.0180
Tracy, L. (1987). Consideration and initiating structure: Are they basic dimensions
of leader behavior? Social Behavior and Personality, 15(1), 21-33.
doi:10.2224/sbp.1987.15.1.21
Tukey, J.W. (1977). Exploratory Data Analysis. Reading, PA: Addison-Wesley
Turban, D. B., & Jones, A. P. (1988). Supervisor-subordinate similarity: Types,
effects, and mechanisms. Journal of Applied Psychology, 73(2), 228-234.
doi:10.1037/0021-9010.73.2.228
Uhl-Bien, M., & Graen, G. B. (1993). Leadership-making in self-managing
professional work teams: An empirical investigation. In K. E. Clark, M. B.
Clark, & D. P. Campbell (Eds.), The Impact of Leadership (pp. 379-387).
West Orange, NJ: Leadership Library of America.
353
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. doi:10.1016/j.leaqua.2013.11.007
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. doi:10.1016/j.leaqua.2013.11.007
Uhlmann, E. L., Leavitt, K., Menges, J. I., Koopman, J., Howe, M., & Johnson, R. E.
(2012). Getting explicit about the implicit: A taxonomy of implicit measures
and guide for their use in organizational research. Organizational Research
Methods, 15(4), 553-601. doi:10.1177/1094428112442750
van Gils, S., van Quaquebeke, N., & van Knippenberg, D. (2010). The X-factor: On
the relevance of implicit leadership and followership theories for leader–
member exchange agreement. European Journal of Work and Organizational
Psychology, 19(3), 333-363. doi:10.1080/13594320902978458
van Kleef, G. A., Homan, A. C., Beersma, B., & van Knippenberg, D. (2010). On
angry leaders and agreeable followers. How leaders' emotions and followers'
personalities shape motivation and team performance. Psychol Sci, 21(12),
1827-1834. doi:10.1177/0956797610387438
van Scotter, J., Motowidlo, S. J., & Cross, T. C. (2000). Effects of task performance
and contextual performance on systemic rewards. Journal of Applied
Psychology, 85(4), 526-535. doi:10.1037/0021-9010.85.4.526
van Vugt, M., Hogan, R., & Kaiser, R. B. (2008). Leadership, followership, and
evolution: some lessons from the past. American Psychologist, 63(3), 182-
196. doi:10.1037/0003-066X.63.3.182
Vanderslice, V. J. (1988). Separating leadership from leaders: An assessment of the
354
effect of leader and follower roles in organizations. Human Relations, 41(9),
677-696. doi:10.1177/001872678804100903
Vecchio, R. P. (1987). Situational Leadership Theory: An examination of a
prescriptive theory. Journal of Applied Psychology, 72(3), 444-451.
doi:10.1037/0021-9010.72.3.444
Velsor, E. V., & McCauley, C. D. (2010). Introduction: Our view of Leadership
Development. In C. D. McCauley & E. V. Velsor (Eds.), The Center For
Creative Leadership handbook of leadership development (2nd ed., pp. 1-22).
San Francisco, CA: Jossey-Bass.
Venkataramani, V., Green, S. G., & Schleicher, D. J. (2010). Well-connected
leaders: The impact of leaders' social network ties on LMX and members'
work attitudes. Journal of Applied Psychology, 95(6), 1071-1084.
doi:10.1037/a0020214
Verhulst, B., Lodge, M., & Lavine, H. (2010). The attractiveness halo: Why some
candidates are perceived more favorably than others. Journal of Nonverbal
Behavior, 34(2), 111-117. doi:10.1007/s10919-009-0084-z
Vestewig, R. E., & Moss, M. K. (1976). The relationship of extraversion and
neuroticism to two measures of assertive behavior. The Journal of
Psychology, 93(1), 141-146. doi:10.1080/00223980.1976.9921385
Wakabayashi, M., & Graen, G. B. (1984). The Japanese career progress study: A 7-
year follow-up. Journal of Applied Psychology, 69(4), 603-614.
doi:10.1037/0021-9010.69.4.603
Walster, E., Walster, G. W., & Berscheid, E. (1978). Equity: Theory and research.
Boston, MA: Allyn & Bacon.
Walumbwa, F. O., Avolio, B. J., Gardner, W. L., Wernsing, T. S., & Peterson, S. J.
355
(2008). Authentic leadership: Development and validation of a theory-based
measure. Journal of Management, 34(1), 89-126
doi:10.1177/0149206307308913
Walumbwa, F. O., & Schaubroeck, J. (2009). Leader personality traits and employee
voice behavior: Mediating roles of ethical leadership and work group
psychological safety. Journal of Applied Psychology, 94(5), 1275-1286.
doi:10.1037/a0015848
Wanberg, C. R., & Kammeyer-Mueller, J. D. (2000). Predictors and outcomes of
proactivity in the socialization process. Journal of Applied Psychology,
85(3), 373-385. doi:10.1037/0021-9010.85.3.373
Watson, B. (1982). An analysis of communication patterns: A method for
discriminating leader and subordinate roles. The Academy of Management
Journal, 25(1), 107-120. doi:10.2307/256027
Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to
experience aversive emotional states. Psychological Bulletin, 96(3), 465-490.
doi:10.1037/0033-2909.96.3.465
Watson, W., & Michaelsen, L. (1984). Task performance information and leader
participation behavior: Effect on leader-subordinate interaction, frustration,
and future productivity. Group & Organization Management, 9(1), 121-144.
doi:10.1177/105960118400900107
Waugh, C. E., & Fredrickson, B. L. (2006). Nice to know you: Positive emotions,
self–other overlap, and complex understanding in the formation of a new
relationship. The Journal of Positive Psychology, 1(2), 93-106.
doi:10.1080/17439760500510569
Wechsler, D. (2008). Wechsler adult intelligence scale–Fourth Edition (WAIS–IV).
356
San Antonio, TX: NCS Pearson
Weick, K. E. (1995). Sensemaking in organizations. London: SAGE
Weisberg, R. W. (1999). Creativity and knowledge: A challenge to theories. In R. J.
Sternberg (Ed.), Handbook of creativity (pp. 226-250). Cambridge:
Cambridge University Press.
Wexley, K. N., & Pulakos, E. D. (1982). Sex effects on performance ratings on
manager–subordinate dyads: A field study. Journal of Applied Psychology,
67(4), 433-439. doi:10.1037/0021-9010.67.4.433
Whicker, M. L. (1996). Toxic leaders: When organizations go bad. Westport, Conn:
Quorum Books.
White, J. K., Hendrick, S. S., & Hendrick, C. (2004). Big five personality variables
and relationship constructs. Personality and Individual Differences, 37(7),
1519-1530. doi:10.1016/j.paid.2004.02.019
Whiteley, P., Sy, T., & Johnson, S. K. (2012). Leaders' conceptions of followers:
Implications for naturally occurring Pygmalion effects. The Leadership
Quarterly, 23(5), 822-834. doi:10.1016/j.leaqua.2012.03.006
Wieseke, J., Ahearne, M., Lam, S. K., & Dick, R. (2009). The role of leaders in
internal marketing. Journal of Marketing, 73(2), 123-145.
doi:10.1509/jmkg.73.2.123
Williams, J. E., & Best, D. L. (1990). Measuring sex stereotypes: A multination
study. Thousand Oaks, CA: SAGE.
Willner, A. R. (1984). The spellbinders: Charismatic political leadership. . New
Haven, CT: Yale University Press.
Wilson, K. S., Sin, H. P., & Conlon, D. E. (2010). What about the leader in leader-
member exchange? The impact of resource exchanges and substitutability on
357
the leader. The Academy of Management Review, 35(3), 358-372.
doi:10.5465/AMR.2010.51141654
Winkler, I. (2010). Contemporary leadership theories: Enhancing the understanding
of the complexity, subjectivity and dynamic of leadership. Heidelberg:
Springer-Verlag.
Wofford, J. C., & Goodwin, V. L. (1994). A cognitive interpretation of transactional
and transformational leadership theories. The Leadership Quarterly, 5(2),
161-186.
Wofford, J. C., Whittington, J. L., & Goodwin, V. L. (2001). Follower motive
patterns as situational moderators for transformational leadership
effectiveness. Journal of Managerial Issues, 13(2), 196-211.
Woltman, H., Feldstain, A., MacKay, J. C., & Rocchi, M. (2012). An introduction to
hierarchical linear modeling. Tutorials in Quantitative Methods for
Psychology, 8(1), 52-69.
Wood, S. (2008). Job characteristics, employee voice and well‐being in Britain.
Industrial Relations Journal, 39(2), 153-168. doi:10.1111/j.1468-
2338.2007.00482.x
Xin, K. R., & Pelled, L. H. (2003). Supervisor–subordinate conflict and perceptions
of leadership behavior: A field study. The Leadership Quarterly, 14(1), 25-
40. doi:10.1016/S1048-9843(02)00185-6
Xiaqi, D., Kun, T., Chongsen, Y., & Sufang, G. (2012). Abusive supervision and
LMX. Chinese Management Studies, 6(2), 257-270.
doi:10.1108/17506141211236695
Xu, J., Liu, Y., & Guo, Y. (2014). The role of subordinate emotional masking in
leader–member exchange and outcomes: A two-sample investigation. Journal
358
of Business Research, 67(2), 100-107. doi:10.1016/j.jbusres.2012.11.011
Yaremko, R. M., Harari, H., Harrison, R. C., & Lynn, E. (1986). Handbook of
research and quantitative methods in psychology: For students and
professionals. HillSDale, NJ: Lawrence Erlbaum.
Yeakley, F. R. (1982). Communication style preferences and adjustments as an
approach for studying effects of similarity in psychological type. Research in
Psychological Type, 5(1), 30-48.
Yukl, G. (1998). Leadership in organizations (4th ed.). Englewood Cliffs, NJ:
Prentice Hall.
Yukl, G. (1999). An evaluation of conceptual weaknesses in transformational and
charismatic leadership theories. The Leadership Quarterly, 10(2), 285-305.
doi:10.1016/S1048-9843(99)00013-2
Yukl, G. A. (2002). Leadership in organizations (5th ed.). Englewood Cliffs, NJ:
Prentice Hall.
Yukl, G., & Fu, P. P. (1999). Determinants of delegation and consultation by
managers. Journal of Organizational Behavior, 20(2), 219-232.
doi:10.1002/(SICI)1099-1379(199903)20:2<219::AID-JOB922>3.0.CO;2-8
Yukl, G., O'Donnell, M., & Taber, T. B. (2009). Influence of leader behaviors on the
leader-member exchange relationship. Journal of Managerial Psychology,
24(4), 289-299. doi:10.1108/02683940910952697
Zaccaro, S. J. (2007). Trait-based perspectives of leadership. American Psychologist,
62(1), 6-16. doi:10.1037/0003-066X.62.1.6
Zaccaro, S. J., Kemp, C., & Bader, P. (2004). Leader traits and attributes. In J.
Antonakis, A. T. Cianciolo, & R. J. Sternberg (Eds.), The nature of
leadership (pp. 101-124). Thousand Oaks, CA: Sage.
359
Zaccaro, S. J., Rittman, A. L., & Marks, M. A. (2001). Team leadership. The
Leadership Quarterly, 12(4), 451-483. doi:10.1016/S1048-9843(01)00093-5
Zaleznik, A. (1965). The dynamics of subordinacy. Harvard Business Review, 43(3),
119-131.
Zeidner, M., & Matthews, G. (2000). Intelligence and personality. In Robert J.
Sternberg (Ed.), Handbook of intelligence (pp. 581-610). New York:
Cambridge University Press.
Zhang, Z., Wang, M., & Shi, J. (2012). Leader-follower congruence in proactive
personality and work outcomes: The mediating role of leader-member
exchange. The Academy of Management Journal, 55(1), 111-130.
doi:10.5465/amj.2009.0865
360
CHAPTER 8 APPENDICES
Below is a list of the contents in each appendix. Please note that the Constructive
Thinking Inventory (Epstein, 2001) is not included due to licencing and copyright
restrictions.
APPENDIX A
Rational Experiential Multimodal Inventory (Norris & Epstein, 2011)
APPENDIX B
ILT Questionnaire (Offerman, Kennedy & Wirtz, 1994)
APPENDIX C
IFT Questionnaire (Sy, 2010)
APPENDIX D
Leader-Member Exchange (LMX-MDM; Liden & Maslyn, 1998)
APPENDIX E
Exploratory Analyses of Gender and Sub-Scales Relationships with Implicit
Leadership Theories
APPENDIX F
Exploratory Analyses of Relationships between
Leader Naïve-Optimism Sub-scales and Follower LMX
361
APPENDIX A
Rational Experiential Multimodal Inventory (Norris & Epstein, 2011)
Instructions: Please read and rate the following statements about your feelings,
beliefs, and behaviours using the scale below. Work rapidly; first impressions are as
good as any.
Statement
Def
init
ely
Fal
se
Mo
stly
Fal
se
Un
dec
ided
or
Eq
ual
ly T
rue
&
Fal
se
Mo
stly
Tru
e
Def
init
ely
Tru
e
I’m not a very spontaneous person. 1 2 3 4 5
I am not very good in solving problems that
require careful logical analysis. 1 2 3 4 5
I trust my initial feelings about people. 1 2 3 4 5
I enjoy reading things that evoke visual images. 1 2 3 4 5
I often go by my instincts when deciding on a
course of action. 1 2 3 4 5
Reasoning things out carefully is not one of my
strong points. 1 2 3 4 5
I’d rather be upset sometimes and happy
sometimes, than always feel calm. 1 2 3 4 5
I have favourite poems and paintings that mean a
lot to me. 1 2 3 4 5
I try to avoid situations that require thinking in
depth about something. 1 2 3 4 5
When I’m sad, it’s often a very strong feeling. 1 2 3 4 5
I tend to describe things by using images or
metaphors, or creative comparisons. 1 2 3 4 5
Knowing the answer without understanding the
reasoning behind it is good enough for me. 1 2 3 4 5
Sometimes I like to just sit back and watch things
happen. 1 2 3 4 5
I have a logical mind. 1 2 3 4 5
I can clearly picture or remember some sculpture
or natural object (not alive) that I think is very
beautiful.
1 2 3 4 5
I identify strongly with characters in movies or
books I read. 1 2 3 4 5
Using logic usually works well for me in figuring
out problems in my life. 1 2 3 4 5
My emotions don’t make much difference in my
life. 1 2 3 4 5
I am not a very analytical thinker. 1 2 3 4 5
362
Statement
Def
init
ely
Fal
se
Mo
stly
Fal
se
Un
dec
ided
or
Eq
ual
ly T
rue
&
Fal
se
Mo
stly
Tru
e
Def
init
ely
Tru
e
When I travel or drive anywhere, I always watch
the landscape and scenery. 1 2 3 4 5
I almost never think in visual images. 1 2 3 4 5
I don’t like to have to do a lot of thinking. 1 2 3 4 5
Emotions don’t really mean much; they come and
go. 1 2 3 4 5
When I have a strong emotional experience, the
effect stays with me for a long time. 1 2 3 4 5
I enjoy intellectual challenges. 1 2 3 4 5
Things that make me feel emotional don’t seem to
affect other people as much. 1 2 3 4 5
Everyday experiences often evoke strong feelings
in me. 1 2 3 4 5
I am much better at figuring things out logically
than most people. 1 2 3 4 5
I don’t react emotionally to scary movies or
books as much as most people do. 1 2 3 4 5
I enjoy imagining things. 1 2 3 4 5
When I’m happy, the feeling is usually more like
contentment than like exhilaration or excitement. 1 2 3 4 5
I like to rely on my intuitive impressions. 1 2 3 4 5
Art is really important to me. 1 2 3 4 5
I don’t think it is a good idea to rely on ones
intuition for important decisions. 1 2 3 4 5
I enjoy problems that require hard thinking. 1 2 3 4 5
I tend to use my heart as a guide for my actions. 1 2 3 4 5
I enjoy learning by doing something, instead of
figuring it out first. 1 2 3 4 5
I can often tell how people feel without them
having to say anything. 1 2 3 4 5
I generally don’t depend on my feelings to help
me make decisions. 1 2 3 4 5
For me, descriptions of actual people’s
experiences are more convincing than discussions
about facts.
1 2 3 4 5
I prefer complex to simple problems. 1 2 3 4 5
My anger is often very intense. 1 2 3 4 5
363
APPENDIX B
ILT Questionnaire (Offerman, Kennedy & Wirtz, 1994)
Instructions: For each trait below, please indicate the extent to which it is
characteristic of your IDEAL LEADER in a business setting ranging from 1 (not at
all characteristic) to 10 (extremely characteristic). For each trait use whichever
definition is meaningful to you.
Trait
No
t at
all
Ch
arac
teri
stic
Ex
trem
ely
Ch
arac
teri
stic
Sympathetic 1 2 3 4 5 6 7 8 9 10
Energetic 1 2 3 4 5 6 7 8 9 10
Dedicated 1 2 3 4 5 6 7 8 9 10
Well-groomed 1 2 3 4 5 6 7 8 9 10
Domineering 1 2 3 4 5 6 7 8 9 10
Masculine 1 2 3 4 5 6 7 8 9 10
Intellectual 1 2 3 4 5 6 7 8 9 10
Strong 1 2 3 4 5 6 7 8 9 10
Sensitive 1 2 3 4 5 6 7 8 9 10
Charismatic 1 2 3 4 5 6 7 8 9 10
Motivated 1 2 3 4 5 6 7 8 9 10
Attractive 1 2 3 4 5 6 7 8 9 10
Pushy 1 2 3 4 5 6 7 8 9 10
Educated 1 2 3 4 5 6 7 8 9 10
Bold 1 2 3 4 5 6 7 8 9 10
Compassionate 1 2 3 4 5 6 7 8 9 10
Inspiring 1 2 3 4 5 6 7 8 9 10
Hard-working 1 2 3 4 5 6 7 8 9 10
Well-dressed 1 2 3 4 5 6 7 8 9 10
Dominant 1 2 3 4 5 6 7 8 9 10
Intelligent 1 2 3 4 5 6 7 8 9 10
Obnoxious 1 2 3 4 5 6 7 8 9 10
Enthusiastic 1 2 3 4 5 6 7 8 9 10
364
Trait
No
t at
all
Ch
arac
teri
stic
Ex
trem
ely
Ch
arac
teri
stic
Goal-oriented 1 2 3 4 5 6 7 8 9 10
Classy 1 2 3 4 5 6 7 8 9 10
Manipulative 1 2 3 4 5 6 7 8 9 10
Wise 1 2 3 4 5 6 7 8 9 10
Sincere 1 2 3 4 5 6 7 8 9 10
Dynamic 1 2 3 4 5 6 7 8 9 10
Power-hungry 1 2 3 4 5 6 7 8 9 10
Knowledgeable 1 2 3 4 5 6 7 8 9 10
Warm 1 2 3 4 5 6 7 8 9 10
Conceited 1 2 3 4 5 6 7 8 9 10
Clever 1 2 3 4 5 6 7 8 9 10
Forgiving 1 2 3 4 5 6 7 8 9 10
Loud 1 2 3 4 5 6 7 8 9 10
Selfish 1 2 3 4 5 6 7 8 9 10
Male 1 2 3 4 5 6 7 8 9 10
Understanding 1 2 3 4 5 6 7 8 9 10
Helpful 1 2 3 4 5 6 7 8 9 10
Demanding 1 2 3 4 5 6 7 8 9 10
365
APPENDIX C
IFT Questionnaire (Sy, 2010)
Instructions: For each trait below, please indicate the extent to which it is
characteristic of your IDEAL follower in a business setting ranging from 1 (not at
all characteristic) to 10 (extremely characteristic). For each trait use whichever
definition is meaningful to you.
Trait
No
t at
all
Ch
arac
teri
stic
Ex
trem
ely
Ch
arac
teri
stic
Hardworking 1 2 3 4 5 6 7 8 9 10
Outgoing 1 2 3 4 5 6 7 8 9 10
Inexperienced 1 2 3 4 5 6 7 8 9 10
Excited 1 2 3 4 5 6 7 8 9 10
Loyal 1 2 3 4 5 6 7 8 9 10
Arrogant 1 2 3 4 5 6 7 8 9 10
Bad-Temper 1 2 3 4 5 6 7 8 9 10
Happy 1 2 3 4 5 6 7 8 9 10
Team Player 1 2 3 4 5 6 7 8 9 10
Easily Influenced 1 2 3 4 5 6 7 8 9 10
Follows Trends 1 2 3 4 5 6 7 8 9 10
Soft-spoken 1 2 3 4 5 6 7 8 9 10
Uneducated 1 2 3 4 5 6 7 8 9 10
Productive 1 2 3 4 5 6 7 8 9 10
Goes above and
beyond 1 2 3 4 5 6 7 8 9 10
Rude 1 2 3 4 5 6 7 8 9 10
Slow 1 2 3 4 5 6 7 8 9 10
Reliable 1 2 3 4 5 6 7 8 9 10
366
APPENDIX D
Leader-Member Exchange (LMX-MDM; Liden & Maslyn, 1998)
Instructions: the statements below ask you about your relationship with your
officially designated manager or supervisor. Indicate the degree to which a
statement is true of you. Do not spend too much time on any one item.
Statement
Str
ong
ly
Dis
agre
e
Dis
agre
e
Dis
agre
e
So
mew
hat
Un
dec
ided
Ag
ree
So
mew
hat
Ag
ree
Str
ong
ly A
gre
e
I admire my supervisor's professional
skills. 1 2 3 4 5 6 7
I am impressed with my supervisor's
knowledge of his/her job. 1 2 3 4 5 6 7
I am willing to apply extra efforts,
beyond those normally required, to
meet my supervisor's work goals.
1 2 3 4 5 6 7
I do not mind working my hardest for
my supervisor. 1 2 3 4 5 6 7
I do work for my supervisor that goes
beyond what is specified in my job
description.
1 2 3 4 5 6 7
I like my supervisor very much as a
person. 1 2 3 4 5 6 7
I respect my supervisor's knowledge
of and competence on the job. 1 2 3 4 5 6 7
My supervisor defends my work
actions to a superior, even without
complete knowledge of the issue in
question.
1 2 3 4 5 6 7
My supervisor is a lot of fun to work
with. 1 2 3 4 5 6 7
My supervisor is the kind of person
one would like to have as a friend. 1 2 3 4 5 6 7
My supervisor would come to my
defence if I were "attacked" by others. 1 2 3 4 5 6 7
My supervisor would defend me to
others in the organisation if I made an
honest mistake.
1 2 3 4 5 6 7
367
APPENDIX E
Exploratory Analyses of Relationships between Follower Information Processing Style and
Constructive Thinking with Implicit Followership Theories
Table 8.1 presents exploratory correlations between REIm main and sub-scale relations with ILT dimensions by gender. Table 8.2 presents
exploratory correlations between CTI main and sub-scale relations with ILT dimensions by gender.
Table 8.1
Exploratory Correlations between REIm and ILT Dimensions
ILT Dimensions
Sensitivity Tyranny˄ Intelligence Attractiveness Dedication
G M F G M F G M F G M F G M F
Rational -.02 .13 -.11 .04 .24* -.07 .35** .40** .32** -.05 .20 -.18** .02 .13 -.01
Experiential .33** .37** .35** -.01 -.04 .02 .01 .09 .01 .13* .15 .09 .29** .24* .27**
Intuition .35** .27* .39** .03 -.07 .08 -.10 -.03 -.11 .27** .29** .25** .19** .18 .17*
Emotionality .18** .23* .18** -.05 -.17 .02 .07 .08 .09 -.02 -.13 -.01 .21** .07 .22**
Imagination .20** .25* .18** -.04 .11 -.10 .03 .11 .02 .02 .18 .06 .21** .25* .18**
Notes: *Indicates p < .05; **Indicates p < .01; ˄Relationships indicated by Spearman’s rho; Group (G) n = 311; Male (M) n = 91; Female (F) n = 220.
Significant differences in male and female correlations are bolded.
368
Table 8.2
Correlations between CTI and Five ILT Factors
ILT Dimensions
Sensitivity Tyranny˄ Intelligence Attractiveness Dedication
G M F G M F G M F G M F G M F
Global Constructive Thinking .05 -.14 .12 .04 .29* -.06 -.25** -.26* -.26** -.01 .17 -.09 .02 .01 .04
Emotional Coping -.07 -.33** .03 .02 .23* -.07 -.30** -.33** -.30** -.02 .13 -.08 -.05 -.12 -.01
Self-Acceptance -.04 -.28* .05 -.01 .17 -.08 -.29** -.31** -.30** -.07 .04 -.11 -.04 -.15 .04
Absence of Negative
Overgeneralisation -.04 -.16 .01 .02 .19 -.05 -.19** -.25* -.17* -.02 .16 -.11 .03 .01 .04
Non-Sensitivity -.11 -.36** .00 .02 .20 -.06 -.24** -.22 -.25** .00 .09 -.05 -.06 -.05 -.07
Absence of Dwelling -.02 -.26* .08 .03 .26* -.06 -.34** -.36** -.33** .00 .14 -.06 -.03 -.06 -.01
Behavioural Coping .19** .17 .21** .08 .26* .01 -.06 -.01 -.08 .06 .32** -.11 .13* .25* .05
Positive Thinking .09 .08 .11 .07 .22 .01 -.12* -.15 -.11 .05 .22 -.06 .11 .10 .08
Action Orientation .19** .19 .20** .09 .25* .02 -.04 .05 -.07 .05 .33** -.11 .12* .30** .02
Conscientiousness .25** .32** .23** .03 .09 -.01 .09 .29* .03 .07 .31** -.03 .09 .27* .03
Personal Superstitious
Thinking -.08 .04 -.11 .03 -.11 .09 .10 .11 .11 .06 -.10 .10 -.06 .05 -.15*
Categorical Thinking -.10 .06 -.14 .09 .15 .12 .09 -.03 .14 .23** .23* .20** -.02 .12 -.13
Polarised Thinking -.02 .10 -.05 .10 .14 .11 -.02 -.14 .02 .28** .24* .28** -.02 .06 -.09
Distrust of Others -.12* -.07 -.14 .04 -.09 .10 .14* .07 .18* .10 .06 .09 -.03 .05 -.01
Intolerance -.23** -.14 -.26** -.00 -.04 .03 .05 -.28* .16* .05 -.04 .06 -.10 -.07 -.16*
Esoteric Thinking .08 -.09 .13 .11 .22 .08 -.03 -.21 .01 .21** .13 .23** .06 -.01 .06
Belief in the Unusual .08 -.06 .11 .09 .10 .07 -.05 -.14 -.02 .22** .19 .23** .07 .06 .07
Formal Sup. Thinking .07 -.10 .12 .10 .24* .08 -.02 -.23* .04 .19** .06 .22** .05 -.07 .06
Naïve-Optimism .10 .04 .13 .26** .28* .26** -.02 -.21 .06 .23** .33** .17* .13* .17 .11
Over-Optimism .07 .15 .04 .24** .25* .23** .11 .00 .14 .15* .31** .10 .07 .15 .05
Stereotypical Thinking .00 .06 -.02 .19** .14 .23** -.21** -.20 -.21** .20** .28* .15* .10 .25* -.00
Pollyanna-ish Thinking .13* -.09 .22** .16** .25* .12 -.01 -.29** .10 .16** .22 .12 .13* .04 .16*
Notes: *Indicates p<.05; **Indicates p<.01; ˄Relationships indicated by Spearman’s rho; Group (G) n = 273 Male (M) n = 76; Female (F) n = 197; Significant differences in
male and female correlations are bolded.
369
APPENDIX F
Exploratory Analyses of Relationships between Leader Naïve-Optimism Sub-
scales and Follower LMX
The following tables provide the exploratory multi-level analyses which were
conducted in study 3, investigating the relationships between leaders’ over-
optimism, stereotypical thinking, and pollyann-ish thinking (subscales of Naïve-
optimism) with followers’ overall LMX. Exploratory analyses are also presented
investigating how the follower IILT dimensions of attractiveness and dedication may
moderate these relationships.
Table 8.1 presents the exploratory analyses investigating the relationship
between leader over-optimism and follower LMX. As can be seen, in step one, a
negative relationship between leader over-optimism and follower LMX was found
but was not significant (p = .105). Step two showed that follower attractiveness IILT
was not significantly related to their LMX rating. However, step three demonstrates
a significant interaction between leader over-optimism and follower attractiveness.
To investigate this interaction, simple slopes analysis was undertaken and is
presented in Figure 8.1. It was observed that the slope at 1SD above the mean of the
moderator was significant (t = 3.800, p < .001). The slope at 1SD below the mean
though was not significant (t = .101, p = .920).
370
Table 8.3.
Exploratory hierarchical linear modelling results predicting follower LMX with
Leader over-optimism and Follower Attractiveness as Moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Over-optimism (A) -.040 -.040 .064^
Moderator
Attractiveness (B) -.009 .796**
Interaction Term
A×B -.020**
Level 1 variance of residuals .481 .429 .430 .351
Variance change N/A .052 -.001 .079
ΔR2 N/A .108 -.002 .184
^p<.10; *p<.05; **p<.01
Figure 8.1. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader over-optimism (1±SD) and follower LMX.
Table 8.2 presents the exploratory analyses investigating the moderating
effect of follower dedication on the relationship between leader over-optimism and
follower LMX. Step one was the same as previously found above. Step two showed
that follower dedication IILT was not significantly related to their LMX rating.
However, Step three demonstrates a significant interaction between leader over-
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Over-Optimism
Low Attractiveness High Attractiveness
371
optimism and follower attractiveness. To investigate this interaction, simple slopes
analysis was undertaken and is presented in Figure 8.2. It was observed that the slope
at 1SD above the mean of the moderator was significant (t = 2.927, p = .004). The
slope at 1SD below the mean though was not significant (t = .494, p = .622).
Table 8.4.
Exploratory hierarchical linear modelling results predicting follower LMX with
Leader over-optimism and Follower Dedication as Moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Over-optimism (A) -.040 -.038 .016^
Moderator
Dedication (B) .135 1.12**
Interaction Term
A×B -.024**
Level 1 variance of residuals .481 .429 .400 .372
Variance change N/A .052 .029 .028
ΔR2 N/A .108 .068 .070
^p<.10; *p<.05; **p<.01
Figure 8.2. Moderating effect of follower dedication IILT (1±SD) on the relationship
between leader over-optimism (1±SD) and follower LMX.
1
2
3
4
5
6
7
1 2
Foll
ow
er L
MX
Leader Over-Optimism
Low Dedication High Dedication
372
Table 8.3 presents the exploratory analyses investigating the relationship
between leader stereotypical thinking and follower LMX. As can be seen, in step
one, a significant positive relationship between leader stereotypical thinking and
follower LMX was found. Step two showed that follower attractiveness IILT was not
significantly related to their LMX rating. However, step three demonstrates a
significant interaction between leader stereotypical thinking and follower
attractiveness. To investigate this interaction, simple slopes analysis was undertaken
and is presented in Figure 8.3. It was observed that the slope at 1SD above the mean
of the moderator was significant (t = 5.513, p < .001). The slope at 1SD below the
mean though was not significant (t = 1.199, p = .234).
Table 8.5.
Exploratory hierarchical linear modelling results predicting follower LMX with
Leader stereotypical thinking and Follower Attractiveness as Moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Stereotypical thinking (A) .037** .037** -.023
Moderator
Attractiveness (B) -.010 -.423**
Interaction Term
A×B .011**
Level 1 variance of residuals .481 .448 .448 .360
Variance change N/A .033 .000 .088
ΔR2 N/A .069 .000 .196
^p<.10; *p<.05; **p<.01
373
Figure 8.3. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader stereotypical thinking (1±SD) and follower LMX.
Table 8.4 presents the exploratory analyses investigating the moderating
effect of follower dedication on the relationship between leader stereotypical
thinking and follower LMX. Step one was the same as previously found above. Step
two showed that follower dedication IILT was not significantly related to their LMX
rating. However, step three demonstrates a significant interaction between leader
stereotypical thinking and follower attractiveness. To investigate this interaction,
simple slopes analysis was undertaken and is presented in Figure 8.4. It was
observed that the slope at 1SD above the mean of the moderator was significant (t =
6.072, p < .001). The slope at 1SD below the mean though was not significant (t =
.617, p = .539).
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Stereotypical Thinking
Low Attractiveness High Attractiveness
374
Table 8.6.
Exploratory hierarchical linear modelling results predicting follower LMX with
Leader stereotypical thinking and Follower Dedication as Moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Stereotypical thinking (A) .037** .037** -.132*
Moderator
Dedication (B) .136 -.631*
Interaction Term
A×B .020**
Level 1 variance of residuals .481 .448 .416 .384
Variance change N/A .033 .032 .032
ΔR2 N/A .069 .071 .077
^p<.10; *p<.05; **p<.01
Figure 8.4. Moderating effect of follower dedication IILT (1±SD) on the relationship
between leader stereotypical thinking (1±SD) and follower LMX.
Table 8.5 presents the exploratory analyses investigating the relationship
between leader pollyann-ish thinking and follower LMX. As can be seen, in step
one, a positive relationship between leader pollyann-ish thinking and follower LMX
was was found but was not significant. Step two showed that follower attractiveness
IILT was not significantly related to their LMX rating. However, step three
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Stereotypical Thinking
Low Dedication High Dedication
375
demonstrates a significant interaction between leader pollyann-ish thinking thinking
and follower attractiveness. To investigate this interaction, simple slopes analysis
was undertaken and is presented in Figure 8.5. It was observed that the slope at 1SD
above the mean of the moderator was significant (t = 4.810, p < .001). The slope at
1SD below the mean though only approached significance (t = 1.806, p = .074).
Table 8.7.
Exploratory hierarchical linear modelling results predicting follower LMX with
Leader pollyann-ish thinking and Follower Attractiveness as Moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Pollyann-ish thinking (A) .020 .020 -.081*
Moderator
Attractiveness (B) .000 -.851**
Interaction Term
A×B .016**
Level 1 variance of residuals .481 .432 .432 .370
Variance change N/A .049 .000 .062
ΔR2 N/A .102 .000 .144
^p<.10; *p<.05; **p<.01
Figure 8.5. Moderating effect of follower attractiveness IILT (1±SD) on the
relationship between leader pollyann-ish thinking (1±SD) and follower LMX.
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Pollyann-ish Thinking
Low Attractiveness High Attractiveness
376
Table 8.6 presents the exploratory analyses investigating the moderating
effect of follower dedication on the relationship between leader pollyann-ish
thinking and follower LMX. Step one was the same as previously found above. Step
two showed that follower dedication IILT was not significantly related to their LMX
rating. However, step three demonstrates a significant interaction between leader
pollyann-ish thinking thinking and follower dedicatopm. To investigate this
interaction, simple slopes analysis was undertaken and is presented in Figure 8.6. It
was observed that the slope at 1SD above the mean of the moderator was significant
(t = 2.997, p = .003). The slope at 1SD below the mean though was not significant (t
= .814, p = .418).
377
Table 8.8.
Exploratory hierarchical linear modelling results predicting follower LMX with
Leader pollyann-ish thinking and Follower Dedication as Moderator Baseline Step 1 Step 2 Step 3
Independent Variable
Pollyann-ish thinking (A) .020 .018 -.078*
Moderator
Dedication (B) .137 -.467^
Interaction Term
A×B .012**
Level 1 variance of residuals .481 .432 .403 .387
Variance change N/A .049 .029 .016
ΔR2 N/A .102 .067 .040
^p<.10; *p<.05; **p<.01
Figure 8.6. Moderating effect of follower dedication IILT (1±SD) on the relationship
between leader pollyann-ish thinking (1±SD) and follower LMX.
1
2
3
4
5
6
7
Low High
Foll
ow
er L
MX
Leader Pollyann-ish Thinking
Low Dedication High Dedication