The Effect of Personality and Emotional Intelligence on Workplace Performance: An Investigation of Hong Kong managers by Reuben Darrell Shaffer Presented to the Faculty of the International Graduate School of Management The University of South Australia in Partial Fulfillment Of the Requirements For the Degree of DOCTOR OF PHILOSOPHY The University of South Australia at Adelaide, Australia October 2004
156
Embed
The Effect of Personality and Emotional Intelligence on Workplace Performance
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The Effect of Personality and Emotional Intelligence on Workplace Performance: An
Investigation of Hong Kong managers
by
Reuben Darrell Shaffer
Presented to the Faculty of the International Graduate School of Management
The University of South Australia in Partial Fulfillment
Of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
The University of South Australia at Adelaide, Australia
October 2004
TABLE OF CONTENTS
Page
TABLE OF CONTENTS i
LIST OF FIGURES iii
LIST OF TABLES iv
ABSTRACT v
DECLARATION vii
ACKNOWLEDGEMENT viii
Chapter
INTRODUCTION 1
Statement of the Problem 1
What is Emotional Intelligence 8
Key Research Issues Involving El 9
Research Purpose and Objectives 10
Overview of this Study 11
LITERATURE REVIEW 13
Emotional Intelligence: An Historical Perspective 13
Empirical Studies 16
Conceptualizations of Emotional Intelligence 17
Summary 43
THEORETICAL FRAMEWORK 45
Socioanalytic Theory 45
A Model of Personality, El, and Workplace Performance 48
Moderating Effects of Emotional Intelligence 69
1
METHOD 72
Research Strategy and Design 72
Data Collection 74
DATA ANALYSES AND RESULTS 88
Analytical Procedures 88
Data Quality 89
Descriptive Statistics 95
Hypothesis Tests 96
DISCUSSION 106
Key Research Findings 106
Limitations of the Research 118
Strengths of the Research 120
Contributions to the Literature 121
Implications of the Results for Organizations 122
Suggestions for Future Research 126
Conclusion 128
REFERENCES 130
11
LIST OF FIGURES
Figure Page
3.1 Theoretical Framework 49
5.1 Interaction Effects of Emotional Intelligence and Agreeableness on TaskPerformance 102
5.2 Interaction Effects of Emotional Intelligence and Agreeableness onContextual Performance 102
5.3 Interaction Effects of Emotional Intelligence and Agreeableness onInnovative Performance 103
5.4 Interaction Effects of Emotional Intelligence and Intellectance onInnovative Performance 103
5.5 Interaction Effects of Emotional Intelligence and Agreeableness onOverall Performance 104
5.6 Interaction Effects of Emotional Intelligence and Conscientiousness onRelationship Disruptive Behaviors 104
5.7 Interaction Effects of Emotional Intelligence and Emotional Stability onRelationship Disruptive Behaviors 105
111
LIST OF TABLES
Table Page
1.1 Borman and Motowidlo's (1997) Taxonomy of ContextualPerformance 4
1.2 Validity Coefficients for Commonly Used Selection Tests 6
2.1 Comparison of Three El Models 18
2.2 Comparison of Major El Measures 25
2.3 The Four-Branch Model of Emotional Intelligence 28
3.1 The Big Five Personality Dimensions 54
4.1 Demographic Profile of Respondents 79
4.2 Items to Assess Performance and Personality 81
4.3 Sample Items from the MSCEIT 84
5.1 Factor Loadings of Independent Variables 92
5.2 Factor Loadings of Dependent Variables 93
5.3 Descriptive Statistics, Correlations and Cronbach's Alphas 98
5.4 Regression Results for the Direct Effects of Personality andEl Abilities on Performance 100
5.5 Regression Results for the Moderator Effects of Emotional Intelligence 101
iv
ABSTRACT
The extant literature on emotional intelligence (El) is replete with claims that
El is an important antecedent (higher than IQ in many cases) of job performance and
success (e.g., Goleman, 1995). Additionally, the El literature continues to debate its
scope and relationship with personality factors (e.g., McCrae, 2000). To clarify these
major issues, I drew upon socioanalytic theory (Hogan & Shelton, 1998a) to develop
a model predicting the direct effects of both the Big Five personality traits and El on
multiple forms of performance (i.e., task, contextual, and innovative performance as
well as relationship supportive and disruptive behaviors) and the moderating effects
of El (conceptualized as a social skill) on the relationship between the Big Five and
performance.
The proposed model was tested with two on-line instruments completed by
116 Hong Kong managers. One instrument was an abilities test of emotional
intelligence (MSCEIT 2.0: Mayer, Salovey, & Caruso, 2002). The other was a survey
of self-reported personality and workplace behavioral data. Multiple hierarchical
(moderator) regression was used to analyze the data.
Hypotheses in the proposed model were generally supported. Extraversion was
a significant positive predictor of relationship supportive behaviors. Agreeableness
had a positive influence on contextual performance and relationship supportive
behaviors; it was a negative predictor of relationship disruptive behaviors.
Conscientiousness was positively associated with task performance, and emotional
stability was a negative influence on relationship disruptive behaviors. Except for
Branch Three (understanding emotions) of El, all branches were significant in
predicting various forms of performance. Branch One (perceiving emotions) had a
negative effect on contextual and relationship disruptive behaviors. Branch Two
(facilitating emotions) was a negative predictor of relationship disruptive behaviors.
Branch Four (managing emotions) had a negative influence on relationship supportive
behaviors. The effects of El on contextual performance and relationship supportive
behaviors were in the opposite direction hypothesized.
Several interactions of El and personality were significant in predicting all
except relationship supportive behaviors. Agreeableness was involved in three
influential interactions with El; for those with high El scores, relationships between
agreeableness and task, contextual, and innovative performance were enhanced.
Intellectance interacted with El to predict innovative performance; in this case, El had
a suppressive effect. For relationship disruptive behaviors, El interacted with both
conscientiousness and emotional stability to counteract the negative effects of those
personality traits.
This study has made several important contributions to the literature. First, it
has clarified the (joint) roles of El and personality on performance. Second, it has
expanded the performance criterion space beyond the traditional focus on task and
contextual performance by including measures of innovative performanceas well as
relationship supportive and disruptive behaviors. Third, it has provided an explicit
test of socioanalytic theory by conceptualizing El as a social skill that interacts with
personality to predict performance. Finally, these findings have significant practical
value to the selection and assessment of managers.
vi
DECLARATION
I declare that this thesis presents work carried out by myself and does not
incorporate without acknowledgement any material previously submitted for a degree
or diploma in any university; and that to the best of my knowledge it does not contain
any materials previously published or written by another person except where due
reference is made in the text.
vii
IcZeReuben Da 1 S '.er
1 October 2004Date
ACKNOWLEDGEMENT
The PhD process can be a lonely, solitary march or, as in my case, a social and
emotional climb. The experience of developing this research made me recognize far
more than in any other former endeavor, the value of teamwork. I am indebted to
many individuals for their support, both directly and indirectly, in making this project
a reality.
First, I would like to thank my two formal advisors. I am grateful to Dr.
Barry Elsey for his patience, understanding and support as I grappled with several
peripheral topics and areas of study before discovering the central themes of the work.
Secondly, I am indebted to Dr. David Harrison for his readiness to act as a principle
supervisor from thousands of miles away. He has reaffirmed that when it comes to
the discourse of ideas, distance and time can be overcome and even leveraged.
I am grateful to many support individuals, in particular, Ms Rebecca Lui for
her assistance in navigating the administrative waters of the project. Her dedication to
service helped to make the process a great deal easier. The help of Mr. W.M. Fu in
web design and administration was invaluable in the data collection as was Ms
Shirley Liu in assisting with the data input. The solid introduction to El testing
provided by Dr. David Caruso and Mr. Charles Wolfe is sincerely appreciated.
Likewise, the able assistance in El testing provided by the support staff of Multiple
Health Systems is acknowledged.
Of course, the research could not have been completed without the
cooperation of the many Hong Kong managers who volunteered to be respondents.
The very act of taking several hours from their busy schedules for the sake of
academic enquiry and to gain greater self-awareness demonstrates that they are the
type of people that today's learning organizations require.
viii
I am indebted to a host of individuals I have traveled with over the last fifty
years that confirmed for me the power that emotions and personality have in
influencing many of the social outcomes of life. While these individuals are too many
to enumerate here, one individual stands highest in this group, Dr. Margaret Shaffer.
After thirty years of marriage and the sharing of life's full range of emotional
peaks and valleys, I think she is the best-demonstrated practice of the power of
emotional management and how it relates to the myriad of life outcomes. As my
"life-partner" she has fulfilled the role of "supporting spouse" far beyond what I
probably deserved or hoped for. However, beyond this conventional role, she served
as an on-going mentor and advisor in this research, sharing her considerable
experience and skills. She has shown what can be done when the right elements of
personality, emotional intelligence abilities and love (of me and academic inquiry) are
focused. Ten years ago she dedicated her PhD thesis to me and I am delighted to
finally reciprocate.
For what is a scholarly and meaningful contribution to the field I am very
indebted to all of the above. For the many errors and omissions, I take personal
responsibility and ask the reader to charge it to my learning account.
ix
CHAPTER 1
INTRODUCTION
Statement of the Problem
A central philosophical and religious question that has historically divided
groups of people is: What is the better part of the human self, its head or its heart?
(Smith, 1991) The question is age-old and relates to the entire range of human
enterprises. However, it is only recently that management researchers have seriously
considered the question in the context of the workplace.
This interest is peaked by a desire of most organizations to improve employee
performance. The research community reflects this search for efficiency by the
attention it has given job performance in recent years: Bommer, Johnson, Rich,
Podsakoff & MacKenzie (1995) observed that job performance is the most
extensively researched criterion variable in both the organizational behavior and the
human resource management literatures. Driving this interest is a need for
organizations to gain an advantage in an increasingly global and competitive economy
(Welbourne, Johnson, & Erez, 1998). According to Lawler (1986), such widely used
initiatives as total quality management, employee involvement, job enrichment, skill-
based pay, autonomous work teams, and gain sharing, have a common goal of
influencing employees' work behavior, level of responsibility acceptance, and
participation in achieving group-based and organizational objectives.
Underlying the challenge of managing performance are changes in the nature
and structure of work. For example, with the shift to a service economy,
organizations have increasingly focused on relationship marketing concepts and
interpersonal skills as ways to enhance service quality (Kotler & Armstrong, 2001).
This new perspective of the relationship that service providers have with their clients
1
has required that many organizations change how work is structured. In particular,
more work is now being accomplished through teams. As organizations have seen the
benefits of diverse perspectives, skills, and knowledge conveyed through enhanced
innovation and improved decision-making, they have increasingly turned to
organizing in teams (Lawler, 1998). The extent of this shift to a team orientation was
reported by Lawler, Mohrman, and Ledford (1995) where 79 percent of Fortune 1000
firms reported that they used self-managed work teams, while 91 percent reported the
use of employee work groups. This emphasis on teams, as a way to organize work,
has motivated researchers to become more interested in investigating team processes
such as cooperation and cohesion, as well as team results (Druskat & Wolff, 2001).
With more cross-functional teams and team based knowledge workers making more
complex, important decisions, it becomes more important to learn how to better
facilitate group effectiveness.
Concurrent with changes in the nature and structure of work is the recognition
that performance is more than the execution and completion of well-defined tasks
(Bommer et al., 1995; Borman & Motowidlo, 1993). During the last twenty-five
years, still other streams of research have emerged that move the focus beyond task
performance to consider other forms of employee performance. These include
Each of these lines of inquiry have made contributions in raising awareness about
what contributes to overall performance in the workplace.
Delineating the distinction between task and contextual performance more
clearly, Van Scotter, Motowidlo and Cross (2000) characterize task performance as
2
patterns of behavior employees exhibit in the production of goods and delivery of
services or more specifically, those activities that contribute indirectly to supporting
the organization's core technical processes. They contend that a more complete
picture of employee performance is gained with the inclusion of contextual
performance behaviors, which are defined as patterns of behavior that support the
social and psychological context in which the tasks are performed. In developing their
taxonomy of contextual performance (Table 1.1) they note that they drew heavily on
work previously done in the areas of organizational citizenship behavior (OCB:
Organ, 1988) and prosocial organizational behavior (POB: Brief & Motowidlo,
1986). Borman and Motowidlo (1997) advanced the argument that contextual
performance and task performance are different in three ways. Whereas task activities
are variable from job to job, contextual activities are seen to be more stable across
jobs. Further, task behaviors, more so than contextual behaviors, tend to be role
prescribed. Finally, task performance is related more directly to cognitive abilities
whereas contextual performance is related more to personality variables. They further
contend that personnel selection can be made more successful by including contextual
performance and, thereby, personality variables.
Other researchers have recently drawn further attention to the practicality of
this extended definition of work performance. For example, Ashforth and Humphrey
(1993) focused on the importance of expressing desirable emotions in the context of
service transactions, referring to such acts as "employee affective delivery" (EAD).
Brown and Sulzer-Azaroff (1994) reported that friendlier service results in higher
levels of customer satisfaction. Pugh (2001) demonstrated a link between EAD and
ratings of service quality and Tsai (2001) showed that displayed emotions result in an
increased willingness for customers to return and to refer others. However, while
3
evidence continues to build as to its importance, the job analysis function typically
does not take non-job-related behavior into account. This leads to problems for
organizations wishing to reward behaviors (such as extraordinary customer service)
that do not fit the traditional task performance definition (We'bourne et al., 1998).
Table 1.1: Borman and Motowidlo's (1997) Taxonomy of Contextual Performance
Persisting with enthusiasm and extra effort as necessary to complete own taskactivities successfully.
Perseverance and conscientiousness (Borman et. al., 1985)Extra effort on the job (Brief & Motowidlo, 1986; Katz & Kahm, 1978)
Volunteering to carry out task activities that are not formally part of own job.Suggesting organizational improvements (Brief & Motowidlo, 1986; Katz &Kahn, 1978)Initiative and taking on extra responsibility (Borman et. al., 1985; Brief &Motowidlo, 1986; Katz & Kahn, 1978)Making constructive suggestions (George & Brief, 1992)Developing oneself (George & Brief, 1992)
Helping and cooperating with others.Assisting/helping coworkers (Borman et al., 1985; Brief & Motowidlo, 1986;Katz & Kahn, 1978)Assisting/helping customers (Brief & Motowidlo, 1986)Organizational courtesy (Organ, 1988)Sportsmanship (Organ, 1988)Altruism (Smith et al., 1983)Helping coworkers (George & Brief, 1992)
Following organizational rules and procedures.Following orders and regulations and respect for authority (Borman et al.,1985)Complying with organizational values and policies (Brief & Motowidlo, 1986)Conscientiousness (Smith et al., 1983)Meeting deadlines (Katz & Kahn, 1978)Civic virtue (Graham, 1986)
Endorsing, supporting, and defending organizational objectivesOrganizational loyalty (Graham, 1986)Concern for unit objectives (Borman et al., 1985)Staying with the organization during hard times and representing theorganization favorably to outsiders (Brief & Motowidlo, 1986)Protecting the organization (George & Brief, 1992)
Source: Borman and Motowidlo, 1997, p.102
A central issue in expanding the definition of employee performance is the
need to clearly understand its antecedents. The last century saw a great body of
research that demonstrated antecedents of individual task performance. As a result,
4
much more is known about what makes for high levels of workplace task performance
than is known about antecedents of contextual performance. For example, a common
measure considered is an individual's general intelligence. This is referred to in the
literature as General Mental Ability (GMA) or (g) and its roots go back a century ago
when Spearman (1904) first introduced the term. Hernstein and Murrray (1994)
reflected how widely accepted the construct of a general intelligence has become and
how it is commonly measured by stating:
Among the experts, it is now beyond much technical dispute that there is sucha thing as a general factor of cognitive ability on which human beings differand that this general factor is measured reasonably well by a variety ofstandardized tests, best of all by IQ tests designed for that purpose. (p.35)
Others agree that IQ testing, as a measure of individual differences, has proven
useful in predicting performance across a number of domains (Schmidt & Hunter,
1998). Gottfredson (1998) contends that, "Intelligence as measured by IQ tests is the
single most effective predictor known of individual performance at school and on the
job" (p. 24). One possible reason for its effectiveness in predicting performance is
that general intelligence can reflect an individual's ability to learn new cognitive
tasks.
In reviewing tests commonly used for selection, Howard and Howard (2001)
listed the wide range of instruments firms have traditionally used to predict
performance along with the average predictive validity of each (see Table 1.2). It is
interesting to note that the highest average predictive validity of all the measures
reported was mental and psychomotor tests at .53. Other measures of note are job
knowledge tests (.50), skill tests (.44), Big Five tests with job analysis (.44),
biographical information forms (.35), and structured interviews (.34). Other measures
commonly used with lower average predictive validities include education, reference
checks and age.
5
However, noticeably absent in this litany of measures are instruments that
attempt to directly assess areas of contextual performance potential. Indeed, there is
no evidence that general intelligence or other psychometric tests are predictive of
contextual performance. That is, an employee may possess a very high GMA and be
very capable of performing required "tasks" but still unable to deliver high "affective"
service.
Table 1.2: Validity Coefficients for Commonly Used Selection Tests
Selection Test or Procedure Average Predictive ValidityMental and psychomotor tests .53Job knowledge tests .50Skill tests .44Big Five test with job analysis .40Biographical information forms .35Structured interviews .35Assessment centers .25Personality tests (pre-Big Five) .24Class rank .21Experience .18Traditional interviews .17Reference checks .13College grades .13Vocational interest tests .10Amount of education .10Handwriting analysis .00Projective personality tests .00Age .00Source: Howard and Howard, 2001, p. 177
Recent studies (Bommer et al., 1995; Motowidlo & Van Scotter, 1994; Van
Scotter & Motowidlo, 1996) have indicated that contextual performance accounts for
substantial variance in supervisory performance ratings. However the influence it
may exert on other outcomes important for employees has not been tested. "Little is
known about the extent to which contextual performance influences employees' job-
related rewards and career advancement over time." (Van Scotter et al., 2000, p. 526).
Research in areas such as personality has proven helpful in identifying
individual differences that offer the hope of better predicting workplace performance
6
in both task and contextual areas (Howard & Howard, 2001). Additionally, and
representative of the diverse range of interest in uncovering antecedents to
performance, investigations in the area of GMA are continuing. This has at times
become very controversial as witnessed by the recent publication of the Bell Curve
argument for the importance of social class and race as a significant determinant
(Herrnstein & Murray, 1994). It is evident that much more needs to be done to
delineate the complexity of performance antecedents. Thus, two research questions
drive this current research:
What individual differences or traits might predict multiple forms of
performance beyond the current established measures?
Do measures of those individual differences or traits exist that have
good psychometric properties and utility in HR selection?
One possible stream of research that has emerged in recent years that has been
associated with the study of effective performance is Emotional Intelligence (El: Bar-
On, 1997b; Goleman, 1995; Salovey & Mayer, 1990). Insofar as the management of
social behavior involves the management of emotions (Hochschild, 1983), Emotional
Intelligence has the potential to be a strong predictor of more contextual and
interpersonal behaviors. As Goleman (1995) notes: " . . . imagine the consequences
for a working group when someone is unable to keep from exploding in anger or has
no sensitivity about what the people around him are feeling. ...When emotionally
upset, people cannot remember, attend, learn, or make decisions clearly" (p.170).
Linking El with the appropriate criterion (e.g., contextual performance) may help to
clarify a controversy in respect to the relative contributions of personality and El to
employee performance and provide organizations with a valid alternative for selecting
and assessing employees.
7
What is Emotional Intelligence?
Typical of the early stages of research with a new construct, there is a lack of
consensus about what constitutes El. Although most constitutive definitions
appearing in the literature share some common themes, most fail to distinguish
between the construct and its consequences adequately. The first investigators to use
the term in the literature, Salovey and Mayer (1990), offered a definition that most
other theoretical researchers accept (and expand on): "Emotional intelligence is the
ability to perceive emotions, to access and generate emotions...to assist thought, to
understand emotions and emotional knowledge, and to reflectively regulate emotions
so as to promote emotional and intellectual growth"(p.186).
Remarking on the complexity and the many definitions that have emerged for
the term, Mayer, Salovey, and Caruso (2000b) examined how emotional intelligence
has come to be used in the last decade. They identify three popular meanings of the
term: zeitgeist, a group of personality traits, and a set of abilities.
Viewed as a zeitgeist, El reflects the tension between emotion and reason in
Western thought. As one of many cultural trends, it reflects a greater recognition of
the importance of emotions both culturally and politically. Mayer et al. (2000b)
acknowledged that it may be a passing fad but also suggested that it could be an
historical movement of a similar import as the historical stoic, classical, and romantic
movements.
These original theorists have been particularly uncomfortable in characterizing
El as personality, pointing out that much of what appears in the El literature does not
belong and should remain the province of personality psychology. By re-labeling
areas of personality as "emotional intelligence" the definition of El becomes confused
8
and undermines long-term research efforts in both emotional intelligence and
personality.
Lastly, the authors have argued that emotional intelligence should be viewed
as a set of abilities that are part of an individual's intelligence system. This system is
characterized in terms of its capacity to identify and process information with both
immediate symbol manipulation and reference to expert knowledge. With this
conceptualization, emotional intelligence can be seen to operate across both the
cognitive and emotional systems (Mayer et al., 2000b). Mayer and his colleagues
also contend that by conceptualizing it as a set of abilities, El meets the requirements
of a standard dimension of intelligence. In other words, emotional intelligence is not
to be viewed as the opposite of cognitive ability "heart versus head". Rather, it is the
combination of cognition and affect; it is cognitive processing of affective
information. This perspective is embodied in their formal definition of the term:
"the ability to perceive accurately, appraise, and express emotions; the ability
to access and/or generate feelings when they facilitate thought; the ability to
understand emotion and emotional knowledge; and the ability to regulate
emotions to promote emotional and intellectual growth" (Mayer & Salovey,
1997, p. 10)
Key Research Issues Involving El
The lack of consensus regarding the definition of El has generated several
research issues that need to be resolved. One issue has to do with the
operationalization of El. Three major theoretical approaches to El have emerged in
recent years and each has proffered unique instrumentation (Gowing, 2001). Using
cognitive psychology as a theoretical base, Mayer and Salovey (1997) developed an
assessment instrument called the Multifactor Emotional Intelligence Scale (MEIS).
9
The original work of Goleman (1995) started with a theory of work
performance in the world of work and resulted in the Emotional Competence
Inventory (Ed). The work of Bar-On (1997a), which appears to have been built on
both personality and performance in the workplace, attempts to measure emotional
and social intelligence with the Bar-On EQ-i (Emotional Quotient Inventory). The
introduction of these various instruments has resulted in an emphasis in the literature
on construct validation, thus obfuscating our understanding of El in relation to other
constructs.
A second issue has to do with the predictive utility of El. Over the last ten
years many claims for the predictive power of Emotional Intelligence have been made
in the popular press. Speculations have been made that El is twice as important as IQ
to career success (Gibbs & Epperson, 1995; Goleman, 1995). However little rigorous
empirical evidence has been offered to support such claims. There is also limited
research investigating the influence of El on performance (Lam & Kirby, 2002).
Furthermore, making comparisons or comparing findings is hampered by differences
in operationalizations of both El and performance across studies.
A third issue has to do with the distinction between El and personality. Some
researchers seem to use El and personality as interchangeable terms (see McCrae,
2000). However, Mayer, Salovey, and Caruso (2000c) contend that El refers to a set
of mental abilities rather than stable traits such as personality. According to them,
mental abilities represent characteristics of individuals who can successfully perform
to a desired standard.
Research Purpose and Objectives
Many organizational researchers have recently called for more focus on the
role of emotions at work. For example, Ashforth and Humphrey (1995) argued that
10
emotions are an integral and inseparable part of organizational life and more attention
should be given to the employee's emotional experience and the relationship between
rationality and emotionality in the organizational context. The interplay of emotional
and rational forces in the workplace is manifested in various organizational outcomes
such as performance.
In this study, I move beyond what has thus far been largely a narrow focus on
construct validation to examine the relationships among El, personality and employee
performance. Specifically, my objectives are to:
Consider the direct effects of El and the Big Five personality traits
on several performance constructs, including task, contextual, and
innovative performance, as well as relationship supportive (RSB)
and relationship disruptive (RDB) behaviors,
Examine how El interacts with the Big Five personality factors to
influence the various domains of employee performance, and
Discuss the relevance of these results for improving the selection
and assessment of employees.
The target population for this study was Hong Kong managers at various
management levels from a diverse range of industries. This group was chosen
because of the inherent social nature of being a manager. Also, by surveying
managers from a wide range of industries, generalizability of the findings will be
enhanced.
Overview of this Study
This investigation is based on an integration of the job performance literature
emotional intelligence literature, and the personality literature. In Chapter 2, I
provide an in-depth review of the El literature, noting in particular emotional
11
intelligence research involving personality and job performance. Building on this
literature and socioanalytic theory (Hogan & Roberts, 2000; Hogan & Shelton,
1998a), I develop a model and testable hypotheses of the direct effects of personality
and El on employee performance and the moderating effects of El abilities on
relationships between personality and performance. This model is presented in
Chapter 3. Chapter 4 provides details about how I collected the data and the measures
used. In Chapter 5, I present analytical procedures and the results of the study. In
Chapter 6, I conclude with a discussion of the results and a description of the
contributions, limitations and suggestions for future research directions.
12
CHAPTER TWO
LITERATURE REVIEW
In this chapter I review both the theoretical and empirical literatures of
Emotional Intelligence (El). The review reflects the field's early stage of conceptual
development and the confusion that exists due to the lack of a common understanding
of the term Emotional Intelligence, including how encompassing it should be.
Therefore, I first provide constitutive definitions of the construct from an historical
perspective. Then I discuss key variables proposed to be associated with emotional
intelligence. Finally, I specify various limitations and gaps in the extant literature.
Emotional Intelligence: An Historical Perspective
Contemporary El theorists owe a great deal to early thinkers in this area. For
example, over eighty years ago Thomdike (1920a) introduced the term social
intelligences to explain individual outcome measures beyond what may have been
accounted for by IQ. According to Thomdike, social intelligences are aspects of a
person's general intelligence; they reflect an ability to relate well with others. Despite
his belief of the importance of this ability, Thomdike was unable to develop
satisfactory laboratory tests to measure it. He concluded that it could not be
accurately measured other than in an individual's "real world" interactions. He found
little acceptance of this broader view of intelligence from other theorists and
researchers in the field of intelligence at the time; thus it was not pursued as a serious
stream of scientific inquiry.
The recognition that GMA measures (including IQ tests) do not account for
enough of the variance in individual performance continued to frustrate intelligence
theorists for decades. For example, Wechsler (1940) commented that "individuals
with identical IQs may differ very markedly in regard to their effective ability to cope
13
with the environment" (p.444). Indeed, most attempts to measure the phenomena
were failures, perhaps because of the complexity of the abilities, social habits, and
attitudes it involves (see Cherniss & Goleman, 2001 for reviews of the history of
social intelligence research; Goleman, 1995).
In the 1940s, some researchers began to consider the importance of this type
of intelligence in management studies. For example, in 1945, leadership studies done
at Ohio State University identified two dimensions of leadership: structure and
consideration. In conceptualizing a manager's performance as being composed of
concern for task performance and concern for people, the importance of a wider range
of requisite management abilities and skills beyond those that simply 'get the task
done' was recognized. A manager high in consideration was seen to be sensitive to
people's feelings and effective in establishing trust, respect and rapport with his or her
group (Fleishman & Harris, 1962). It is interesting to note that these are common
elements to several contemporary emotional intelligence theorists' models (e.g.,
Goleman, Boyatzis, & Mckee, 2002b). Additionally, recent research in the area of
leadership has demonstrated a link between El and the performance of leaders (Wong
& Law, 2002).
Further evidence of some early acceptance of an extended view of
intelligence is reported by Gowing (2001). In the late 1940s, the U.S Office of
Strategic Services developed a whole person assessment process, which included a
measurement of both cognitive and noncognitive abilities based on the earlier work of
Murray (1938). This was the beginning of what we now know as assessment centers,
used widely by government and business organizations today.
In the 1980s, renewed attention was given to intelligence measures that strive
to go beyond the widely accepted intelligence quotient (IQ) when Gardner (1983)
14
outlined his framework of multiple intelligences. In exploring his proposed seven
"intelligences", he found no significant relationships with IQ measures. According to
Dulewicz and Higgs (2000), this demonstrated that Gardner's "other" intelligences
were not the same construct as IQ. Two of Gardner's suggested seven types of
intelligence are of particular interest with respect to the construct of emotional
intelligence. Reminiscent of Thorndike's (1920b) idea of social intelligence, Gardner
broke down what he termed personal intelligences into interpersonal and intrapersonal
dimensions. The interpersonal intelligence dimension reflects an individual's ability
to understand others; intrapersonal intelligence reflects an understanding of one's self.
This is also mirrored in a later definition offered by Marlowe (1986). For him, social
intelligence is the "...ability to understand the feelings, thoughts, and behaviors of
persons, including oneself, in interpersonal situations and to act appropriately upon
that understanding" (p.52). However, Gardner (1993) has more recently modified his
position of how personal intelligences relate to emotions. Goleman (1995) reports
that while Gardner (1993) originally viewed emotions as a central component of the
intrapersonal intelligence model, in practice his theory of multiple intelligences
evolved to focus on metacognitions. Such cognitions represent awareness of mental
processes rather than particular emotional abilities.
In the 1990's, immediate interest in emotional intelligence arose after Salovey
and Mayer (1990), building on the work of Gardner and others, introduced the term
emotional intelligence to the research community. According to Mayer, Salovey, and
Caruso (2002), Gardner's definition.. ."described emotional abilities but failed to
admit explicitly the possibility of a separate emotional intelligence" (p.5). To
delineate emotional intelligence as a subset of general intelligence, it is critical to both
describe and measure emotional intelligence as ability. Using the work of Salovey
15
and Mayer (1990) and Gardner (1983) as a foundation, Goleman (1995) wrote the
best-selling book, Emotional Intelligence, creating unprecedented popular interest in
emotional or social intelligence research. As businesspersons and researchers debated
the value of early theories, new and similar views of what the construct should mean
emerged. For example, the concepts of Emotional Literacy (Steiner, 1997);
Emotional Intelligence As an Ability: Mayer and Salovey
Salovey and Mayer (1990) were the first to attempt a formal definition of
emotional intelligence. Their initial work proposed that much of the research in
aesthetics, brain research, intelligence measurement, artificial intelligence, and
clinical psychology shared a common focus on a form of intelligence that had not
been previously examined. They defined emotional intelligence (El) as "...the subset
of social intelligence that involves the ability to monitor one's own and others'
feelings and emotions, to discriminate among them and to use this information to
guide one's thinking and actions" (p.189). This is consistent with Gardner and
Hatch's (1989) concepts of interpersonal and intrapersonal intelligences.
Interpersonal intelligence is the ability to judge and respond properly to the
Method Informant Ability Self-Report
Emotional General MoodManagement
motivations, desires, moods, and temperaments of others; intrapersonal intelligence is
the ability to access one's own feelings, discriminate among them and to use them to
guide one's behavior. Goleman (2001b) characterized Salovey and Mayer's model as
having a distinctively cognitive emphasis that stresses the importance of measuring
the process of "thinking about feeling". This is in contrast to other models, which
only account for perceiving and regulating feelings. In accord with this, Matthews,
Zeidner, and Roberts (2002) characterize this abilities approach as being more
sophisticated theoretically than others and having its roots in cognitive psychology.
As has been noted, there is an ongoing debate in the El literature, emanating
from the lack of common definitions, as to whether or not El can be distinguished
from personality (see McCrae, 2000). Just as emotional intelligence has been
confused with performance, some researchers have also blended emotional
intelligence and personality to form what Mayer and his colleagues (e.g., Mayer et al.,
2000c) refer to as "mixed models". However, from its conception and on through
limited empirical research, the abilities approach has clearly demarcated emotional
intelligence from other well-tested constructs of personality. McCrae (2000)
highlights the importance of "breaking out" individual personality variables from a
global construct of El (counter to the competency frameworks) as being both practical
and scientifically sound. He further observed that the distinction between the El
abilities (as described by Mayer and his colleagues) and personality traits may be
difficult to discern at times but it can be seen. He illustrates this by suggesting that a
person can be an optimist owing to a cheerful disposition and this does not necessitate
intelligence of any kind. However, a person can display a form of El abilities by
deliberate cognitions, i.e., manipulating his or her emotional state to create an
optimistic assessment.
26
Mayer, DiPaolo, and Salovey (1990) operationalized their El construct with
the development of the first abilities scale designed to measure these elements of
emotional intelligence. The scale's objective was to test emotional intelligence in a
similar manner as IQ is measured by the Wechsler Adult Intelligence Scales
(Wechsler, 1958). As a subset of social intelligence, and thereby general intelligence,
Mayer and his colleagues see significant commonalities with IQ. For example, Wolfe
and Caruso (2002) stress that, like IQ, El is fairly static, changing only marginally as
a person matures after his or her formative years. An individual may, however, learn
to compensate for lower "areas" of emotional intelligence by employing strategies
that take advantage of their areas of strength.
The early administration of the scale generated data that allowed Mayer and
his colleagues to better identify the abilities involved with El and to refine their
model. The result was the "Four-Branch Model of Emotional Intelligence" (Mayer &
Salovey, 1997) and a revised definition: "the ability to perceive accurately, appraise,
and express emotion; the ability to access and/or generate feelings when they facilitate
thought; the ability to understand emotion and emotional knowledge; and the ability
to regulate emotions to promote emotional and intellectual growth" (p.10). In
essence, the framework organizes the various abilities involved in the adaptive
processing of emotionally relevant information (Hedlund & Sternberg, 2000). Mayer,
Salovey, and Caruso (2002, p.7) provide a current overview of the Four Branches
Model and the skills involved in each branch (see Table 2.3).
In their review of the empirical El research, Ciarrochi et al. (2001a) compare
and contrast the two main measurement methods developed thus far (self-report
instruments and performance instruments) in terms of their ability to discriminate
between El and personality. They observe that self-report measures of El show
27
Table 2.3: The Four-Branch Model of Emotional Intelligence
Branch Skills
1 Perceiving Emotions
2 Facilitating Thought
3 Understanding Emotions
4 Managing Emotions
28
Perceiving emotions in self, others,objects, art, stories, music, and otherstimuli. With emotions being sensed, anautomatic influence is generated oncognition.To generate, use, and feel emotion tocommunicate feelings, or in othercognitive processes. In short, emotionsand emotion-related information isattended to.Understanding information that emotionsconvey, how they come together andmove through relationship transactions,and understand their meanings. Theimplications of emotions, their feelingsand their meanings are processedBeing open to one's feelings, managingthem in self and others to promotepersonal understanding and growth bothpsychologically and intellectually.Management of emotions promotesfurther openness to feelings
Source: Mayer, Salovey, and Caruso (2002, p. 7)
moderate to strong overlap with temperament. Because two major components of
temperament related to the Big Five personality factors are neuroticism and
extroversion, they suggest that this overlap indicates that self-report measures are
perhaps measuring aspects of personality. However, they also report that ability-
based measures of El, such as the MEIS, do not generally overlap with temperament
but do reflect a small to medium overlap with traditional measures of intelligence.
Indeed, the ability of the MEIS to achieve high levels of distinctiveness is perhaps one
of its greatest strengths (Ciarrochi, Chan, & Caputi, 2000). Further, a recent critique
of the MEIS reported that correlations with broad personality traits generally are less
than .30 and that it correlates positively but modestly as would be expected with
cognitive-ability measures (Matthews et al., 2002). This is consistent with the view
that El is a mental ability and should be set apart from traditional conceptualizations
of personality.
Measures of El as Ability
The Multibranch Emotional Intelligence Scale (MEIS) was developed Mayer,
Salovey, and Caruso (1997) to test their model of El. It and its successor, the
MSCEIT, render a range of scores: Total El, two area scores (experiential and
strategic), a score for each of the four branches, and two subscale scores for each
branch.
The MEIS was normed on a general population sample of 5000 North
American respondents and provided empirical evidence to support their theory and
the capacity of the test to measure emotional intelligence abilities (Mayer et al.,
2002). The researchers also tested the scale with expert opinion, and targeted
approaches to score emotional intelligence. The findings showed that all three
methods were related, with the general population approach being the highest in
predictive value. Mayer et al. (2002) reported that emotional intelligence emerged as
a unified intelligence with three distinct sub factors: emotional perception, emotional
understanding, and emotional management. These correspond with the first, third,
and fourth branches. Limited evidence was found for "Integrating Emotions,"
representing the second branch of the model. The findings also showed that the test
results, while related to both general intelligence and self-reported empathy, were still
fairly independent. This supported the researchers' view that measures of emotional
intelligence as ability can measure qualities not covered by other tests.
Recognizing limitations of length (402 items) and its weakness in measuring
Branch Two subfactors, Mayer et al. (2002) modified the test and renamed it. The
29
current version is the Mayer, Salovey, Caruso Emotional Intelligence Test (MSCEIT).
With 141 items, most respondents can complete it in 30 to 45 minutes.
In reviewing the tests, Gowing (2001) states "They predict, with empirical
justification, that ...internal abilities have external consequences" (p.93). This
appraisal is supported by more recent empirical research. Mayer et al., (1999) report
that the MSCEIT achieved reasonable levels of reliability and a confirmatory factor
analysis supported the theoretical model. After administering the test to 503 adults
and 229 adolescents, they found that performance across 12 diverse tasks was highly
positively correlated across samples. Additionally, a factor analysis of those tasks
gave support for defining them as one general factor as well as indications that they
could each fall into three or four subgroups of skills that generally match the branch
model of El. Finally, evidence was found indicating that El ability increases from
childhood to adulthood.
Basing their work on that of Salovey and Mayer (1990), Schutte et al. (1998)
developed a 33 item, self-report measure that purported to assess El. Significant
correlations were demonstrated with eight theoretically related constructs, which
include awareness of emotion, outlook on life, mood, ability to regulate emotions, and
impulsivity. Other results also add general support for the abilities model: (1) group
differences were demonstrated where they would have been expected, i.e., men and
women, psychotherapists and prisoners; (2) El scores predicted end-of-year GPA for
college freshmen but did not correlate with SAT or ACT scores; (3) with respect to
personality, they found that of the Big Five personality traits, only openness to
experience was related significantly to El (Hedlund & Sternberg, 2000).
To investigate the possible linkages between El, personality and general
cognitive ability, Davies, Stankov, and Roberts (1998) conducted three separate
30
studies. These included both self-report scales and objective tests, which included
recognition of emotions in faces, colors and music. Their findings revealed that verbal
ability and personality factors such as neuroticism, extraversion, and agreeableness
were inclined to load on self-reported measures of El. However, they also found that
the emotional perception factor of El was unrelated to personality or cognitive
abilities. Thus, challenging whether El should be considered in the same manner as
other cognitive abilities, they argue for a more limited definition of El that
encompasses only the ability to perceive emotional information. However, in
commenting on this research, McCrae (2000) notes that this study employed measures
developed before 1990 and that more recent measures (e.g., the MEIS and the
MSCEIT) are more reliable and valid.
Mayer and his colleagues take a measured view as to the potential for El and
the abilities model to predict performance. For example, Mayer (1999) suggests that
El may be able to predict important life outcomes but probably at about the same
level of other personality variables. This may be from 2 to 25 percent of the variance
explained-- much lower than some popular writers have advanced and which have yet
to be substantiated by systematic and rigorous scientific approaches. Mayer sees the
common goal of both the popular and scientific treatments of El to be a broadened
understanding of what intelligence means, and this should be accomplished without
expanded definitions or sensational claims.
In his review of El and its relationship to personality dimensions, McCrae
(2000) observes that the Salovey and Mayer (1990) construct is so appealing that
many researchers have extended it to include motivational, interpersonal, and
intrapsychic attributes that look more like personality traits than abilities. By arguing
for broader notions of "intelligence" and highlighting the adaptive values of flexible
31
planning, social adroitness, and interpersonal considerateness, a theoretical license
was granted to theorists such as Goleman (1995) and others ..."who in effect argued
that any beneficial noncognitive trait might be construed as emotional intelligence"
(p.264). This movement prompted Mayer, Salovey, and Caruso (2000c) to clarify the
construct and to distinguish between ability models and mixed models. McCrae and
John (1992) point out that most of the traits that are included in the research of
emotional intelligence can be found in a comprehensive taxonomy of personality
traits, i.e. the five-factor model (FFM: Digman, 1990). Of the Big Five traits,
openness to experience seems to be most directly related to the original ability
conception of El (McCrae, 2000, p.263).
Continuing his review, McCrae (2000) examined the more recent work of
Mayer, et al., (2000c), and commented that the conceptually related abilities under
their model are distinct (although sometimes in subtle ways) from personality traits.
For example, the personality trait of optimism can be simply the result of a cheerful
disposition, requiring no intelligence. However, individuals with psychological
mindedness (as conceived under the Salovey & Mayer model) can purposefully set
out to change their emotional state by changing their mind set or gaining social
support from others. He concludes by suggesting that while the abilities approach to
evaluating emotional intelligence is promising, much more research is required to
understand it well.
Emotional Intelligence as a Competence
Between 1994 and 1997 the term emotional intelligence was popularized by
Daniel Goleman (1995; 1996; 1997). The foundation for this school of thinking was a
theory of performance in the workplace; the goal is to identify what El
"competencies" lead to high performing employees. However, Gowing (2001), in her
32
review of the conceptual underpinnings of the El domain, cautions that a distinction
must be made between emotional intelligence (Mayer & Salovey, 1997) and
emotional competence (Bar-On, 1997a; Cooper, 1997; Goleman, 1997). On one
hand, emotional intelligence as outlined by Mayer and his colleagues, deals with
abilities to perceive, express, assimilate, understand, reason, and regulate emotions in
self and others. On the other hand, emotional competence for Goleman and his
contemporaries is focused more on outcomes in the workplace.
Goleman's (1995; 1996) first influential books cite many stories and research
studies from the field of education that demonstrate the importance of emotional
intelligence as a competence. Indeed, much of the early work of intelligence testing
has been with children. Contemporary investigations in this sphere support the view
that children generally learn content and how to value their work, social skills of
relating to their peers and how they feel about others (H8pfl & Linstead, 1997).
Dulewicz and Higgs (2000) suggest such studies demonstrate that successful learning
and performance results from both rational and emotional development.
This is consistent with Goleman's (1996) observation that: "IQ and
emotional intelligence are not opposing competencies, but rather separate ones. We
all mix intellect and emotional acuity; people with a high IQ, but low emotional
intelligence (or a low IQ and high emotional intelligence) are, despite the stereotypes,
relatively rare "(p. 44). Thus, he argues that individuals with a mix of high emotional
intelligence and high IQ will generally be more successful in their careers than
individuals with high IQ and low EQ. He further contends that IQ levels tend to be a
threshold to entry for many professions and EQ will then become more salient,
distinguishing those with high levels after entering. For example, a reasonably high
IQ is required to enter the field of law. However, once on the job, advancement and
33
other measures of success are more a matter of emotional intelligence than general
cognitive ability.
Goleman (1998) calls for a "new yardstick" to be used in measuring
employees when making hiring, retention, and promotion decisions. His reasoning
carries a degree of face validity, "...today's workforce is being judge by a new
yardstick.. .we are being judged.. .not just by how smart we are, or by our training and
expertise, but also by how well we handle ourselves and each other" (p.1). The
breadth of his framework is also revealed when he comments: "In a time with no
guarantees of job security, when the very concept of a 'job' is rapidly being replaced
by 'portable skills', these are prime qualities that make and keep us employable.
Talked about loosely for decades under a variety of names, from 'character' and
'personality' to 'soft skills' and 'competence,' there is at last a more precise
understanding of these human talents, and a new name for them: emotional
intelligence" (p.4).
Offering a descriptive definition, Goleman and his colleagues state that El is:
"...observed when a person demonstrates the competencies that constitute self-
awareness, self-management, social awareness, and social skills at appropriate times
and ways in sufficient frequency to be effective in the situation" (Boyatzis et al.,
2000, p.344). Goleman's (1998) emotional intelligence framework suggests that an
individual's potential competencies can be measured across four areas: self-
awareness, self-management, social awareness, and relationship management. After
reviewing nearly three hundred different company-sponsored studies, encompassing a
wide-range of jobs, he concluded that the importance of emotional competencies over
cognitive abilities in making a "star performer" is clear. Admittedly, IQ remains an
important predictor of performance. However, star performers also possess important
34
emotional intelligence competencies that set them apart from others. He argues that
such competencies as persuasion, drive to achieve and inner discipline can play a
critical role in performance outcomes for professionals as diverse as salespersons,
technicians and scientists (Goleman, 1998). Thus, a cornerstone of his model is
emotional competence, which is based on emotional intelligence that results in
outstanding performance at work. By basing his model on competencies, Goleman is
open to critical challenges from those that believe that competencies are difficult to
define and differentiate from more established constructs of ability, skills and
personality (see Matthews et al., 2002, pp. 10-15).
In Goleman's (2001a) most recent model, he identified clusters or behavioral
groups of desired competencies. According to Boyatzis et al. (2000), this is a
practical approach because: "They (clusters) are often linked conceptually and
defined by a 'theory' as a convenient way to describe which competencies are
associated with others. Clustering provides parsimony" (p.349). In her review,
Gowing (2001) offers useful definitions of each of these clusters and their constituent
components. Personal competence, or how we manage ourselves, is made up of the
self-awareness, and self-management clusters. The self-awareness cluster is
composed of three distinct competencies: emotional self-awareness - the ability to
identify one's emotions and their effects; accurate self-assessment having a sense of
ones strengths and limitations; and self-confidence a sense of one's self-worth and
capabilities. The self-management cluster is a set of five competencies: self-control
controlling emotions or impulses that are disruptive; trustworthiness showing
integrity and honesty; conscientiousness displaying responsibility in self-
management; adaptability being able to adapt to the changing environment;
35
achievement orientation drive to meet a personal standard of excellence; and
initiative one's readiness to act.
What Goleman (2001a) terms as social competence is made up of two main
clusters: social awareness and relationship management. The social awareness cluster
includes: empathy understanding others and having a genuine interest in their
concerns; service orientation seeing and meeting the customers' needs; and
organizational awareness empathizing at the organizational level. The last group of
competencies, relationship management, reflects a modification to Goleman's (1995)
original conceptualization.
Boyatzis et al. (2000) reports that the need for the modification was
demonstrated through the analyses of data collected from nearly six hundred
corporate managers and professionals as well as engineering, management, and social
work graduate students. The original model was collapsed and much of what was
grouped in the "social skills" domain was placed under the relationship management
cluster. The relationship management cluster now includes the following social
competencies: developing others perceiving the developmental needs of others and
supporting the growth of their abilities; influence using interpersonal influence
tactics; communication transmitting unambiguous and persuasive messages; conflict
management resolving disagreements; visionary leadership inspiring and guiding
groups; catalyzing change managing or initiating change; build bonds nurturing
instrumental relationships; teamwork and collaboration creating a shared vision; and
synergy in teamwork working with others toward shared goals.
This conceptualization has drawn intense criticism from many scholars.
According to Mayer, Salovey, and Caruso (2000b), many of the listed traits are, to a
large degree, the product of genetic, biological or early-learning. Thus, they are not
36
readily amenable to training or change. Goleman (1998), however, contends that
emotional competencies are skills related to the job, which employees can and must
learn. Unlike the relatively stable IQ, emotional intelligence can improve with
maturity. More pointedly, he suggests that "...maturity itself describes this process of
becoming more intelligent about our emotions and our relationships." (p. 240)
The Goleman Emotional Competency Inventory (ECI)
Boyatzis et al. (2000) chronicles how the Emotional Competency Inventory
(ECI) was developed to operationalize Goleman's (1998) model. They report that
while many instruments were available to assess competencies in their clusters
through simulations, assessment centers, and behavioral event interviews, these
methods were not easy to implement, desirable in 360-degree applications,
comprehensive enough, or high in validity. As a basis for the ECI, they used an
assessment questionnaire developed earlier by Boyatzis. This had demonstrated good
validity in testing for competencies against performance in hundreds of previous
studies of North American managers, executives and leaders. About 60 percent of the
inventory consisted of new items, which were added to test for a broader set of
competencies applicable across all jobs and life settings. The instrument was
designed for use as a developmental tool only and includes a 360-degree assessment
instrument administered to self, subordinate, peer and supervisory respondents. Using
a 6-point scale, it calls for respondents to describe self or another person on each
competency. With 110 items, it can be completed in about thirty-five minutes
(Gowing, 2001).
Reviews of the instrument are generally mixed. For example, Gowing (2001)
reports that it ..."is supported by construct validity evidence, content validity
evidence, and validity generalization evidence from its predecessor instrument, the
37
self-Assessment Questionnaire At present, there is no evidence of convergent or
discriminant validity with measures of similar or different constructs" (p.92). Further,
Hedlund and Sternberg (2000) are also critical, saying that the validity evidence
Goleman advances does not support his definition of El and its ability to account for
variance in education or job performance beyond IQ. The major weakness of the
work for them is that it is based primarily on anecdotal evidence.
Many challenge the assertion that this approach to measuring an individual's
competencies represents a "new yardstick". For example, Sternberg (1999) comments
that for Goleman to suggest that how well people handled themselves in the past had
little bearing on hiring decisions or performance evaluations is certainly inaccurate.
He also notes that the breadth of Goleman's framework includes a combination of
abilities, personality traits, motivations, and emotional characteristics that stretches
the definition of intelligence to include anything that is not IQ. Indeed, some
"clusters" of the model may even include what have traditionally been considered
aspects of IQ measurement (Sternberg, 1999). Mayer (1999), in addressing the
accuracy of the "popular versions" of El, echoes this by saying:" ...the meaning of
emotional intelligence has been stretched. Emotional intelligence is now defined by
popular authors in dozens of waystypically as a list of personality characteristics,
such as empathy, motivation, persistence, warmth and social skills" (p.2). Because
these models blend many diverse parts of personality, he refers to them as mixed. His
point is that many of the variables contained in these models go beyond what is meant
by "emotion" or "intelligence" and are simply new ways to sell personality research
and prediction.
In his review of Goleman's (1998) book Working with Emotional
Intelligence, O'Shaugnessy (1999) observes that Goleman's claims are the same that
38
humanistic psychologists made in the 1950's and 1960's. Further, Gowing (2001)
refers to several researchers over the last fifty years who were advocates of
measurement approaches similar to Goleman's. For example, she notes that four
decades ago Katz (1955) advanced the idea that an effective administrator needs three
sets of fundamental skills: technical, conceptual, and human.
McCrae (2000), in evaluating "mixed models" of emotional intelligence
(Goleman's and others), draws attention to how many aspects of the Goleman model
"overlap" a widely used operationalization of the Five Factor Model (FFM) -- the
revised NE0 Personality Inventory (NEO-PI-R: Costa & McCrae, 1992b). As he
sees it, the mixed models are attractive to many because they bring together the
evaluatively positive extremes of each of the five factors. Thus, a high El evaluation
would be characterized by high scores for four factors i.e., extraversion, openness,
agreeableness, and conscientiousness, and low scores for neuroticism. He continues
his argument by suggesting that if El is simply a combination of personality traits,
decades of personality research can be drawn from to better define the domain and
that measures of each of the five personality factors should be used. Along this line,
other researchers have simply suggested that emotional intelligence is little more than
a set of personality variables and, as such, good measures have already been
developed (Davies et al., 1998).
If El is learned, or as malleable as Goleman (1998) suggests, then it would not
be genetically based. However, research in personality traits shows the importance of
genetics in their formation (Riemann, Angleitner, & Strelau, 1997). Additionally,
traits appear to be fairly stable in adulthood (Costa & McCrae, 1997). This does not
add support to Goleman's and other proponents of mixed models in respect to the
developmental potential of El. For example, in a recent interview, Goleman
39
contended that "people can change from being pessimists to being optimists in a
matter of weeks" (Toms, 1998, P. 115). This is contrary to empirical research,
indicating that while specific attitudes, behaviors, and organizational polices can be
changed, changes in personality are more problematic (Costa & McCrae, 1986).
What can be said about personality trait development is characterized by
recent research that suggests maturational trends (McCrae et al., 1999). In early
adulthood (late adolescence to age thirty) there is a decline in neuroticism,
extraversion, and openness and an increase in agreeableness and conscientiousness.
Even after age thirty, the changes appear to be in the same direction, albeit much
slower. With a decrease in neuroticism and increased agreeableness and
conscientiousness, the maturing process should reflect a higher El. However,
Sapolsky (1998) points out that with age extraversion and agreeableness decline,
suggesting the older one grows, the lower his or her El. McCrae (2000) suggests that
this may be a matter of being emotionally intelligent in different ways at different
ages. For example, where young persons may be more optimistic and more aware of
their emotions, they may be less effective in areas of persistence and impulse control.
This points out the need for differentiating between traits rather than relying on one
unified construct, such as a global measure of El.
In their comprehensive review of Goleman's work, Matthews et al., (2002)
summarized the views of other academics by saying that his "...model of El simply
represents a journalist distilling scientific information for the consumption of the
populist, rather than a legitimate scientific theory" (p.14). However, they did note
that the work has had value as a source of ideas that may prove to be of worth if they
can be developed and tested empirically.
40
Bar-On: Emotional Intelligence as Personality and Performance
A major theorist aligned, in part, with Goleman is Bar-On (2000). He began
his work in 1983 with a focus on emotional and social intelligence, which he defined
as: ..."an array of emotional, personal, and social abilities that effect one's overall
ability to effectively cope with daily demands and pressures; this ability is apparently
based on a core capacity to be aware of, understand, control, and express emotions
effectively." Drawing from both personality theory and work performance theory,
Bar-On (1997b) developed the Emotional Quotient Inventory (EQ-i) as a measure as
of emotional and social intelligence. Like Goleman and his colleagues, he contends
that non-cognitive intelligence is more important for life success. His theoretical
framework is based on five conceptual components encompassing fifteen factors. The
conceptual components and their associated factors are: intrapersonal EQ, including
emotional self-awareness, assertiveness, self-regard, self-actualization, and
independence; interpersonal EQ, including empathy, social responsibility, and
interpersonal relationship; adaptability EQ, including reality testing, flexibility, and
problem solving; stress management EQ, including stress tolerance and impulse
control; and general mood EQ, including optimism and happiness (Bar-On, 1997b).
Gowing (2001) draws attention to the similarities between Goleman's
emotional competence framework and the Bar-On EQ-i scales and subscales (see
Table 2.1). Bar-On's Intrapersonal scales are similar to Goleman's Personal
Competence. Similarly, emotional self-awareness relates to emotional awareness;
self-regard to accurate self-assessment and self-confidence; impulse control to self-
regulation and self-control; social responsibility to trustworthiness, conscientiousness,
and collaboration and cooperation; flexibility to adaptability; problem solving to
innovation; self-actualization is not unlike achievement drive; and optimism is much
41
the same as optimism. The interpersonal skills map generated from the EQ-i is very
similar to social competence. Empathy is a direct match with empathy; interpersonal
relationship is akin to building bonds. However, not included under Goleman's
framework are Bar-On's competencies of reality testing, stress tolerance, and
happiness.
Bar-On: The EQ-i
The Emotional Quotient Inventory (EQ-i) uses a self-assessment method to
measure emotional and social competent behaviors. Bar-On (2000) stresses that the
instrument was developed to measure emotional and social intelligence and not traits
or cognitive capacity. It has 133 items and uses a five-point Likert scale where
respondents rate their own behavior from "very seldom or not true of me" to "very
often true of me or true of me." The scores of the inventory are expressed as a total
EQ score and five EQ composite scale scores that comprise fifteen subscale scores.
Bar-On (2000) relates the EQ-i's effectiveness based on several years of
implementation and analysis. The findings obtained to date suggest that the total EQ
scale score correlates with various other measures that are thought to tap this
construct as well as closely related aspects of it. In detailing research conducted over
twelve years with over 6,300 respondents, Bar-On (1997b) reports reasonably high
internal consistency on the fifteen subscales. The average coefficient alphas ranged
from .69 to .86 across samples. While Bar-on asserts that the instrument has been
validated with many other measures such as personality inventories, his research has
not been published in peer-reviewed journals. Hedlund and Sternberg (2000)
comment that while exploratory and confirmatory factor analyses point to a good fit
of the model, its internal consistency is uncertain. This is important to know, given
42
the number of factors included in the EQ-i. They wonder if all fifteen factors are
adequately represented under the single general construct.
Summary
A recent review of the literature relating to "nonacademic intelligences"
highlights that while research on social intelligence enjoys some breadth, empirical
research on emotional intelligence is still very limited (Hedlund & Sternberg, 2000).
The empirical El literature has been dominated by attempts to validate the
operationalizations developed by various researchers. The acceptance of the construct
by the academic community has generally been weak. As one set of reviewers noted,
"Indeed, with the possible exception of Mayer, Salovey, Caruso, and colleagues, the
intelligence component of El receives short shrift in leading reviews and empirical
research devoted to the topic" (Matthews et al., 2002, p. 82). Although there is a
small but growing body of research that has investigated outcomes of El (e.g., (Law,
Song, & Wong, 2002; Law, Wong, & Song, 2004; Wong & Law, 2002), more
research in this area is needed. A general critique of much of the research is its lack
of transparency. For example much of Goleman's work has relied on studies
conducted by businesses and commercial organizations that (for competitive reasons)
are reluctant to readily share research findings. This lack of peer review retards the
exploration process greatly. Additionally, much of the research relies heavily on
atheoretical or anecdotal evidence (see (Matthews et al., 2002). There is a lingering
debate about the definitional distinctions between personality and what some consider
a new construct of El, or whether such a distinction is necessary. A final limitation is
the narrow focus, thus far, on developing antecedents, correlates and consequences of
El.
43
In the next chapter I develop a definition and model that attempts to address
some of these limitations. By considering El as ability (Mayer & Salovey, 1997),
1997), I construct hypotheses about the possible direct influence of El on a range of
performance dimensions. Additionally, the model posits how El abilities may
influence the personality-performance relationship.
44
CHAPTER 3
THEORETICAL FRAMEWORK
In this chapter, I develop a conceptual model of workplace performance and
individual differences based on the socioanalytic theory of Hogan and colleagues
behaviors, and a negative predictor of (e) relationship
disruptive behaviors.
To summarize, the model developed in this chapter predicts that both
personality and the various dimensions of Emotional Intelligence will have direct
influences on performance. Also, exploratory hypotheses regarding the interactive
effects of personality and Emotional Intelligence are offered. In the next chapter, I
present the methodology used to test these hypotheses.
71
CHAPTER 4
METHOD
In this chapter I describe the methods used to test the proposed model. First, I
discuss the overall research strategy and design. Next I discuss the data collection
procedures employed. Then I provide an overview of the sample that was surveyed.
This is followed by a description of the instruments used to operationalize the
variables. Finally, I describe the analytical procedures used to assess the hypotheses.
Research Strategy and Design
This was a quantitative, cross-sectional field study. The major advantage of a
field study is that it occurs in a non-contrived, natural context so that realism is
enhanced and the effects are stronger (McGrath, 1982). Collecting data in a non-
contrived setting by use of surveys goes to developing strong external validity
(Sekaran, 2000). The use of two or more independent measures, or triangulation, is
considered to be advisable to prevent the research from becoming method-bound
(Easterby-Smith, 1991). To this end, data triangulation was accomplished by
collecting data with two on-line instruments (one was an abilities test for assessing
Emotional Intelligence and the other was a self-report survey for assessing all other
variables) at two different time periods.
According to Sekaran (2000), the survey technique is generally acknowledged
as the most efficient means to carry out data collection, especially when dealing with
a large population with limited time. Arguably, the most significant communications
technological development in the last fifty years, the Internet or the World Wide Web
(WWW), is being used increasingly in academic research. Many recent scholarly
journals attest to its usefulness (e.g., Kelly-Milburn & Milburn, 1995: Landis, 1995).
These and other advocates of the use of this new technology as a research tool
72
compare it favorably with traditional pencil and paper media. Kraut and Saari (1999)
observed that the technology is increasingly a vital part of an organization's
infrastructure as employees are connected via e-mail, Internet, intranet and electronic
surveying has increased. The potential for lower data collection costs is significant.
Donovan (2000) highlighted the convenience and global reach of online data
collection. Schaefer and Dillman (1998) pointed out that the speed of data collection
over paper instruments is enhanced. Still, another writer has noted that the electronic
survey can grant the researcher greater control (Dillman, 2000). Several researchers
further note that employees may actually prefer this mode of data collection
(Christianson DeMay & Toquam, 2001; Church, 2001; Thompson, Surface, Martin
and Sanders, 2003).
In respect to the effectiveness and the integrity of data generated by online
methods, many scholars offer evidence that supports its adoption. For example,
comparison studies with online and paper instruments report no difference in response
rates (Fenlason, 2000; Yost and Homer, 1998) ; online instruments display the same
or a lower proportion of missing values (Stanton, 1998; Fenlason, 2000); and
completion rates are generally the same (Church, 2001) .
Thompson, Surface, Martin and Sanders (2003) showed that influences of
Web response rates included past satisfaction with past online surveys, the absence of
problems with technology, and the subjects' overall evaluation of the usefulness of
the media. They found also that demographic differences such as gender, race, and
military versus civilian occupation did not influence participation rates. Finally, no
differences in measurement equivalence between paper and Web-based modes have
been found by many recent investigators (e.g., Donovan, Drasgow and Probst, 2000;
Fenlason, 2000; Magnan, Lundby and Fenlason, 2000; Young, Daum, Robie and
73
Macey, 2000). Thus, the choice of an online questionnaire and an abilities test as
data collection instruments was made based on the increased economies, speed and
effectiveness they offered. I also thought that the subjects would benefit by having
the ability to complete the instruments at a time of their choosing, thus increasing
their willingness to participate.
Using two different surveys meant that data were collected at two different
points in time, with two types of instruments a self-report survey and an abilities
test. A large proportion of the bias in estimated connections between constructs
measured with common methods is due to transient factors. Thus possible problems
associated with common method variance (CMV) were mitigated. Additionally, most
theories of CMV argue that cognitive limitations/tendencies are the main source of
problems (see (Podsakoff, MacKenzie, Lee, & Paodsakoff, 2003). Separating by time
greatly reduces these (see (Ostroff, Kinicki, & Clark, 2002).
Data Collection
To collect data, I targeted managerial-level employees working in Hong Kong.
To maximize variability across occupations and industries, I used a convenience
sampling technique to solicit subjects to participate in the study. As a form of
nonprobablity sampling, convenience sampling is generally believed to undermine the
generalizability of research findings (e.g. Aaker, Kumar and Day, 1995; Sekaran,
2000). However, as the focus of this research is model-based rather than person-
based, I felt it justified. Mook (1983) argued that a preoccupation with external
validity may indeed be detrimental to good research, especially when generalizations
to real life are not its intention.
Two basic criteria were applied to screen for potential subjects. First,
subjects were selected based on their fluency in English. University education was
74
used as a proxy for judging fluency because the language of instruction at all Hong
Kong universities is English. Several other reasons influenced this decision to use
English rather than Chinese: (1) The El measures employed (MSCEIT) were first
developed and validated in English, and are not yet available (as copyrighted
instruments) in Chinese; (2) as a former British colony, Hong Kong's second official
language is English; and (3) the need for English communications skills is high,
especially for mid to senior level Hong Kong mangers as is witnessed by continuing
reports in the popular press and in academic research. For example, Lundelius (1997)
gave some indication of its importance in the Hong Kong workplace when he noted,
"For use in nearly all types of written business communications, nearly twice as many
respondents say that English is used rather than Chinese."(p.112). Thus, the basic
criterion of English proficiency was established for selection.
The second criterion used for selection was management experience.
Individuals were invited to take part if they had experience in a management position
at front-line, middle or senior levels. Once again, enrollement in university programs
designed for working adults and executive development programs was used as an
initial screen. This was done both as a matter of convenience and to ensure
respondents had previous or current managerial work experience. No control group
was employed. Thus, effects that these developmental programs may have had on the
subjects' responses were not explored. However, insofar as all participants had
experienced executive development programs, the influence of such programs is
constant across participants. Thus, it should not have any effect if terms of the
putative connections between variables.
On-line surveying was conducted over a six-month period from September
2003 to February 2004 in two phases. The first phase involved soliciting participation
75
through email invitations. To find subjects that met the basic criteria, I approached
students in post-graduate classes at three Hong Kong universities and participants
from executive development programs that I had conducted for two Hong Kong
organizations (the Hong Kong Government and an insurance firm). As a high level of
English fluency is required for enrolment in these classes, the language requirement
was met. These classes were composed of current and former part-time students who
held management positions in a wide-range of industries. Thus, the second criterion of
management experience was also achieved. I then sent an email invitation to
participate. The invitation directed the potential subject to the first of two an Internet
websites.
Labeled "eiplace.com" the initial website included more background
information outlining the scope of the study, the participants' involvement and
instructions on completing the first part of the survey. The first instrument measured
personality factors, demographics, and performance variables. After completing part
one of the survey, respondents were directed to part two, the MSCEIT V2 emotional
intelligence abilities test. This was also an on-line instrument administered by
Multiple Heath Systems.
Although there are no figures available in respect to expected participation
rates for on-line instruments, Harzing (1997) noted that for international academic
mail surveys, Hong Kong ranked lowest among 22 countries at 7.1%. Various
measures were used to increase the response rate. First, to obtain email addresses for
sending the invitation, I (or a trained representative) visited each class explaining the
general nature of the study and requested their involvement. Secondly, all email
correspondence featured the letterhead of the International Graduate School of
Management, The University of South Australia and the endorsement of my
76
supervisors, Dr. David Harrison and Dr. Barry Elsey and me. According to Bruvold
and Corner (1988) and Schneider and Johnson (1995) questionnaires endorsed by
reputable persons or renowned organizations will elicit higher response willingness.
Secondly, because participation in this study was voluntary, every effort was
made to prepare easy-to-read and appealing on-line questionnaires. Erdogan and
Baker (2002) observed that once the respondents invest the time to read a
questionnaire, they will develop a degree of psychological commitment to complete
it.
Third, every effort was made to assure the respondents of confidentiality. The
study's purpose was explained explicitly and promises that the data collected would
be used solely for research were given in the invitation and the background
information sheet given to the participants before they gave their informed consent.
Throughout the study, respondents were requested to only identify themselves by
means of their email address and my local server and the MHS testing authority's
secure Internet server captured all the data. Fourth, in return for their participation, I
offered them a personal summary of their Emotional Intelligence scores. Finally, in
instances where invited subjects did not respond within two weeks of the invitation, a
reminder email was sent. Subsequently, after one month another, final email
remainder was sent.
These procedures resulted in an above-average response rate for Hong Kong
surveys. Out of the 509 email invitations sent, 484 were confirmed received (25 were
not delivered, perhaps due to incorrect email addresses). Thus, of the 484 confirmed
invitations, 116 subjects completed both instruments, for a response rate of 24%.
Based on a power analysis, this sample size is adequate for detecting small to
moderate effects. According to Cohen (1988), for a power of .80 and a significance
77
criterion of .05, a sample size of 84 is needed to detect a small to medium (i.e., .30)
effect size.
The on-line surveys were presented in a way that participants were required to
complete the performance and personality measures before accessing the Emotional
Intelligence test. This order was intended to ensure that participants completed both
surveys. The inducement was that they would receive an assessmentof their
Emotional Intelligence, so this survey was administered last. Nevertheless, several
potential participants (N = 72) only completed the first survey, thus rendering their
(partial) input invalid.
Sample
The sample represents a diverse group of management-level employees. Only
13% of the sample reported that their native language was English, and 84% of the
respondents reported that it was Cantonese or Mandarin. Another 3% reported that it
was another language. However, all were fluent in English. Correspondingly, most
(88%) were of Asian descent and the rest (12%) were Caucasians. The gender
distribution was 37% male and 63% female. The majority of respondents were over
30 years of age, and they had, on average, 7.41 years of work experience. The
respondents were highly educated, with 91% having at least a bachelor's degree.
Most (86%) were in management positions, and 14% were in supervisory posts. They
represented a wide variety of industries, including the public sector, professional
services, financial/insurance services, and manufacturing. Details of these
characteristics are presented in Table 4.1.
Measures
In this section, I describe the measures used to operationalize the variables.
All measures were taken from existing studies, and all have demonstrated sound
78
Table 4.1. Demographic Profile of Respondents
79
Demographic Characteristics Number Percentage
GenderMale 43 37
Female 73 63
Age20-25 7 6
26-30 18 16
31-35 27 23
36-40 33 28
41-45 13 11
46-50 9 8
51+ 9 8
EthnicityAsian 102 88
Caucasian 14 12
LanguageCantonese 94 81
Mandarin 3 3
English 15 13
Other 4 3
EducationForms 5-7 3 3
Associate's Degree 5 4
Bachelor's Degree 55 47Post-graduate Diploma 11 9
Master's Degree 33 28
Doctoral Degree 9 8
PositionTop-level Management 18 16
Middle-level Management 58 50
Lower-level Management 23 21
Supervisory Level 16 13
IndustryAgriculture, Forestry, Hunting and Fishing 1 1
Hotel/tourism 1 1
Transport and Allied Services 1 1
Manufacturing 5 4
Communication 6 5
Electricity, Gas and Water 1 1
Financial Services/Insurance 16 14
Construction 2 2
Wholesale, Import/Export 3 3
Business/Professional Services 19 16
Retail 5 4
Community, Social and Personal Services 6 5
Restaurant/Food Services 2 2
Government 35 30
Education 13 11
psychometric properties. The performance, personality and demographic items were
on the first on-line survey (see Table 4.2 for a complete listing of the performance and
personality items). After completing that survey, participants were directed to the
second survey, which assessed their Emotional Intelligence (see Table 4.3 for sample
items). Both on-line surveys were in English.
Performance. Performance was conceptualized and operationalized in terms
of five variables: task and contextual performance, innovative behaviors, relationship
supportive behaviors, and relationship disruptive behaviors. The first three
performance variables were assessed with role-based performance scales developed
by Welbourne, Johnson, and Erez (1998). The response format for these scales
ranged from 1 = needs much improvement to 5 = excellent. Task performance was
measured with four items; a sample item was 'quantity of work output.' The
contextual performance scale consisted of four items about behaviors directed toward
the organization. A sample item was 'working for the overall good of the company.'
Innovative behaviors were assessed with four items, such as 'working to implement
new ideas.' Welbourne et al. (1998) reported reliabilities ranging from .59-.87, .72-
.84, and .87-.91 for the task, contextual, and innovator scales, respectively.
Relationship supportive and relationship disruptive behaviors were measured
with scales developed by Sin, Harrison, Shaffer and Lau (2004). These scales
comprised ten items each, and responses were on a 7-point frequency scale ranging
from "never" (0) to "always" (6). Response anchors taken from the psychometric
research of Bass, Cascio, and O'Connor (1974). A sample item for the relationship
supportive behavior scale is 'shows interest in and knowledge of many different
topics of conversation.' For the relationship disruptive scale, a sample item is 'has a
80
Table 4.2. Items to Assess Performance and Personality
81
Variable ItemsPerformance
Task 1. Quantity of work outputQuality of work outputAccuracy of workCustomer service provided (internal and/or externalcustomers)
Contextual 1. Doing things that help others when it's not part ofthe jobWorking for the overall good of the companyDoing things to promote the companyHelping other coworkers so that the company is agood place to be
Innovative 1. Coming up with new ideasWorking to implement new ideasFinding improved ways to do thingsCreating better processes and routines
Relationship 1. Remembers birthdays and special eventsSupportive Behaviors 2. Others go to him/her for advice when making
important decisionsDiscusses problems openlyIs willing to talk about his/her personally matters,family or non-work interestsKeeps track of what's happening to others' familymembers and close friendsVisits others when they're sick or in the hospitalShows interest in and knowledge of many differenttopics of conversationOrganizes informal social activitiesMediates differences of opinions among othersSends postcards or email messages back to his/herworkplace while he/she is on holiday/vacation
Relationship 1. Tends to ignore feedback from othersDisruptive Behaviors 2. Has a number of ongoing conflicts with colleagues
Interrupts or cuts off others who are talkingBrags about his/her contributionsIn case of disagreements, he or she will insult her/hisadversarySeems to enjoy stirring up argumentsTreats others at work as his/her competitorsIs often suspicious of the well-intentioned acts ofothers
Loses his/her temper easilyRefuses to provide information to others
PersonalityExtraversion 1. Is the life of the party
Doesn't talk a lotFeels comfortable around people
Variable
Agreeableness
Conscientiousness
Emotional Stability
Intellectance
ItemsStarts conversationsKeeps in the backgroundHas little to sayTalks to a lot of different people at partiesDoesn't like to draw attention to him/herselfDoesn't mind being the center of attentionIs quiet around strangersFeels little concern for othersIs interested in peopleInsults peopleSympathizes with others' feelingsIs not interested in other people's problemsHas a soft heartIs not really interested in othersTakes time out for othersFeels others emotionsMakes people feel at ease
Is always preparedLeaves his/her belongings aroundPays little attention to detailsMakes a mess of thingsGets chores done right awayOften forgets to put things back in the proper placeLikes orderShirks his/her dutiesFollows a scheduleIs exacting in his/her workGets stressed out easilyIs relaxed most of the timeWorries about thingsSeldom feels blueIs easily disturbedGets upset easilyChanges his/her mood a lotHas frequent mood swingsGets irritated easilyOften feels blue
Has a rich vocabularyHas difficulty understanding abstract ideasHas a vivid imaginationIs not interested in abstract ideasHas excellent ideasDoes not have a good imaginationIs quick to understand thingsUses difficult wordsSpends time reflecting on thingsIs full of ideas
82
number of ongoing conflicts with colleagues.' Sin et al. (2004) report reliabilities
ranging from .74-.75 and .75-.87 for the supportive and disruptive behavior scales,
respectively.
Personality. Big Five factor markers from Goldberg (2000b) were used to
assess personality traits. Ten items assessed each of five subscales: extraversion (e.g.,
"is the life of the party"), agreeableness (e.g., "is interested in people"),
most of the time"), and openness to experience (e.g., "have a vivid imagination").
Responses were made using a 5-point Likert scale, 1 (very inaccurate) to 5 (very
accurate). Goldberg (2000a) reported estimated reliabilities ranging from .75-.90.
Emotional Intelligence. Respondents' emotional intelligence was assessed by
means of the on-line version of the Mayer Salovey Caruso Emotional Intelligence
Test (MSCEIT V2). This is an ability-based assessment designed by the authors of
the four-branch model of emotional intelligence and administered by the professional
assessment organization, Multi-Health Systems Inc. (MHS). Managers were invited
to "log-on" to a designated secure server at their convenience any time during a two-
week period to take the 30 to 45 minute test. MHS scored the test and returned
individual and aggregate results directly to me electronically.
The test consists of 141 items that are designed to measure the specific skills
associated with each of the four branches of the Mayer and Salovey (1997) model:
perceiving emotions, using emotions to facilitate thought, understanding emotions,
and managing emotions (see Table 4.3 for sample items). The four branches are each
measured by two tasks. The test developers varied the response formats to gain better
generalizability across tasks and to reduce the possibility of correlated measurement
error. That is, where some tasks were measured using a 5-point scale, others used a
83
multiple- choice structure. Test validity across educational levels, occupational and
ethnic groups, and geographic locations has been reported (see Mayer, Salovey,
Caruso, & Sitarenios, 2003afor a detailed discussion of the MSCEIT).
Table 4.3: Sample Items from the MSCEIT
El BranchBranch 1 PerceivingEmotions
Indicate theemotionsexpressed by thisace.
Happines
Branch 2 Facilitating What mood (s) might be helpful to feel when meetingEmotions in-laws for the very first time?
Not UsefulUseful
Branch 3 UnderstandingEmotions
Branch 4 ManagingEmotions
Tom felt anxious, and became a bit stressed when hethought about all the work he needed to do. Whenhis supervisor brought him an additional project, hefelt . (Select the best choice.)
OverwhelmedDepressedAshamedSelf ConsciousJittery
Debbie just came back from vacation. She wasfeeling peaceful and content. How well would eachaction preserve her mood?
Action 1: She started to make a list of things at homethat she needed to do.
Very Ineffective..1 2 3 4 5..Very Effective
To measure the ability to perceive emotions (Branch One), participants
responded to 20 items that required viewing a series of 4 faces. For each face, they
84
Tension 1 2 3 4 5
Surprise 1 2 3 4 5
Joy 1 2 3 4 5
Sam le Item
000:11
11111111001411Sadness
001411
indicated on a 5-point (1 = no feeling to 5 = extreme feeling) the extent to which they
perceived five specific emotions. For example, for one face, respondents rated how
much happiness, fear, surprise, disgust, and excitement were expressed by the face.
Another 30 items consisted of respondents viewing 6 pictures of landscapes and
abstract designs and indicating the extent to which each of 5 feelings (e.g., happiness,
fear, anger, surprise and disgust) was expressed by the picture.
Facilitating thought (Branch Two) was measured with 15 items that asked
respondents to match 3 sensations with each of 5 emotions. For example, respondents
were asked to imagine feeling guilty and then they rated how alike (1 = not alike to 5
= very much alike) that feeling was to sensations such as cold and sweet. Another 15
items consisted of 5 situations involving cognitive or behavioral tasks. For each
situation, respondents rated three moods that would be helpful in carrying out the task
(1 = not useful to 5 = useful). For example, when creating decorations for a birthday
party, respondents were asked to indicate how useful it would be to feel boredom or
joy.
Understanding emotions (Branch Three) was measured with 12 items that
asked respondents to identify emotions that may be combined to form other emotions.
A sample item is 'acceptance, joy, and warmth often combine to form
Respondents then chose from one of 5 emotions to complete the statement. Another
20 items required respondents to select an emotion that may result from the
intensification of another feeling. For example, respondents were asked to indicate
the emotion (e.g., depression) that would be the most likely outgrowth of intensified
sadness and fatigue.
Managing emotions (Branch Four) was measured with 20 items. Respondents
were presented with 5 stories and asked to judge the effectiveness of 4 options for
85
reaching a specified emotional outcome. For example, respondents were asked to
judge what a person might do to reduce anger or extend joy. Another 9 items required
respondents to judge the effectiveness of behaviors in the management of another
person's feelings.
Just as its predecessor, the MSEIT, the MSCEIT renders many scores: a total
measure of El, two areas scores (experiential and strategic); a score for each of the
four branches; and eight task scores. Scoring can be done in two ways i.e., using
either general consensus or expert methods. Using the general consensus method
involves scoring the respondent's answers against the proportion of the sample that
selected the same answer. For example, if an individual answered that anger was
"definitely present" in a face, and the same answer was given by 50% of the sample,
the individual's score would be incremented by the proportion .50. The individual's
total raw score is the total of the 141 items across the test. Alternatively, comparisons
can be made of individual responses to a group of experts' responses. The expert
scoring method used by the MSEIT was criticized for its use of only two experts
(Matthews et al., 2002). The MSCEIT V2.0 addressed this criticism by incorporating
21 experts in its development and testing.
In scoring the tests by general consensus and expert methods, differences in
reliabilities were reported. At the total El score level, the general method reliability
was .93 and the expert method was .91. Reliabilities for the four branch scores were
more varied. For the general scoring method, reliabilities were: Branch 1= .91, Branch
2=.79, Branch 3=.80, and Branch 4=.83. In contrast, the expert scoring method
resulted in reliabilities of: Branch 1=.90, Branch 2=.76, Branch 3.77 and Branch
4=.81 (Mayer et al., 2003a).
86
In commenting on the differences between the two methods, Mayer and his
colleagues report that for emotions "...experts are more reliable judges, and converge
on correct answers where research has established clear criteria for answers"(Mayer et
al., 2003a). Therefore this was the scoring method of choice for the current study.
Control Variables. Three control variables were used in the analyses. Gender
was assessed by asking respondents to indicate whether they were male (coded 1) or
female (coded 2). To measure age, respondents selected from one of 7 categories:
20-25, 26-30, 31-35, 36-40, 41-45, 46-50, and over 50. Ethnicity was assessed by
having respondents indicate whether they were Asian (scored 1) or Caucasian (scored
2). An option was available for participants to write-in another ethnic choice, but
none was reported.
87
CHAPTER FIVE
DATA ANALYSES AND RESULTS
In this chapter, I explain how I analyzed the data and the results. First, I give
an overview of the analytical procedures used to assess the data. Then I present an
overview of the quality of the data. I next present descriptive statistics and
correlations of the variables personality, emotional intelligence, performance and
controls. Following this, I provide the regression results of the tests of the
hypotheses.
Analytical Procedures
Before testing the hypotheses, I first examined the quality of the data by
checking to ensure it was normally distributed and that common method variance was
not a major issue. Next, I assessed the internal consistency of each scale by
calculating Cronbach alphas. I then looked at the means and standard deviations. An
examination of the zero-order correlations indicated preliminary evidence for
proposed relationships.
To test the hypotheses, I conducted multiple hierarchical (moderated)
regressions on all performance measures (task, contextual, innovative, relationship
supportive behaviors, and relationship disruptive behaviors). To test the direct effects
of the Big Five personality traits and the four branches of El, I entered these into
regression equations after entering the control variables. To test for the moderating
effects of El, I first calculated a total El score by adding the scores from the four
branches together. Then, I created interaction terms by multiplying each of the Big
Five personality traits by the total El score. For the regressions, I entered the
variables in this sequence: controls, personality, total El, and interaction terms (El x
personality). To minimize multicollinearity among interaction terms and their
88
constituent terms in the regression models, I centered all variables with the mean (i.e.,
I transformed the scales to 1 = -2, 3 = 0, and 5 = 2 for 5-point scales: Aiken & West,
1991).
For each regression model, the coefficient of determination, R2 or the
goodness-of-fit measure, was computed. R2 ranges from 0 to 1 and is the proportion
of variation in the dependent variable explained by the regression model. Small
values indicate that the model does not fit the data well. Hypotheses were tested by
analyzing the significance of the coefficients of the regressions. Positive (negative)
coefficients denote a positive (negative) relationship between the predictor variable
and the criterion variable. The t statistic determines the importance of each variable
in the model. A significance value of p < .05 supports the relationship between the
dependent and independent variables.
Data Quality
To evaluate the quality of the data, I first examined the normality of the
distributions (Hair, Anderson, Tatham, & Black, 1998). To do this, I looked at the
skewness and the kurtosis. Skewness is a measure of the asymmetry of a distribution.
In skewness equals zero, the distribution is symmetric. A distribution with a
significant positive skewness exhibits a long right tail, called a downward straggle; a
significant negative skewness possesses a long left tail, called an upper straggle. If
skewness is less than one, it indicates that the distribution would not differ
significantly from a normal, symmetric distribution (Norusis, 2001). All of the
variables had skewness scores less than 1.0, therefore assumption of data normality
was valid.
Kurtosis is a measure of the extent to which the observations gather around a
central point. In a normal distribution, the value of the kurtosis statistic is zero.
89
Positive kurtosis, termed leptokurtic, suggests that the observations gather more and
have longer tails than those in the normal distribution. Negative kurtosis, called
platokurtic, indicates that the observations pack less and form shorter tails. A
distribution is considered normal if the kurtosis value is within the range of 3 to +3
(Norusis, 2001). All variables were within this range, so normal distributions were
assumed.
The El measures were assessed at a separate time and with a separate
instrument (i.e., an abilities test), so common method variance should not be a
problem with these scores (Harrison, McLaughlin, & Coalter, 1996). However, the
personality and performance measures were collected at the same time. Consequently
there was a potential threat of common method variance, in which any defect in the
source would contaminate both measures in the same fashion and in the same
direction (Podsakoff & Organ, 1986). Although some researchers (e.g., Crampton &
Wagner, 1994) argued that common method variance will not necessarily invalidate
empirical findings, Harman's one factor test (Harman, 1967) was performed to assess
the presence of common method variance. The underlying assumption of this method
is that if a substantial amount of common method variance exists in the data, either a
single factor will emerge or one general factor will account for the majority of the
covariance among the variables. Although this test does not completely rule out the
existence of common method effects, it provides some post hoc statistical support for
the absence of such bias in the findings, and it increases confidence in the substantive
interpretations made based on the results.
To assess the possibility of CMV, I entered all the independent and dependent
variables from the self-report survey into a factor analysis (Promax rotation). Sixteen
factors emerged, and the results of the un-rotated factor analysis indicated that no
90
single factor accounted for over 50% of the covariation in the variables. Factor 1
explained 10.73 of the variance, and all other factors explained less than 5% variance
each. These results suggest that common method variance is probably not a problem
in the data. See Tables 5.1 and 5.2 for factor loadings for independent and dependent
variables, respectively. Because some items failed to load or loaded on more than one
factor, they were eliminated from the scales.
Descriptive Statistics
Having established the quality of the data, I next computed descriptive
statistics. In a normal distribution, two summary measures, mean and variance, are
sufficient to describe an entire distribution. Table 5.3 reports the means, standard
deviations, zero-order correlations, and internal consistency reliabilities (i.e.,
Cronbach's alphas) for all variables. Cronbach's alphas are reported on the diagonal
in parentheses.
The Pearson correlation coefficient is a measure of the linear association
between two variables. Values range from to +1, with the absolute value denoting
the strength (i.e., larger values reflect stronger relationships) and the sign showing the
direction of the relationship. In this study, all of the values except one (the correlation
between task and innovative performance was r = .81) were below r = .75, which is
the threshold for rejecting multicollinearity (Sekaran, 2000). To further examine
multicollinearity, which will reduce the power of the test of coefficients in regression
analyses, the variance inflation factor (VIF) associated with each independent variable
in the regression equations was examined. The maximum VIF calculated for each
independent variable was less than six, which is the cutoff threshold (Maruyama,
1998). As a result, multicollinearity was ruled out as a problem in this study.
91
Tab
le 5
.1: F
acto
r L
oadi
ngs
of I
ndep
ende
nt V
aria
bles
92
Item
sE
mot
iona
l Sta
bilit
yE
xtra
vers
ion
Inte
llect
ance
Con
scie
ntio
usne
ssA
gree
able
ness
I fe
el c
omfo
rtab
le a
roun
d pe
ople
..6
2
I st
art c
onve
rsat
ions
..8
0I
talk
to a
lot o
f di
ffer
ent p
eopl
e at
par
ties.
.72
I do
n't m
ind
bein
g th
e ce
nter
of
atte
ntio
n..7
6I
am in
tere
sted
in p
eopl
e..4
4
I sy
mpa
thiz
e w
ith o
ther
s' f
eelin
gs.
.79
I ha
ve a
sof
t hea
rt.
.76
I fe
el o
ther
s' e
mot
ions
..5
7
I m
ake
peop
le f
eel a
t eas
e.V
.41
I am
alw
ays
prep
ared
..4
7
I pa
y at
tent
ion
to d
etai
ls.
.67
I ge
t cho
res
done
rig
ht a
way
..6
2
I lik
e or
der.
.64
I fo
llow
a s
ched
ule.
.65
I am
exa
ctin
g in
my
wor
k..6
2
I ge
t str
esse
d ou
t qui
ckly
..6
4I
wor
ry a
bout
thin
gs.
.57
I am
eas
ily d
istu
rbed
..6
0I
get u
pset
eas
ily.
.85
I ch
ange
my
moo
d a
lot.
.67
I ha
ve f
requ
ent m
ood
swin
gs.
.76
I ge
t irr
itate
d ea
sily
..7
8I
ofte
n fe
el b
lue.
.62
I ha
ve a
ric
h vo
cabu
lary
..7
5
I ha
ve a
viv
id im
agin
atio
n..6
0I
have
exc
elle
nt id
eas.
.44
Tab
le 5
.2: F
acto
r L
oadi
ngs
of D
epen
dent
Var
iabl
es
93
Item
sIn
nova
tive
Con
text
ual
Tas
kR
SBPe
rfor
man
ceR
DB
Perf
orm
ance
Perf
orm
ance
Qua
ntity
of
wor
k ou
tput
..6
6
Qua
lity
of w
ork
outp
ut.
.76
Acc
urac
y of
wor
k..7
6
Cus
tom
er s
ervi
ce p
rovi
ded
(int
erna
l and
/or
exte
rnal
).5
6
Com
ing
up w
ith n
ew id
eas.
.83
Wor
king
to im
plem
ent n
ew id
eas.
.84
Find
ing
impr
oved
way
s to
do
thin
gs.
.75
Cre
atin
g be
tter
proc
esse
s an
d ro
utin
es.
.65
Doi
ng th
ings
that
hel
p ot
hers
whe
n its
not
par
t of
my
job.
.61
Wor
king
for
the
over
all g
ood
of th
e or
gani
zatio
n..7
3
Doi
ng th
ings
to p
rom
ote
the
orga
niza
tion.
.68
Hel
ping
cow
orke
rs s
o th
at th
e or
gani
zatio
n is
a g
ood
plac
e to
be.
.77
Rem
embe
r bi
rthd
ays
and
spec
ial e
vent
s..7
2D
iscu
ss p
robl
ems
open
ly.
.50
Am
will
ing
to ta
lk a
bout
my
pers
onal
mat
ters
, fam
ily, o
r no
n-.7
0
wor
k in
tere
sts.
Kee
p tr
ack
of w
hat's
hap
peni
ng to
oth
ers'
fam
ily m
embe
rs a
nd.8
3
clos
e fr
iend
s.V
isit
othe
rs w
hen
they
're s
ick
or in
the
hosp
ital.
.77
Item
sE
mot
iona
l Sta
bilit
yE
xtra
vers
ion
Inte
llect
ance
Con
scie
ntio
usne
ssA
gree
able
ness
I us
e di
ffic
ult w
ords
..6
9
I sp
end
time
refl
ectin
g on
thin
gs.
.55
I am
ful
l of
idea
s..5
8
94
Item
sIn
nova
tive
Con
text
ual
Tas
kR
SBPe
rfor
man
ceR
DB
Perf
orm
ance
Perf
orm
ance
Show
inte
rest
in a
nd k
now
ledg
e of
man
y di
ffer
ent t
opic
s of
conv
ersa
tion.
.49
Org
aniz
e in
form
al s
ocia
l act
iviti
es.
.67
Bra
g ab
out m
y co
ntri
butio
ns.
.63
Enj
oy s
tirri
ng u
p ar
gum
ents
..6
8
Tre
at o
ther
s at
wor
k as
com
petit
ors.
.65
Am
oft
en s
uspi
ciou
srof
the
wel
l-in
tent
ione
d ac
ts o
f ot
hers
..5
9
Send
pos
tcar
ds o
r em
ail m
essa
ges
back
to m
y w
orkp
lace
whi
le I
am o
n ho
liday
/vac
atio
n.
.71
Los
e m
y te
mpe
r ea
sily
..6
7
Ref
use
to p
rovi
de in
form
atio
n to
oth
ers.
.56
However, the correlations among the three role-based forms of performance
(i.e., task, contextual, and innovative) were high especially the relationship between
task and innovative performance as noted above. This indicates that these three
dimensions are all tapping into a general overall measure of performance.
Consequently, I formed a measure of 'overall performance' by adding the scores for
these three forms together. I then included this as another dependent variable and
conducted the same regressions as with other dependent variables (see Tables 5.4 and
5.5).
The zero-order correlations (see Table 5.3) were stronger for the personality-
performance relationships than for the El-performance relationships. Task
performance and innovative performance were positively and significantly related to
all of the Big Five personality traits. Contextual performance was only related to
extraversion and agreeableness; RSBs were significantly related to extraversion,
agreeableness, and intellectance; RSBs were negatively related to extraversion,
conscientiousness, and intellectance. Correlations involving El were primarily related
to RSBs, with all four branches correlating negatively with this dimension of
performance. Branch One (perceiving emotions) was negatively associated with
contextual performance and Branches Two (facilitating emotions) and Three
(understanding emotions) were positively related to innovative performance.
The reliability coefficient alphas for each scale are recorded on the diagonal in
Table 5.3. Cronbach alpha is a measure of internal consistency based on the average
inter-item correlations among scale items. It measures the degree to which the items
are measuring the same theoretical construct. Alpha values in this study ranged from
.66 to .90, all of which were greater than the minimum threshold of .60 (Nunnally &
Bernstein, 1994). While Cronbach alpha reliability coefficients below .60 are
95
generally considered to be poor and those at .70 or better are acceptable (Sekaran,
2000) those ranging from .60 to .69 may also be acceptable, albeit modest. For
example Schmidt and Hunter (1996) showed that the average pairwise inter-rater
estimates of reliability for job performance are rarely above .60. Yet we still continue
to use pairs of raters in research.
Hypothesis Tests
Effects of Personality on Performance
The results for the effects of personality on performance are reported in Table
5.4. As predicted by Hypothesis 1, extraversion was a significant positive predictor of
relationship supportive behaviors (B = .35, p < .001). Agreeableness was significantly
and positively predictive of contextual performance (B = .28, p < .01) and relationship
supportive behaviors (B = .25, p < .01); it was also a significant, negative predictor of
relationship disruptive behaviors (B = -.22, p < .01). Thus, Hypotheses 2b, d and e
were supported, but Hypotheses 2a and c were not. Conscientiousness had a
significant influence only on task performance (13= .28, p < .01) and on overall
performance (B = .19, p < .05), so Hypothesis 3a was supported but Hypotheses 3b-e
were refuted. Emotional stability was an important negative predictor of relationship
disruptive behaviors (B = -.22, p < .050), as predicted by Hypothesis 4e. However, it
had no effect on the other forms of performance; thus Hypotheses 4a-d were not
supported. Intellectance had no effect on any of the performance dimensions, so
Hypotheses 5a-e were also not supported.
Effects of El on Performance
Table 5.4 presents the results for the influence of El on performance.
Contrary to expectations, Branch One (perceiving emotions) was a significant
negative predictor of contextual performance and Branch Two (managing emotions)
96
was a significant negative predictor of relationship supportive behaviors. However,
Branches one (perceiving emotions: B = -.22, p < .01) and Two (facilitating emotions:
B = -.24, p < .01) were significant negative predictors of relationship disruptive
behaviors. Thus, Hypothesis 6e was partially supported, but Hypotheses 6a-d were
not supported.
Moderating Effects of El
According to Hypotheses 7a-e, El would moderate the relationship between
personality and performance (see Table 5.5). Significant moderating effects were
found for all forms of performance except relationship supportive behaviors.
Agreeableness was involved in significant interactions with El for task (B = .52, p <
.05), contextual (B = .47, p < .05), innovative (B = .73, p < .001), and overall (B = .69,
p < .001) performance. Two significant interactions affected relationship disruptive
behaviors: conscientiousness x El (B = .22, p < .05) and emotional stability x El (B =
.18, p < .05). As predicted, El enhanced the effects of agreeableness on task,
contextual and innovative performance However, the relationship between
intellectance and innovative performance was stronger for those with low levels of El.
For relationship disruptive behaviors, El counteracted the negative relationships
involving conscientiousness and emotional stability. Thus, Hypotheses 7a, b, c, and e
were partially supported, but Hypothesis 7d was not. Graphical representations of
these moderating relationships are depicted in Figures 5.1 to 5.7.
97
Tab
le 5
.3: D
escr
iptiv
e St
atis
tics,
Cor
rela
tions
and
Cro
nbac
h's
Alp
has
Var
iabl
esM
ean
s.d.
Tas
k Pe
rfor
man
ce3.
76.6
5
Con
text
ual P
erfo
rman
ce3.
66.6
8In
nova
tive
Perf
orm
ance
3.42
.81
.57*
**.5
8
RSB
4.28
1.08
.30*
**.4
5***
.38*
**(.
83)
RD
B1.
93.6
7-.
06.0
6-.
00.0
7(.
77)
Ext
rave
rsio
n3.
19.7
7.2
7**
.35*
**.3
4***
.49
Agr
eeab
lene
ss3.
84.5
2.2
3*.3
6***
.22*
.43*
**-.
27**
Con
scie
ntio
usne
ss3.
60.5
9.3
5.2
4**
.14
-.09
Em
otio
nal S
tabi
lity
3.38
.75
.25*
*.1
2.2
4**
.17
-.20
*
Inte
llect
ance
3.32
.64
.20*
.32
.40*
**.2
4*-.
10
Perc
eivi
ng E
mot
ions
.48
.15
-.06
-.20
*-.
09-.
07-.
37**
*
Faci
litat
ing
Em
otio
ns.4
3.1
0.1
6.1
2.2
4*.0
8-.
46**
*
Und
erst
andi
ng E
mot
ions
.53
.12
.05
.12
.21*
-.06
-.44
***
Man
agin
g E
mot
ions
.38
.09
-.01
-.04
.03
-.13
-.33
***
Tot
al E
l.4
5.0
8.0
3-.
02.1
2-.
06-.
55**
*
Gen
der
1.63
.49
.03
-.12
.31
.16
-.22
*
Age
36.3
78.
14.1
9*.1
7-.
12**
*.1
0-.
19*
Eth
nici
ty.8
8.3
3-.
10-.
24*
-.22
*-.
02.1
8
*p <
.05;
**
p <
.01;
***
p <
.001
Coe
ffic
ient
alp
has
indi
catin
g sc
ale
relia
bilit
ies
are
in p
aren
thes
es.
**.
RSB
= R
elat
ions
hip
Supp
ortiv
e B
ehav
iors
; RD
B =
Rel
atio
nshi
p D
isru
ptiv
eB
ehav
iors
; El =
Em
otio
nal I
ntel
ligen
ce
Gen
der:
1 =
Mal
e, 2
= F
emal
e; E
thni
city
: 1 =
Asi
an, 2
= C
auca
sian
(.80
).5
2***
(.82
)
67
8
(.78
).4
1***
(.66
).2
1*.2
0*(.
69)
.31*
**.1
5.1
9*
.46*
**.2
3*.1
4.0
6.0
7-.
04.2
4**
.20*
.07
.10
.12
.00
.11
.13
-.02
.16
.17
-.00
.02
.23*
-.06
.15
.15
.18*
-.07
-.06
.07
12
Tab
le 5
.3: D
escr
iptiv
e St
atis
tics,
Cor
rela
tions
and
Cro
nbac
h's
Alp
has
(Con
t'd)
4.*
***
*13
< .0
5;p
<.0
1;p<
.001
Coe
ffic
ient
alp
has
indi
catin
g sc
ale
relia
bilit
ies
are
in p
aren
thes
es
Var
iabl
es1.
Tas
k Pe
rfor
man
ce2.
Con
text
ual P
erfo
rman
ce3.
Inn
ovat
ive
Perf
orm
ance
4. R
SB5.
RD
B6.
Ext
rave
rsio
n7.
Agr
eeab
lene
ss8.
Con
scie
ntio
usne
ss9.
Em
otio
nal S
tabi
lity
10. I
ntel
lect
ance
11. P
erce
ivin
g E
mot
ions
12. F
acili
tatin
g E
mot
ions
13. U
nder
stan
ding
Em
otio
ns14
. Man
agin
g E
mot
ions
15. T
otal
El
16. G
ende
r17
. Age
18. E
thni
city
9
(.85
).1
6-.
01 .18
.15
.13
.14
-.20
*.3
7***
-.02
10 (.78
).0
3.2
0*.3
0.1
3.2
2*-.
21*
.27*
*
-.43
***
11 (.90
).3
2
.33
.70*
**
.23*
-.08
-.01
12 .54
.75*
**
.16
.33*
**
-.10
13 .40*
**
.73*
**
.03
.36*
**
-.46
***
14 (.69
).6
9***
.06
.19*
-.23
*
15 (.90
).1
8.2
5**
-.27
**
16
-- -.23
*.1
5
17 -.36
***
Tab
le 5
.4: R
egre
ssio
n R
esul
ts f
or th
e D
irec
t Eff
ects
of
Pers
onal
ity a
nd E
l Abi
litie
s on
Per
form
ance
< .0
5<
.01
p <
.001
100
Tas
kC
onte
xtua
l
Stan
dard
ized
Reg
ress
ion
Coe
ffic
ient
sR
elat
ions
hip
Supp
ortiv
eR
elat
ions
hip
Dis
rupt
ive
Inno
vativ
eO
vera
llPr
edic
tors
Perf
orm
ance
Perf
orm
ance
Perf
orm
ance
Perf
orm
ance
Beh
avio
rsB
ehav
iors
Con
trol
sG
ende
r.1
0-.
12-.
03-.
02.1
6-.
14
Age
.01
-.05
.08
.02
.05
.03
Eth
nici
ty-.
16-.
21-.
10-.
18-.
07.0
9
Pers
onal
ity.0
7.2
0.1
2.1
535
***
.34*
**E
xtra
vers
ion
Agr
eeab
lene
ssC
onsc
ient
ious
ness
.07
.28*
*
.28*
*
.08
.05
.12
.15
.19*
.25*
*
-.01
-.22
**-.
08
Em
otio
nal S
tabi
lity
Inte
llect
ance
.18
.05
-.03 .0
6.0
8.2
0.0
9.1
3.0
7.0
6-.
22*
-.04
El B
ranc
hes
Perc
eivi
ng-.
09-.
22*
-.12
-.17
-.07
-.22
*
Faci
litat
ing
.16
.14
.17
.19
.06
-.24
*
Und
erst
andi
ng-.
12-.
00.0
1-.
04-.
15-.
17
Man
agin
g-.
09-.
11-.
10-.
12-.
20*
-.04
F2.
56**
*3.
96**
*3.
23**
*4.
26**
*4.
96**
*6.
97**
*
R2
.23
.32
.28
.33
.37
.45
Adj
. R2
.14
.24
.19
.26
.29
.39
df12
, 102
12, 1
0212
, 102
12, 1
0212
, 102
12, 1
02
Tab
le 5
.5: R
egre
ssio
n R
esul
ts f
or th
e M
oder
ator
Eff
ects
of
Em
otio
nal I
ntel
ligen
ce
101
Pred
icto
rsT
ask
Perf
orm
ance
Con
text
ual
Perf
orm
ance
Stan
dard
ized
Reg
ress
ion
Coe
ffic
ient
sIn
nova
tive
Ove
rall
Perf
orm
ance
Perf
orm
ance
RSB
sR
DB
s
Step
12
12
12
12
12
12
Con
trol
Var
iabl
esG
ende
r.1
1.1
0-.
11-.
11-.
02-.
03-.
01-.
02.1
8:.1
6-.
15:
-.15
:
Age
.05
.04
.02
.01
.14
.13
.09
.08
.06
.02
.01
.04
Eth
nici
ty-.
10-.
07-.
20:
-.20
-.07
-.10
-.14
-.15
-.02
-.02
.06
.07
Pers
onal
ity.0
9.0
7.0
8.0
4.2
1*
.28*
*
.19
.24*
.13
.05
.08
-.00
.17:
.15
.14
.11
.36*
**
.25*
*
.37*
**
.25*
*
.33*
**
-.22
*.3
0**
-.21
*E
xtra
vers
ion
Agr
eeab
lene
ssE
mot
iona
l Sta
bilit
yC
onsc
ient
ious
ness
.28*
*
.17:
.33*
**
.13
.08
-.03
.14
-.03
.13
.08
.21*
.06
.19*
.09
.26*
*
.06
-.01 .0
6.0
1.1
1
-.08
-.21
*-.
10-.
30**
*
Inte
llect
ance
.06
.08
.08
.11
.22*
.26*
.15
.19:
.07
.06
-.05
-.05
Em
otio
nal I
ntel
ligen
ce-.
09-.
49*
-.16
:-.
35-.
03-.
44*
-.11
-.51
**-.
24**
-.20
-.49
***
-.84
***
Tot
al E
lIn
tera
ctio
n T
erm
sE
xtra
vers
ion
X E
l-.
02.0
1-.
04-.
02-.
15.0
5
Agr
eeab
lene
ss X
El
.52*
.47*
.73*
**.6
9***
.12
.14
Con
scie
ntio
usne
ss X
El
-.09
-.20
-.19
-.19
-.13
.22*
Em
otio
nal S
tabi
lity
X E
l.0
5-.
04.0
5.0
3-.
04.1
8*
Inte
llect
ance
X E
lF
3.11
**
.01
2.67
**4.
47-.
093.
68**
*3.
78**
*
-.21
:4.
22**
*4.
88**
*
-.13
4.83
***
6.27
***
-.01
4.30
***
9.18
***
-.04
7.21
***
R2
.21
.27
.28
.34
.24
.37
.29
.40
.35
.38
.44
.50
Adj
. R2
.14
.17
.22
.25
.18
.28
.23
.32
.29
.29
.39
.43
df9,
105
14,1
009,
105
14,0
09,
105
14,0
09,
105
14,1
009,
105
14,1
009,
105
14,1
00
P <
.10
*p<
.05
**p
< .0
1p
< .0
01
Figure 5.1: Interaction Effects of Emotional Intelligence and Agreeableness on TaskPerformance
2 3
Agreeableness
Figure 5.2: Interaction Effects of Emotional Intelligence and Agreeableness onContextual Performance
4-3.5
3
2.5
2a)"
0- 1.5-c-t
1
t' 0.5
0
-0.5
-1
3
Agreeableness
102
4
4
5
5 EmotionalIntelligence (H)
0 EmotionalIntelligence (M)
EmotionalIntelligence (L)
4 EmotionalIntelligence (H)
0 EmotionalIntelligence (M)
EmotionalIntelligence (L)
Figure 5.3: Interaction Effects of Emotional Intelligence and Agreeableness onInnovation Performance
4
3
3
0
2EmotionalIntelligence (H)
1
Intellectance
4 5
103
EmotionalIntelligence (H)
0 EmotionalIntelligence (M)
1
Emotional
0 Intelligence (M)
EmotionalIntelligence (L)
-1 Emotional
AgreeablenessIntelligence (L)
Figure 5.4: Interaction Effects of Emotional Intelligence and Openness to Experienceon Innovation Performance
4
1 2 0
Figure 5.5: Interaction Effects of Emotional Intelligence and Agreeableness onOverall Performance
4
2
2
0
-1
Agreeableness
2
Conscientiousness
104
-4- EmotionalIntelligence (H)
EmotionalIntelligence (M)
- - EmotionalIntelligence (L)
Figure 5.6: Interaction Effects of Emotional Intelligence and Conscientiousness onRelationship Disruptive Behaviors
EmotionalIntelligence (H)
-E- EmotionalIntelligence (M)
1 3 4 EmotionalIntelligence (L)A,..
Figure 5.7: Interaction Effects of Emotional Intelligence and Emotional Stability onRelationship Disruptive Behaviors
-- Emotional Intelligence(H)
2
-3
Emotional Stability
105
4 5
Emotional Intelligence(M)
Emotional Intelligence(L)
CHAPTER 6
DISCUSSION
In this chapter, I discuss the central findings of the study. First, I review the
key research findings in respect to the direct effects of personality and El on
workplace performance dimensions of task, contextual, innovative as well as
relationship supportive behaviors and relationship disruptive behaviors. Then I
elaborate on the interactive effects of El and personality on the workplace
performance variables. Next, I discuss the limitations and strengths of the research
and offer suggestions for future research paths. Finally, I present the contributions
that this research makes to the literature and its benefits to practitioners.
Key Research Findings
The purpose of this research was to develop and test a socioanalytic model of
workplace performance by synthesizing three literatures: personality, El and
performance. In the process, I proposed and examined, as have many earlier
researchers, the direct effects that the Big Five personality factors have on workplace
performance. In this regard, my findings are fairly consistent with those of the
literature.
Direct effects of Personality on Performance
As expected, the study confirmed previous research findings in respect to task
and contextual performance. Addressing each of the Big Five factors in turn, I will
review the key findings and discuss unanticipated results.
Extraversion. Extraversion was a strong, significant predictor of relationship
supportive behaviors. Individuals high in social abilities are more sensitive to the
ongoing maintenance of their groups as compared to introverted personalities.
Although not predicted, extraversion was also a significant predictor of relationship
106
disruptive behaviors. Insofar as the zero-order correlation between these two
variables was not significant, it is likely that this is a statistical artifact. While no
other significant relationships with dependent variables were noted from the
regression analyses, the picture is somewhat different in respect to the zero-order
correlations. Here, extraversion was significantly correlated with task performance,
contextual performance, innovative performance and relationship supportive
behaviors.
Extraverts are generally considered to be more social, enthusiastic, and active,
with a desire to be "part of the action" (Barrick & Mount, 1991a). Hogan (1986)
considered extraversion to consist of two parts. One part is Ambition which includes
such traits as initiative, surgency, ambition, and impetuous. The other component is
Sociability which includes traits of being sociable, exhibitionist, and expressive.
Following this thinking, those characteristics of extraversion under Ambition, would
likely influence task and innovative performance; those under Sociability would be
played out in the relationship performance areas. Further, it could be argued that an
extravert will be more likely to work toward maintaining and facilitating a team for
the purpose of task completion, in part by performing supportive behaviors. Future
research targeting the more specific facets of extraversion would help to clarify its
effects on performance.
Agreeableness. Agreeableness was a positive significant predictor of
contextual performance and relationship supportive behaviors and a negative
predictor of relationship disruptive behaviors. Although no direct effects were found
for task performance or innovative performance, at the zero-order correlation level,
agreeableness correlated significantly with all performance variables. This suggests
support for the view that individuals who have a disposition to accommodate to others
107
and who are team players with a service orientation will perform better, across the
board, than those who do not.
Barrick and Mount (1991a) report that traits associated with this dimension
1995), workplace incivility (Andersson & Pearson, 1999), and interpersonal conflict
(Spector & Jex, 1998). Sin et al. (Sin et al., 2004) demonstrated the relationship
between RDB and outcome variables at the individual (e.g., job performance, social
network size) and organizational levels (e.g., firm performance,-employee loyalty).
Direct Effects of El Abilities on Performance
To provide a more specific examination of El on performance, I assessed the
effects of each of the four branches of El on the multiple dimensions of employee
performance. Below, I review the major findings for each branch.
Branch One Perceiving Emotions. In regression analyses, Branch One was a
significant negative predictor of contextual performance. This would seem to indicate
111
that individuals that are better in perceiving emotions in themselves and others are
less likely to perform well contextually in the workplace. Findings from studies of
emotional eavesdropping may offer an explanation for these results. These studies
(e.g., Elfenbein & Ambady, 2002a) report that workplace ratings from colleagues and
supervisors are higher for individuals that are better in their ability to "read" positive
emotions and lower for those with abilities to "read" negative emotions in others. It
could be conjectured that subjects in the present study who registered high in
emotional perception abilities find it difficult to perform well contextually. At
operation here may simply be a felt sense of vulnerability to others in the workplace
context with respect to being able to read negative emotions.
Regression analysis also revealed, as expected, a significant negative impact
of the ability to perceive emotions on relationship disruptive behaviors. This
indicates that those high in emotion detection abilities would be predisposed, through
their awareness of their own emotions and those of others, to process this information.
This "intake of more information" may be conducive to creating a higher level of
empathy for others. Although in the present research greater ability to perceive
emotions did not result in more supportive behaviors, it did increase the likelihood
that managers would engage in less disruptive or negative behaviors in the workplace.
Branch Two Facilitating Thought About Emotions. Similar to findings for
Branch 1, regression analyses revealed a significant negative relationship between
facilitating abilities and relationship disruptive behaviors. This would suggest that if
an individual, after perceiving emotions, actually engages in cognitive processing
about them, the resulting behavior is less likely to result in disruptive behaviors.
Interestingly, perceiving emotions was positively associated with innovative
performance at the zero-order correlation level. One possible explanation may
112
involve the acuity skills that are common to both Branch One and innovativeness. For
example, just as an individual must be able to scan the face, gestures and voice of
others to see the emotions display, an innovative person is served well by being able
to scan the environment to select meaningful information that may facilitate the
process of creativity.
Branch Three Understanding Emotions. While regression analysis indicated
no significant relationships with the performance measures, two significant
correlations did emerge. First, understanding of emotions correlated with innovative
performance. It may be conjectured that this reflects a cognitive ability for processing
information that is common to making sense of emotions and the creative process.
The second significant correlation was with respect to relationship disruptive
behaviors. Consistent with the first two branches, a negative correlation between
understanding and relationship disruptive behaviors was reasonably high. Following
the same reasoning to explain the previous branches' relationships with this
performance dimension, it may be argued that by simply understanding emotions of
self and others the need or desire to engage in disruptive behavior is mitigated.
Branch Four Managing Emotions. Interestingly, unlike the other branches,
the regression analysis showed no significant relationship with relationship disruptive
behaviors, although managing emotions was a strong, negative correlate of this
performance dimension. However, a significant negative relationship between this
branch and relationship supportive behaviors was recorded. This seems to indicate
that an individual that can regulate the emotions of others and themselves will have a
decreased tendency to engage in relationship supportive behaviors.
The manifestation of this results in the personality characteristic of
Machiavellianism (Mach), which has been defined as "the degree to which an
113
individual is pragmatic, maintains emotional distance, and believes that ends can
justify means" (Robbins, 1991, p. 724). Early research showed that high-Machs were
more manipulative, more apt to win, less susceptible to persuasion and more likely to
persuade others (Vleeming, 1979). Further, research as to how this personality trait is
moderated by situational factors has shown that high-Machs do well in direct
interpersonal interactions especially when the social context has few rules and
regulations that allows them to improvise and when low-Machs may be distracted
with the details of emotional involvement irrelevant to winning (see Christie and
Geis, 1970 for a review). In essence, people who can control their own emotions
more readily can use them strategicallypressing the emotional buttons to serve
themselves but not others.
Total El Ability. Consistent with branch-level findings, total El abilities
registered a very strong correlation with relationship disruptive behaviors. This may
indicate the power of emotional intelligence in regulating negative behavior. That is,
an individual with high El abilities may be more aware of what behaviors will disrupt
social bonds and be quite good at not committing them. On the other hand, no
evidence emerged that high El abilities are manifested in positive behaviors (e.g.,
contextual performance or relationship supportive behaviors). This may be reflective
of passivity in respect to extra-familial relationships in the context of Hong Kong (Lo,
2003; Lo, Stone, & Ng, 2003). It could be speculated that to avoid engaging in
relationship disruptive behaviors requires no extra effort beyond knowing the rules of
the social context. However, to engage in relationship supportive behaviors requires
an extra degree of emotional labor and emotional risk taking, especially when it is
extra-familial.
114
In summary, with respect to the El abilities and performance dimensions, El
had no significant effects on task performance or innovative performance. However,
it did have significant effects on the social forms of performance. This was especially
apparent in the zero-order correlations where all four branches of the El abilities test
showed significant correlations with relationship disruptive behaviors. Another
interesting observation is that Branches Two and Three correlated positively with
innovative performance. Thus, facilitating thought about emotions and understanding
emotions was highly correlated with innovative performance measures, indicating that
thinking about emotional information and understanding what the information means
in a given context may share a common dynamic with the creative thought process.
Interactive effects of El and Personality on Performance
A major objective set out for this study was to examine what, if any,
interactive effects exist between the independent variables of personality and El in
respect to workplace performance. The regression analyses supported the supposition
that there are several interactive effects involving the personality traits of
agreeableness, intellectance, conscientiousness, and emotional stability. I will discuss
each in turn.
Agreeableness El Interaction. Agreeableness was involved in three
significant interactions with El. Together, El and agreeableness predicted task,
contextual and innovative performance. Agreeable individuals with high levels of El
displayed higher task performance than those with medium or low El abilities (see Fig
5.1). This may indicate that an individual possessing a personality trait that
predisposes them to get along with others, such as team players, are even more
effective in task roles when they also possess high El abilities. The relationship
between agreeableness and contextual performance was also stronger for individuals
115
with high levels of El than for those with medium or low El abilities (see Fig 5.2).
Here again, this may indicate that people with high agreeableness traits will leverage
them when they also have high abilities of El. This may indicate that a highly
agreeable, emotionally intelligent individual will be a better facilitator of
relationships. Finally, El also enhanced the effects of agreeableness on innovative
performance (see Fig 5.3). This would seem to suggest that an employee with high El
abilities and high agreeableness tends to contribute more to the job than only getting
the job done and getting along with others. For example, their increased innovative
performance level may include contributing insights and perspectives to a greater
extent than those with low El abilities.
IntellectanceEl Interaction. There was a significant interaction involving
intellectance and El with respect to innovative performance, but the direction was
counter to what might have been expected (see Fig 5.4). Individuals with high levels
of intellectance displayed lower innovative performance if they also had high levels of
El. This may suggest that El has a suppressing effect on the relationship between
intellectance and innovative performance. Perhaps the cognitive vent to being highly
sensitive to the emotions of others and self detracts from the expansive perspectives
that seem to characterize an individual with a high degree of openness and curiosity.
ConscientiousnessEl Interaction. Individuals who reported both extremely
low levels of conscientiousness and extremely low levels of El abilities were more
likely to engage in relationship disruptive behaviors (see Fig. 5.5). A possible
explanation is that low conscientiousness and low El abilities may reflect a common
predisposition factor. For example, an individual with low conscientiousness may be
less "engaged" with the job, while an individual with low El may have less interest in
the relationship aspects of the job. Combined, these qualities may reflect an
116
individual that is disconnected from the task and the social context of the job.
Consequently, their behaviors on the job are more likely to be disruptive.
A look at the other end of the continuum presents a different picture. Those
individuals with high levels of conscientiousness and high levels of El abilities also
engaged in high levels of relationship disruptive behaviors. While this may be
startling, it may reflect a fairly intuitive truth about personality types. High
conscientiousness qualities could include being focused, organized, perfectionist and
ambitious (Howard & Howard, 2001). Further, those with high El abilities are
considered to be superior in terms of perceiving and managing the emotions of others
and themselves. The combination of these two sets of qualities does not necessitate
that the resultant behavior should be entirely socially acceptable. For example, one
can think of sociopathic behaviors that draw upon qualities of both conscientiousness
and El such as a confidence man who needs to be very focused and organized in terms
of playing his game while he monitors and manipulates emotional states. This also
illustrates the "value neutral" nature of El abilities and personality factors.
Emotional Stability El Interaction. The interaction between emotional
stability and El parallel that of the conscientiousness El interactions in regard to
relationship disruptive behaviors (see Fig 5.6). That is, individuals who displayed
low levels of emotional stability and El were more likely to engage in relationship
disruptive behaviors than those with moderate levels of each. Likewise, those with
high levels of emotional stability and El also registered a higher predisposition to
perform relationship disruptive behaviors. Further, those registering high levels of
emotional stability and low levels of El engaged in much less relationship disruptive
behaviors than those with high emotional stability and high El. This may add support
to the earlier observation that a neurotic person, who reflects more anxiety, stress and
117
with a reactive nature and who is not equipped in respect to managing his or her
emotions, will be prone to disruptive social behavior. At the same time, those with
high emotional stability and high El may reflect a personality consistent with the
profile of a sociopath.
Limitations of the Research
All research has its limitations, and this is no exception. The findings reported
here need to be considered against the backdrop of the study's limitations. One
limitation has to do with generalizability. As the data collection was confined to
English speaking Hong Kong managers who were currently or formerly engaged in
executive development programs, a caveat in respect to the applicability of its results
to other Hong Kong mangers or managers in other cultural settings is required. First,
as no control group was established, it is not possible to determine what (if any)
influence the respondents' participation in executive developmental programs may
have had on their responses. It is possible that these types of programs work to
making managers more sensitive to some of the research issues than would be the
case with managers without such training.
While Hong Kong is considered to be cosmopolitan and one of the world's
most attractive environments for conducting business, it has many unique qualities
that set it apart. One such quality is the blend of languages, cultures and races that
one generally experiences in the workplace. While the data collection instruments
were written in English, some of the subjects' English level proficiency was probably
less than that of a native speaker.
Hong Kong's business culture is quite similar to major cities in North
America and Europe. However, the influence that some artifact of Hong Kong's
national culture such as Confucianism, specifically its precepts in respect to
118
interpersonal relationships, should be considered. As the primary data collection
instruments were of Western origin, the results may have been mitigated by this
cultural difference.
Data collection was done on-line, which may be considered a strength of the
study because it allowed for economies of time on the part of the participants and the
researcher. However, it may also constitute a limitation in that it presupposed
respondent computer and Internet literacy. While Hong Kong businesses are
approaching the saturation point for this relatively new technology, the method of
data collection may have posed a self-selection hurdle that many did not jump.
Another possible limitation of the study was its sample size. While an attempt
was made to gather a more robust sample, only 116 subjects completed both on-line
instruments. While I had hypothesized 50 effects, only 11 (about 22%) proved to be
significant. That is, only 1 of 5 predictions of El's effects on performance; and 5 out
of 25 predicted interactions proved significant. The small sample size may have made
for a low statistical power resulting in the large number of insignificant results.
Further, while an attempt was made to gather performance measures from the
respondents' workplace supervisors by asking the respondents to "nominate" them,
few complied. Again, this very request may have intimidated many potential
subjects. Thus, participant self-selection may have resulted in a sampling that was
less representative of Hong Kong's general business environment than was desired.
The reliance on single source data collection measures raises questions about
common method bias (Podsakoff & Organ, 1986). An additional concern is the
accuracy of respondents' perceptions and their willingness to respond honestly.
While these concerns were partially address by use of recognized, previously
l"
119
validated instruments and the incorporation of a test of ability as opposed to self-
reports of El, social desirability remains as a possible limitation.
Another limitation is the fact that this was a cross-sectional study. Although
data triangulation was achieved by having participants complete two separate
instruments (a survey and a test) at two separate times, all of the data is from the same
respondent, and no repeat measures were taken. A longitudinal study would serve to
increase the internal validity of the study and help to establish causal relationships.
A final limitation has to do with the exploratory nature of the moderating
effects of Emotional Intelligence. Under such conditions, a Bonferroni correction
should be done. This is an adjustment of the critical values of the significance tests
when multiple comparisons are being made. Adopting a .05 level of significance, and
conducting 20 different moderator tests, the significant level needs to be .05/20 or
.0025. Two of the interactions reached this level of significance: emotional
intelligence interacted with agreeableness to influence both innovative and overall
performance. Recognizing that this correction can cause a substantial loss in the
precision of my findings, however, I reported all interactions that reached traditional
levels of significance. These findings are intended to inform future researchers who
may want to confirm such relationships.
Strengths of the Research
Despite several limitations, this research has many strong points. One is the
use of a standardized El test that has been well-validated with good internal
consistency indicators. As mentioned earlier, the use of web based survey
instruments facilitated data collection and data entry. Steps to safeguard
confidentiality were taken seriously. Finally, the diversity of subjects in terms of
gender, age, ethnicity, education, managerial experience, management level, years of
120
tenure, and industries represented resulted in a heterogeneous sample that allows more
confidence in generalizing the findings.
Contributions to the Literature
The literature is replete with research that has investigated the relationship of
personality with various dimensions of performance. However, none has investigated
the interplay of some of the important antecedents of performance in a manner that
this study did. Three contributions to the literature are most significant. First, the
study illuminated the joint roles that El and personality have on workplace
performance. Second, the criterion space of performance was expanded beyond the
traditional focus on task performance and contextual performance to encompass
innovative performance and relationship supportive and disruptive behaviors. Third,
the study incorporated an explicit test of socioanalytic theory by conceptualizing El as
a social skill that interacts with personality to predict performance.
Other findings of note include the use of the El abilities test in an Asian
culture. As this is the first empirical study of the MSCEIT V2 in the Hong Kong
context it will contribute by bringing a better understanding of its validity outside
Western cultures. The research results also contribute to personality and El literatures
by clarifying the distinct nature of El abilities. These findings help inform the
ongoing debate as to whether El is independent or simply measuring the same
individual traits as personality. The two constructs are distinct. By testing for direct
effects of both independent variables on performance, differences were revealed in
respect to their differential influences on task, contextual, and innovative performance
as well as the newer constructs of relationship supportive and relationship destructive
behaviors.
121
In this study, the proposed relationships were all based on sound theoretical
arguments. As predicted by socioanalytic theory (Hogan & Roberts, 2000; Hogan &
Shelton, 1998a), El moderated the personality performance relationship. Although
in general El had enabling effects, some suppressor effects also occurred. More
research in this area is needed to clarify the dynamic interactive effects.
Implications of the Results for Organizations
The desire to develop a good theory that is practical was one of the prime
objectives of this research. In many respects I believe a foundation has been built to
achieve this goal. This stream of research can address many of the primary concerns
of human resource practitioners, especially in the areas of selection and assignment.
Selection
Personal selection has been defined as a process of measurement, decision-
making and evaluation with a goal to hire individuals that will do well on the job.
(Fisher et al., 2003, p. 283). Because of today's rapidly expanding, competitive global
economy, corporations find that they are more frequently engaged in a talent war for
the best managers and leaders. Selection and promotion decision are becoming
increasingly critical to the success of a company.
A recent study reported in Hong Kong's popular press highlighted the
important role selection and training can play in the competition between regional and
national economies (Lo, 2004). According to the British-based manpower assessment
company SHL, managers in Hong Kong spend twenty percent of their time correcting
their employees' mistakes. The $HK39b a year that this costs, is nearly equivalent to
Hong Kong entire budget for education. In contrast, the six other economies studied
reported significant (but lower) figures, i.e., India 16 percent, the US 13 percent,
Australia 12 percent, the Netherlands 11 percent, Britain 9 percent and in Sweden it
122
was 7 percent. If workforce productivity is regarded as a significant competitive
element in the new global economy, Hong Kong is certainly lagging behind. The
study further reported that twenty-five percent of Hong Kong employees quit their
jobs before they reach a competent level. The researchers suggested that the root of
these very costly inefficiencies could be traced back to a 'fire-and-hire attitude' on the
part of Hong Kong employers and advocated the use of better assessment tools to aid
decision-making in the selection process.
After several decades of abandonment as a primary selection device,
personality measures are becoming increasingly popular. What seems to have
fostered this new interest is an increased sense of their predictive validity relative to
workplace performance (Fisher et al., 2003). Increasingly, Human Resource
professionals are turning to a selection instruments that go beyond measuring task
performance (e.g., Hough & Oswald, 2000).
The importance of the selection process to an organization's bottom line is
highlighted by research of the costs involved. Spencer (2001) notes that staffing adds
value in demonstrable ways. By hiring, placing and promoting higher performers
organizational goals are more readily achieved. The costs associated with poor
selection decisions are significant. For example, an analysis of lost productivity
revealed that on average, it takes between 55 to 57 days or about two months of sales
or production costs to fill a management position. The author estimates this to be
about one third of an employee's first-year salary. Further, he contends that it takes a
new hire on average about twelve months for a new hire to become proficient on his
or her job. If direct costs incurred for relocation and training of technical or
professional staff are added, the total cost can be as much as two to three times their
direct salary (McClelland, 1998; Spencer, 1986; Spencer & Spencer, 1993). Thus it is
123
evident that the stakes are high in the search for assessment instruments that offer
better performance predictability.
All three of the selection process stages (measurement, decision-making,
evaluation) offered by Fisher et al., (2003) can be aided by a better understanding of a
potential employee's personality traits and El abilities. In choosing selection
measurement instruments, care must be given as to their reliability and validity. The
present study, testing the reliability and validity of an El abilities test in a new cultural
context has added to the ongoing search for more effective tools that can predict
performance with mangers in Asia. Evidence was obtained that may aid the Human
Resource professional in choosing viable measures for selecting employees. For
example, while no evidence was found of a direct link between El and task
performance, evidence was revealed as to the power of El to predict contextual
performance. Additionally, by shedding light on the joint roles that El and personality
have on workplace performance, the instruments that measure these abilities and traits
can be better evaluated.
Selection decision-making needs to incorporate a balanced approach as to the
relative value of the instruments used such that the information collected about an
individual leads to a successful hiring decision. The present study offers evidence that
the predictive value of performance gained by an El abilities test is more in line with
Mayer (1999) who suggested that El may be able to predict important life outcomes
but probably at about the same level of other personality variables. This more
measured view refutes popular press accounts, which advance that El is twice as
important as GMA in its predictive powers (i.e., Goleman, 1995). Thus while, El
measurement can be considered to have marginal utility, it should be used in
conjunction with other instruments to aid selection decisions. This study illustrates the
124
promise of El abilities tests, especially when used in conjunction with Big Five
personality measures, as an effective assessment tool.
Job Assignment
Reporting on a recent study of work-life balance in Hong Kong, Michael
(2004) outlined some of the personal and organizational consequences in relationship
to poor job assignment decisions. For example, one of the outcomes of a more
balanced life is to allow people to find fulfillment inside their work situation. A poor
fit between the employee and the job undermines the achievement of this goal with a
resultant dissatisfaction on the part of the employee and lower productivity for the
organization. According to the study, only five out of 1,000 people polled reported
that that they were very happy in their work, seventy-five percent suffered from
stress, thirty-three percent reported being depressed by their job and twenty-eight
percent were planning to quit in the next twelve months.
This illustrates why assigning the right person to the right job is an important
goal for most Human Resource managers. The merits of a proper "job fit" may
extend to a wide range of outcomes. Across all levels of a firm (organizational, group
and individual) such outcomes as performance, job satisfaction, turnover,
absenteeism, team functioning and customer service all stand to improve through the
proper match of the employee to the position. Better job satisfaction and lower
turnover (through better job fit analysis) results from better staffing decisions.
Spencer (2001) illustrated this in a meta-analysis of the effects of eight selection
systems that incorporated emotional intelligence factors. The effects of including El
competencies in the selection were: median productivity improved 19%, median
turnover decreased 63 percent, median economic value added was $1.6 million and
125
the median return on investment was over 1,000 percent. Placing the right employee
in the right jobs has dramatic consequences in respect to productivity increases.
However, the decision to use El assessment instruments should take into
account their purpose and relevance for specific occupational settings under
consideration. Matthews et al.(2002) recommend their judicious use, suggesting that a
distinction should be made between occupations where emotional skills are relevant
to successful job performance and those in which it may not be as important. For
example, helping and service professions such as clergy, teachers, sales,
psychotherapist would seem to require higher skills to do their jobs than mechanical
engineers, software programmers and brain surgeons. In essence, the task of filling
jobs that require more social and emotional involvement may be made more effective
by measuring the candidates' El.
A major implication of this study is that by including measures of El ability,
along with other measurements such as the Big Five, decision-making in respect to
employee assignments and job design can be improved. Additionally, having a
clearer picture of employee personality and El abilities can be of practical use in the
structuring of groups and work teams. This study's findings suggest that team
composition variables can, and should, include other factors than demographics such
age, gender, education and position. By considering the personality and emotional
qualities of potential team members before organizing a team, the likelihood of
building teams with compatible members is increased. Such teams are more likely to
be high performing and experience less negative conflict (Druskat & Wolff, 2001).
Suggestions for Future Research
Many future research opportunities are open in respect to El abilities. As a
relatively new, multi-dimensional construct, continued effort should be placed on
126
validating instruments. While this study illustrates the promise of El abilities tests,
especially when used in conjunction with Big Five personality measures, as an
effective assessment tool much remains untested. Based on the results of this study,
more research to clarify the interactive effects of individual traits on workplace
performance is warranted. The wider perspective that this offers can help improve the
selection, assignment and job design processes of organizations. Additionally, more
research is in order to test for the malleability of El abilities. Is it relatively static
after a certain age or can it be increased through training as some suggest? More
comprehensive, longitudinal studies in respect to coaching may result in significant
insights in this regard.
Demographic differences in the Hong Kong workplace in respect to El abilities
would also be of great interest to explore. Because of significant differences involving
ethnicity, gender and age with other constructs in the present study's model, I
controlled for these demographic factors in the regressions. Additionally, subjects
reported being employed across 15 different industries. Grouped under
manufacturing, government and services; 7 subjects were in manufacturing, 35 were
in government and 76 were employed in services. Analysis revealed no significant
differences as a result of their respective industries. While these various demographic
variables were not of theoretical interest in the present work, I recommend that future
researchers consider looking at their possible influence further.
Further research in respect to the individual El abilities and team composition
is also needed. What is the most advantageous grouping of individuals based on
individual El abilities for raising the level of group performance? This becomes an
increasingly important issue as more organizations rely on teams as a major way of
structuring work. In conjunction with this, more concerted investigation of how the
127
dynamics of social interaction in general are influenced by individual El abilities may
be rewarding.
Another potentially valuable stream of research would be investigations into
the cross-cultural differences that may exist relative to El abilities. Examining what
commonalities may exist between El abilities of managers coming from different
cultures could aid our understanding in such areas as international human resource
management, expatriate selection and assignment procedures, and even international
business negotiations.
Conclusion
The three-fold objectives of this study have been accomplished to varying
degrees. First, an examination of the direct effects of El and personality on task
performance, contextual performance and innovative performance was accomplished
with significant, mixed results. The inclusion of relationship supportive and disruptive
behaviors as more interpersonal forms of performance yielded important findings,
especially regarding the relationship between El and the darker side of employee
actions. Secondly, the exploration of the interaction effects between El abilities and a
range of workplace performance variables was completed, with significant findings
surfacing. Lastly, the results set a base for discussing the implications of El and
performance on important areas of human resource management.
I conclude the study with a degree of amazement. While its objectives were
accomplished and answers to the research questions I initially proposed were
generally enlightening and gratifying, many more questions were raised in the
process. This may be common to all research, i.e., it raises more questions than it
answers. I end with a favorite quotation by George Bernard Shaw that has sustained
me and that embodies the spirit of this endeavor: "All progress is initiated by
128
challenging current conceptions". I hope that in some small way this work has
challenged the traditional view of workplace performance and added to the
understanding of the complexities of how the head and the heart work together to
determine employee performance.
129
REFERENCES
Aaker, D. A., Kumar, V., & Day, G. S. 1995. Marketing Research. New York, NY:John Wiley & Sons, Inc.
Adler, P. S. & Kwon, S. W. 2002. Social capital: Prospects for a new concept.Academy of Management Review, 27: 17-40.
Adler, S. 1996. Personality and work behavior: Exploring the linkages. AppliedPsychology: An International Review, 45: 207-214.
Aiken, L. S. & West, S. G. 1991. Multiple regression: Testing and interpretinginteractions. Thousand Oaks, California: Sage Publications.
Andersson, L. M. & Pearson, C. M. 1999. Tit for tat? The spiraling effect of incivilityin the workplace. Academy of Management Review, 24: 452-471.
Argyle, M. 1969. Social Interaction. Chicago: Aldine.
Ashforth, B. E. & Humphrey, R. H. 1993. Emotional labor in service roles: Theinfluence of identity. Academy of Management Review, 18: 88-115.
Ashforth, B. E. & Humphrey, R. H. 1995. Emotion in the workplace: A reappraisal.Human Relations, 48(2): 97-125.
Austin, J. T. & Villanova, P. 1992. The criterion problem: 1917-1992. Journal ofApplied Psychology, 77: 836-874.
Averill, J. R. & Nunley, E. P. 1992. Voyages of the heart: Living an emotionallycreative life. New York: Free Press.
Barchard, K. A. & Hakstian, A. R. 2004. The Nature and Measurement of EmotionalIntelligence Abilities: Basic Dimensions and Their Relationships With OtherCognitive Ability and Personality Variables. Educational and PsychologicalMeasurement, 64(3).
Bar-On, R. 1997b. Development of the Bar-On EQ-i: A measure of emotional andsocial intelligence. Chicago: The American Psychological Association.
Bar-On, R. 2000. Emotional and Social Intelligence: Insights from the EmotionalQuotient Inventory. In R. Bar-On & J. D. A. Parker (Eds.), The Handbook ofEmotional Intelligence: Theory, development, assessment, and application athome, school, and in the workplace.: 363-388. San Francisco: Jossey-Bass.
Bar-On, R. & Parker, J. D. A. (Eds.). 2000. The Handbook of Emotional Intelligence:Theory, development, assessment, and application at home, school, and in theworkplace. San Francisco, California: Jossey-Bass.
130
Barrick, M. R. & Mount, M. K. 1991a. The Big Five personality dimensions and jobperformance: A meta-analysis. Personnel Psychology, 44(1): 1-26.
Barrick, M. R. & Mount, M. K. 1991b. The big five personality dimensions and jobperformance: A meta-analysis. Personnel Psychology, 44: 1-26.
Barrick, M. R. & Mount, M. K. 1993. Autonomy as a moderator of the relationshipbetween the Big Five personality dimensions and job performance. Journal ofApplied Psychology, 78: 111-118.
Barrick, M. R., Stewart, G L., Neubert, M. J., & Mount, M. K. 1998. Relatingmember ability and personality to work-team processes and teameffectiveness. Journal of Applied Psychology, 83: 377-391.
Bass, B. M., Cascio, W. G, & O'Connor, E. J. 1974. Magnitude estimations offrequency and amount. Journal ofApplied Psychology, 59: 313-320.
Behling, 0. 1998. Employee selection: Will intelligence and conscientiousness do thejob? Academy of Management Executive, 12: 77-86.
Berry, D. S. & Sherman Hansen, J. 2000. Personality, nonverbal behavior, andinteraction quality in female dyads. Personality & Social Psychology Bulletin,26: 278-292.
Bies, R. J. & Moag, J. S. 1986. Interactional justice: Communication criteria forfairness. In B. Sheppard (Ed.), Research on negotiation in organizations, Vol.1: 43-55. Greenwich, CT: JAI Press, Inc.
Bobko, P. 1985. Removing assumptions of bipolarity: Towards variation andcircularity. Academy of Management Review, 10: 99-109.
Bommer, W. H., Johnson, J. L., Rich, G A., Podsakoff, P. M., & MacKenzie, S. B.1995. On the interchange ability of objective and subjective measures ofemployee performance. Personnel Psychology, 48: 587-605.
Borman, W. C., Motowidlo, S., Rose, S. R., & Hanser, L. M. 1985. Development of amodel of soldier effectiveness. Minneapolis, MN: Personnel DecisionsResearch Institute.
Borman, W. C., White, L. A., Pulakos, E. D., & Oppler, S. H. 1991. Models ofsupervisory performance ratings. Journal of Applied Psychology, 76: 863-872.
Borman, W. C. & Motowidlo, S. J. 1993. Expanding the criterion domain to includeelements of contextual performance. In N. Schmitt & W. C. Borman (Eds.),Personnel Selection in Organizations: 71-98. San Francisco: Jossey-Bass.
Borman, W. C. & Motowidlo, S. 1997. Task Performance and ContextualPerformance: The Meaning for Personnel Selection Research. HumanPerformance, 10(2): 99-109.
Boyatzis, E. E. 1982. The Competent Manager: A Model for Effective Performance.New York: John Wiley & Sons.
131
Boyatzis, R., Goleman, D., & Rhee, K. S. 2000. Clustering competence in emotionalintelligence: Insights from the Emotional Competence Inventory (ECI). In R.Bar-On & J. D. A. Parker (Eds.), The Handbook of Emotional Intelligence:Theory, development, assessment, and application at home, school, and in theworkplace.: 343-362. San Francisco: Jossy-Bass.
Brief, A. P. & Motowidlo, S. 1986. Prosocial organizational behaviors. Academy ofManagement Review, 11: 710-725.
Brown, C. S. & Sulzer-Azaroff, B. 1994. An assessment of the relationship betweencustomer satisfaction and service friendliness. Journal of OrganizationalBehavior Management, 14: 55-75.
Bruvold, N. T. & Corner, J. 1988. A model for estimating the response rate to a mailedsurvey. Journal of Business Research, 16(2): 101-116.
Buck, R. 1984. The communication of emotion. New York: The Guilford Press.
Campbell, J. P., McCloy, R. A., Oppler, S. H., & Sager, C. E. 1993. A theory ofperformance. In N. Schmitt & W. C. Borman (Eds.), Personnel selection inorganizations: 35-70. San Francisco: Jossey-Bass.
Cantor, N. & Kihlstrom, J. F. 1987. Personality and social intelligence. EnglewoodCliffs NJ: Prentice Hall.
Carson, R. C. 1969. Interaction concepts of personality. Chicago: Aldine.
Caruso, D. R. & Wolfe, C. J. 2001. Emotional Intelligence in the Workplace. In J.Ciarrochi & J. P. Forgas & J. D. Mayer (Eds.), Emotional Intelligence inEveryday Life.. A Scientific Inquiry: 150-167. New York: Psychology Press.
Caruso, D. R. & Wolfe, C. 2002. Emotionally Intelligent Certification WorkshopTraining Manual. Mashantucket, Connecticut.
Caruso, D. R. & Salovey, P. 2004. The Emotionally Intelligent Manager. SanFrancisco, CA: Jossey-Bass.
Chatman, J. A., Caldwell, D. F., & 0-Reilly, C. A. 1999. Managerial personality andperformance: A semi-idiographic approach. Journal of Research inPersonality, 33: 514-545.
Cherniss, C. & Goleman, D. (Eds.). 2001. The Emotionally Intelligent Workplace. SanFrancisco: Jossey-Bass.
Christianson DeMay, C. & Toquam, H. J. 2001. 360-degree assessment at the Boeingcompany: A dual- process (web/paper) approach. In M. H. (Chair) (Ed.), TheInternet and I/O psychology: Applications and issues.: Symposium conductedat the 16th Annual Conference of the Society for Industrial and OrganizationalPsychology.
Christie, R. & Geis, F.L. 1970. Studies in Machiavellianism. New York, NY:Academic Press.
132
Church, A. H. 2001. Is there a method to our madness? The impact of data collectionmethodology on organizational survey results. Personnel Psychology, 54: 937-969.
Ciarrochi, J., Chan, A., & Caputi, P. 2000. A critical evaluation of the emotionalintelligence construct. Personality and Individual Differences, 28: 539-561.
Ciarrochi, J., Chan, A., Caputi, P., & Roberts, R. 2001a. Measuring EmotionalIntelligence. In J. Ciarrochi & J. P. Forgas & J. D. Mayer (Eds.), EmotionalIntelligence in Everyday Life: A Scientific Inquiry: 25-45. New York:Psychology Press.
Ciarrochi, J., Forgas, J., & Mayer, J. (Eds.). 200 lb. Emotional intelligence ineveryday life: A scientific inquiry. Washington, DC: American PsychologicalAssociation.
Cobb, C. D. & Mayer, J. D. 2000. Emotional intelligence. Educational Leadership,58(3): 14-18.
Cohen, J. 1988. Statistical power analysis for the behavioral sciences (2nd ed.).Hillsdale, NJ: Erlaum.
Cooper, R. K. 1997. Applying emotional intelligence in the workplace. Training&
Development, 51(12): 31-38.
Cooper, R. K. & Sawaf, A. 1997. Executive EQ: Emotional Intelligence in Leadershipand Organizations. New York, New York: Gosset, Putnam.
Costa, P. T. & McCrae, R. M. 1992a. Revised NEO personality inventory [NEO-PI-R]and NEO five factor inventory [NEO-FFI] professional manual. Odessa, FL:Psychological Assessment Resources, Inc.
Costa, P. T., Jr. & McCrae, R. R. 1986. Personality stability and its implications forclinical personality. Clinical Psychology Review, 6: 407-423.
Costa, P. T., Jr. & McCrae, R. R. 1992b. Revised NEO Personality Inventory (NE0-PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual.Odessa, FL: Psychological Assessment Resources.
Costa, P. T., Jr. & McCrae, R. R. 1997. Longitudinal stability of adult personality. InR. Hogan & J. A. Johnson & S. R. Briggs (Eds.), Handbook of personalitypsychology: 269-290. New York: Academic Press.
Crampton, S. M. & Wagner, J. A. 1994. Percept - percept inflation inmicroorganizational research: An investigation of prevalence and effect.Journal of Applied Psychology, 79: 67-76.
Davies, M., Stankov, L., & Roberts, R. D. 1998. Emotional Intelligence: In search ofan elusive construct. Journal of Personality and Social Psychology, 75: 989-1015.
133
Digman, J. M. 1990. Personality structure: Emergence of the Five-Factor Model.Annual Review of Psychology, 41: 417-440.
Dillman, D. A. 2000. Mail and Internet surveys: The tailored design method. NewYork, NY: John Wiley & Sons, Inc.
Donovan, M. A. (Ed.). 2000. Web-based attitude surveys: Data and lessons learned.New Orleans, LA: 15th Annual Conference of the Society for Industrial andOrganizational Psychology.
Donovan, M. A., Drasgow, F., & Probst, T. M. 2000. Does computerizing paper-and-pencil job attitude scales make a difference? New IRT analyses offer insight.Journal of Applied Psychology, 85(2): 305-313.
Druskat, V. U. & Wolff, S. B. 2001; Group Emotional Intelligence and Its Influenceon Group Effectiveness. In C. Cherniss & D. Goleman (Eds.), TheEmotionally Intelligent Workplace: 132-155. San Francisco: Jossey-Bass.
Dulewicz, V. & Higgs, M. 2000. Emotional intelligence a review and evaluationstudy. Journal of Managerial Psychology, 15(4): 341-372.
Easterby-Smith, M., Thorpe, R. & Lowe, A. 1991. Management Research: anintroduction. London: Sage Publication, Ltd.
Elfenbein, H. A. & Ambady, N. 2002a. On the universality and cultural specificity ofemotion recognition: A meta-analysis. Psychological Bulletin, 128: 2003-2235.
Elfenbein, H. A. & Ambady, N. 2002b. Predicting Workplace Outcomes from theability to eavesdrop on feelings. Journal of Applied Psychology, 87(5): 963-971
Elfenbein, H. A., Marsh, A. A., & Ambady, N. 2002. Emotional Intelligence and theRecognition of Emotion from Facial Expressions. In L. F. Barrett & P. Salovey(Eds.), The Wisdom in Feeling: Psychological Processes in EmotionalIntelligence: 37-59. New York: Gulford Press.
Erdogan, B. Z. & Baker, M. J. 2002. Increasing mail survey response rates from anindustrial population: A cost-effectiveness analysis of four follow-uptechniques. Industrial Marketing Management, 31: 65-73.
Fenlason, K. J. 2000. Multiple data collection methods in 360 feedback programs:Implications for use and interpretation. In T. A. (Chair) (Ed.), Current issuesand challenges in the use of survey-based data. New Orleans, LA: Symposiumconducted at the 15th Annual Conference of the Society for Industrial andOrganizational Psychology.
Fisher, C. D., Schoenfeldt, L. F., & Shaw, J. B. 2003. Human Resource Management(5th ed.). Boston: Houghton Mifflin Company.
Fleishman, E. A. & Harris, E. F. 1962. Patterns of leadership behavior related toemployee grievances and turnover. Personnel Psychology, 15: 43-56.
134
Forgas, J. 2001. Affective intelligence: The role of affect in social thinking andbehavior. In J. Ciarrochi & J. P. Forgas & J. D. Mayer (Eds.), EmotionalIntelligence in Everyday Life: A Scientific Inquiry: 46-63. New York:Psychology Press.
Frida, N. H. 1993. Moods, Emotion Episodes and Emotions. In M. Lewis & J. M.Haviland (Eds.), Handbook of Emotions: 381-403. New York, NY: Guildford.
Furnham, A. & Heaven, P. 1999. Personality and social behaviour. London: Arnold.
Gardner, H. 1983. Frames of Mind: The theory of multiple intelligences. New York:Basic Books.
Gardner, H. & Hatch, T. 1989. Multiple intelligences go to school. EducationalResearcher, 18(8): 8-14.
Gardner, H. 1993. Multiple Intelligences: The Theory in Practice. New York: BasicBooks.
Gardner, L. & Stough, C. 2002. Examining the relationship between leadership andemotional intelligence in senior level managers. Leadership & OrganizationDevelopment Journal, 22(2): 38-78.
Gellatly, I. R. & Irving, P. G 2001. Personality, autonomy, and contextualperformance of managers. Human Performance, 14(3): 231-245.
Gibbs, N. & Epperson, S. E. 1995. The EQ factor, Time, Vol. 146: 60-67.
Gohm, C. L. & Clore, G. L. 2002. Affect as Information: An Individual-DifferencesApproach. In L. F. Barrett & P. Salovey (Eds.), The Wisdom in Feeling:Psychological Processes in Emotional Intelligence: 89-113. New York:Guilford Press.
Big 5 Factor Markers from.2000a. http://ipip.ori.org/ipip/.qform50.htm.March 11.
The IPIP Items in each of the preliminary scales measuring the five NE0domains.2000b.http://ipip.ori.org/ipip/appendixb.htm1.3/11/2000.
Goldberg, L. R. 1993. The structure of phenotypic personality traits. AmericanPsychologist, 48: 26-34.
Goleman, D. 1995. Emotional Intelligence. New York: Bantam Books.
Goleman, D. 1996. Emotional Intelligence: Why It Can Matter More Than IQ.London: Bloomsbury Publishing.
Goleman, D. 1997. Beyond IQ: developing the leadership competencies of emotionalintelligence. Paper presented at the 2nd International Competency Conference,London.
Goleman, D. 1998. Working with Emotional Intelligence. New York: Bantam Books.
135
Goleman, D. 2001a. An El-based theory of performance. In C. Cherniss & D.Goleman (Eds.), The Emotionally Intelligent Workplace: 27-34. San Francisco:Jossey-Bass.
Goleman, D. 200 lb. Emotional Intelligence: Issues in Paradigm Building. In C.Cherniss & D. Goleman (Eds.), The Emotionally Intelligent Workplace. NewYork: Jossey-Bass.
Goleman, D., Boyatzis, R., & Mckee, A. 2002a. The New Leaders: Transforming theArt of Leadership Into the Science of Results. Boston, Mass: Harvard BusinessSchool Press.
Goleman, D., Boyatzis, R., & Mckee, A. 2002b. Primal Leadership: Realizing thePower of Emotional Intelligence. Boston, Mass: Harvard Business SchoolPress.
Gomez-Mejia, L. R., Balkin, D. B., & Cardy, R. L. 1998. Managing HumanResources. Upper Saddle River, NJ: Prentice-Hall.
Gottfredson, L. S. 1998. The general intelligence factor. Scientific American PresentsIntelligence, 9: 24-29.
Gowing, M. K. 2001. Measurement of Individual Emotional Competence. In C.Cherniss & D. Goleman (Eds.), The Emotionally Intelligent Workplace: 83-131. San Francisco: Jossey-Bass.
Gross, J. J. & Keltner, D. 1999. Functional accounts of emotion. Cognition andEmotion, 13(September).
Gross, J. J. & John, 0. P. 2002. Wise Emotion Regulation. In L. F. Barrett & P.Salovey (Eds.), The Wisdom in Feeling: Psychological Processes in EmotionalIntelligence: 297-318. New York: Guilford Press.
Guion, R. M. & Gottier, R. F. 1965. Validity of personality measures in personnelselection. Personnel Psychology, 18: 135-164.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. 1998. Multivariate dataanalysis (4th ed.). New York: MacMillan Publishing Company.
Harman, H. H. 1967. Modern factor analysis (2nd ed.). Chicago: University ofChicago Press.
Harrison, D. A., McLaughlin, M. E., & Coalter, T. M. 1996. Context, cognition, andcommon method variance: Psychometric and verbal protocol evidence.Organizational Behavior and Human Decision Processes, 68: 641-665.
Harzing, A. W. 1997. Response rates in international mail surveys: Results of a 22-country study. International Business Studies, 6: 641-665.
Hedlund, J. & Sternberg, R. J. 2000. Too Many Intelligences? Integrating Social,Emotional, and Practical Intelligence. In R. Bar-On & J. D. A. Parker (Eds.),The Handbook of Emotional Intelligence: Theory, development, assessment,
136
and application at home, school, and in the workplace.: 136-167. SanFrancisco: Jossey-Bass.
Hernstein, R. J. & Murray, C. 1994. Race, genes, and IQ--an apolgia., The NewRepublic: 35-37.
Herrnstein, R. J. & Murray, C. 1994. The Bell Curve: Intelligence and class structurein American life. New York: Free Press.
Higgs, M. 2001. Is there a relationship between the Myers-Briggs type indicator andemotional intelligence? Journal of Managerial Psychology, 16(7): 509-533.
Hochschild, A. R. 1979. Emotion Work, Feeling Rules, and Social Structure.American Journal of Sociology(November): 555-575.
Hochschild, A. R. 1983. The managed heart: Commercialization of human feeling.Berkeley, CA: University of California Press.
Hogan, J., Rybicki, s. L., Motowidlo, S. J., & Borman, W. C. 1998. Relations betweencontextual performance, personality, and occupational advancement. HumanPerformance, 11(2-3): 189-207.
Hogan, J. & Holland, B. 2003a. Using Theory to Evaluate Personality and Job-Performance Relations: A Socioanalytic Perspective. Journal of AppliedPsychology, 88(1): 100-112.
Hogan, J. & Holland, B. 2003b. Using theory to evaluate personality and job-performance relations: A socioanalytic perspective. Journal of AppliedPsychology, 88: 100-112.
Hogan, R. 1983. A socioanalytic theory of personality. In M. M. Page (Ed.), 1982Nebraska Symposium on Motivation: 55-89. Lincoln: University of NebraskaPress.
Hogan, R. 1986. Manual for the Hogan Personality inventory. Minneapolis: NationalComputer Systems.
Hogan, R. 1991. Personality and personality measurement. In D. L. M. Hough (Ed.),Handbook of industrial and organizational psychology, 2nd ed., Vol. 2: 327-396. Palo Alto, CA: Consulting Psychologists Press.
Hogan, R. 1996. A socioanalytic perspective on the five-factor model. In J. S. Wiggins(Ed.), The five-factor model of personality: 163-179. New York: GuilfordPress.
Hogan, R. & Shelton, D. 1998a. A socioanalytic perspective on job performance.Human Performance, 11: 129-144.
Hogan, R. & Shelton, D. 1998b. A socioanalytic perspective on job performance.Human Performance, 11: 129-144.
137
Hogan, R. & Roberts, B. W. 2000. A socianalytic perspective on person-environmentinteraction. In W. B. Walsh & K. H. Craik (Eds.), Person-environmentpsychology: New directions and perspectives., 2nd ed. Mahwah, NJ: Erlbaum.
H6pfl, H. & Linstead, S. 1997. Learning to feel and feeling to learn: emotion andlearning in organisations. Management Learning, 28(1): 5-12.
Hough, L. M. 1992. The "Big Five" personality variables -- construct confusion:Description versus prediction. Human Performance, 5: 139-156.
Hough, L. M. & Oswald, F. L. 2000. Personnel Selection: Looking Toward theFuture--Remembering the Past. Annual Review of Psychology, 51: 631-664.
Howard, P. J. & Howard, J. M. 2001. The Owner's Manual For Personality at Work.Marietta, GA: Bard Press.
Howell, J. P., Dorfman, P. W., & Kerr, S. 1986. Moderator variables in leadershipresearch. Academy of Management Review, 11: 88-102.
Hurtz, G M. & Donovan, J. J. 2000. Personality and job performance: The Big Fiverevisited. Journal of Applied Psychology, 85(6): 869-879.
Ickes, W. 1997. Empathic accuracy. New York: Guilford.
Janovics, J. & Christiansen, N. D. 2001. Emotional intelligence at the workplace.Paper presented at the 16th Annual Conference on the Society of Industrialand Organizational Psychology, San Diego.
John, 0. P. 1990. The "Big Five" factor taxonomy: Dimensions of personality in thenatural language and in questionnaires. In L. A. Pervin (Ed.), Handbook ofpersonality: Theory and research, Vol. 65: 66-100. New York: Guilford.
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 AppliedPsychology, 86(1): 80-92.
Jung, C. 1921. Psychological types (r. R.F.C. Hull (Rev. Trans.), Trans.). Princeton,NJ: Princeton University Press.
Katz, R. L. 1955. Skills of an effective administrator. Harvard BusinessReview(January-February): 33-42.
Kelly-Milburn, D. & Milburn, M. A. 1995. Cyberpsych: Resources for psychologiston the Internet. Psychological Science, 6: 203-211.
Konovsky, M. A. & Organ, D. W. 1996. Dispositional and contextual determinants oforganizational citizenship behavior. Journal of Organizational Behavior, 17:253-266.
Kotler, P. & Armstrong, G 2001. Principles of Marketing (9th ed.). Upper SaddleRiver, New Jersey: Prentice-Hall.
138
Kraut, A. I. & Saari, L. M. 1999. Organization surveys: Coming of age for a new era.In A. I. Kraut & A. K. Korman (Eds.), Evolving practices in human resourcemanagement: 302-327. San Francisco, CA: Jossey-Bass.
Lam, L. T. & Kirby, S. L. 2002. Is Emotional Intelligence an Advantage? AnExploration of the Impact of Emotional and General Intelligence on IndividualPerformance. Journal of Social Psychology, 142(1): 133-143.
Landis, C. 1995. An exploratory study of science educators' use of the Internet.Journal of Science Education and Technology, 4(3): 181-190.
Lane, R. D. & Pollermann, B. Z. 2002. Complexity of Emotion Representations. In L.F. Barrett & P. Salovey (Eds.), The Wisdom in Feeling: PsychologicalProcesses in Emotional Intelligence: 271-298. New York: Guilford Press.
Law, K., S., Song, L. J., & Wong, C. S. 2002. Emotional Intelligence as an abilityfacet: construct validation and its predictive power ofjob outcomes. Paperpresented at the Academy of Management, Denver Co.
Law, K., S., Wong, C.-S., & Song, L. J. 2004. The Construct and Criterion Validity ofEmotional Intelligence and Its Potential Utility for Management Studies.Journal of Applied Psychology, 89(3): 483-496.
Lawler, E. E. 1986. High-involvement management: Participative strategies forimproving organizational performance. San Francisco: Jossey-Bass.
Lawler, E. E., Mohrman, S. A., & Ledford, G E. 1995. Creating high performanceorganizations: Practices and results of employee involvement and total qualityin Fortune 1000 companies. San Francisco: Jossey-Bass.
Lawler, E. E. 1998. Strategies for high performance organizations. San Francisco:Jossey-Bass.
Leary, M. R. & Kowalski, R. M. 1990. Impression management: A literature reviewand two-component model. Psychological Bulletin, 107: 34-47.
Leary, T. 1957. Interpersonal diagnosis of personality. New York, NY: Ronald Press.
Leyman, H. 1996. The content and development of mobbing at work. EuropeanJournal of Work and Organizational Psychology, 5: 165-184.
Lo, A. 2004. Fire-and-hire attitude costs Hk$39b a year, survey finds, South ChinaMorning Post: 1. Hong Kong.
Lo, S. 2003. Perceptions of work-family conflict among married female professionalsin Hong Kong. Personnel Review, 32(3): 376-390.
Lo, S., Stone, R., & Ng, C. W. 2003. Work-family conflict and coping strategiesadopted by female married professionals in Hong Kong. Women inManagement Review, 18(3/4): 182.
139
Lundelius, J. 0. 1997. Languages and business: The Hong Kong Mix. BusinessCommunication Quarterly, 60(4): 112-125.
Magnan, S. M., Lundby, K. M., & Fenlason, K. J. 2000. Dual media: The art andscience of paper and Internet employee survey implementation. In T. A.(Chair) (Ed.), Current issues and challenges in the use of survey-based data.New Orleans, LA: Symposium conducted at the 15th Annual Conference ofthe Society for Industrial and Organizational Psychology.
Marlowe, H. A. J. 1986. Social Intelligence: Evidence for multidimensionality andconstruct independence. Journal of Educational Psychology, 78(1): 52-58.
Maruyama, G M. 1998. Basics of structural equations modeling. Thousand Oaks,CA: Sage.
Matthews, G., Zeidner, M., & Roberts, R. D. 2002. Emotional Intelligence: Scienceand Myth. Cambridge, Mass: The MIT Press.
Mayer, J., Salovey, P., Caruso, D. R., & Sitarenios, G 2003a. Measuring EmotionalIntelligence with the MSCEIT V2.0. Emotion, 3(1): 97-105.
Mayer, J. D., DiPaolo, M. T., & Salovey, P. 1990. Perceiving affective content inambiguous visual stimuli: A component of emotional intelligence. Journal ofPersonality Assessment, 54: 772-781.
Mayer, J. D. & Salovey, P. 1993. The intelligence of emotional intelligence.Intelligence, 17: 433-422.
Mayer, J. D. & Salovey, P. 1997. What is emotional intelligence? In P. Salovey & D.Sluyter (Eds.), Emotional development, emotional literacy, and emotionalintelligence: 3-31. New York: Basic Books.
Mayer, J. D., Salovey, P., & Caruso, D. R. 1997. Emotional IQ test (CD-ROM).Needham, MA: Virtual Knowledge.
Mayer, J. D. 1999. Emotional intelligence: popular or scientific psychology?, APAMonitor Online, Vol. 30: [Shared Perspectives column].
Mayer, J. D., Caruso, D. R., & Salovey, P. 1999. Emotional intelligence meetstraditional standards for an intelligence. Intelligence, 27: 267-298.
Mayer, J. D., Caruso, D. R., & Salovey, P. 2000a. Selecting a Measure of EmotionalIntelligence: The Case for Ability Scales. In R. Bar-On & J. D. A. Parker(Eds.), The Handbook of Emotional Intelligence: Theory, Development,Assessment, and Application at Home, School, and in the Workplace: 320-342.San Francisco: Jossey-Bass.
Mayer, J. D., Salovey, P., & Caruso, D. R. 2000b. Emotional Intelligence as Zeitgeist,as Personality, and as a Mental Ability. The Handbook of EmotionalIntelligence: Theory, development, assessment, and application at home,school, and in the workplace.: 92-117.
140
Mayer, J. D., Salovey, P., & Caruso, D. R. 2000c. Emotional Intelligence. In R. J.Sternberg (Ed.), Handbook of intelligence, 2nd ed.: 396-421. New York:Cambridge University Press.
Mayer, J. D., Salovey, P., Caruso, D. R., & Sitarenios, G 2001. Emotional IntelligenceAs a Standard Intelligence: A Reply. Emotion, 1: 232-242.
Mayer, J. D., Salovey, P., & Caruso, D. R. 2002. Mayer-Salovey-Caruso EmotionalIntelligence Test: User's Manual. Toronto, Canada: Multi-Health Systems Inc.
Academic Researcher's Home Page for Emotional Intelligence & theMSCEIT.2003b.http://www.emotionaliq,org/Model.htm
McClelland, D. C. 1998. Identifying competencies with behavioral-event interviews.Psychological Science, 9(5): 331-340.
McCrae, R. R. & John, 0. P. 1992. An Introduction to the Five-Factor Model and itsapplications. Journal of Personality, 60: 175-215.
McCrae, R. R., Costa, P. T., Jr., Lima, M. P. d., Simoes, A., Ostendorf, F., Angleitner,A., Marui, I., Bratko, D., Caprara, G V., Barbaranelli, C., Chae, J. H., &Piedmont, R. L. 1999. Age differences in personality across the adult life span:Parallels in five cultures. Developmental Psychology, 35: 466-477.
McCrae, R. R. 2000. Emotional Intelligence from the perspective of the Five-FactorModel of Personality. In R. Bar-On & J. D. A. Parker (Eds.), The Handbook ofEmotional Intelligence: Theory, development, assessment, and application athome, school, and in the workplace.: 263-276. San Francisco: Jossey-Bass.
McGrath, J. E. 1982. Dilemmatics: The study of research choices and dilemmas. InR. A. Kulka (Ed.), Judgment calls in research: 69-102. Beverly Hills, CA:Sage.
Michael, P. 2004. Overworked and stressed? Most Hongkongers are, survey finds,South China Morning Post: 1. Hong Kong.
Mook, D. G 1983. In defense of external invalidity. American Psychologist, 38(4):379-387.
Motowidlo, S. J. & Van Scotter, J. R. 1994. Evidence that task performance should bedistinguished from contextual performance. Journal of Applied Psychology,79: 475-480.
Motowidlo, S. J., Borman, W. C., & Schmit, M. J. 1997. A theory of individualdifferences in task and contextual performance. Human Performance, 10: 71-83
Mount, M. K., Barrick, M. R., & Strauss, J. P. 1994. Validity of Observer Ratings ofthe Big Five Factors. Journal of Applied Psychology, 79(2): 279-281.
Mount, M. K. & Barrick, M. R. 1995. The Big Five personality dimensions:Implications for research and practice in human resources management. In K.
141
M. R. G. Ferris (Ed.), Research in personnel and human resourcesmanagement, Vol. 13: 153-200. Greenwich, CT: JAI Press.
Murray, H. A. 1938. Explorations in personality. New York: Oxford Press.
Norusis, M. S. 2001. SPSS 10.0 guide to data analysis. New Jersey: Prentice Hall.
Nunnally, J. C. & Bernstein, I. H. 1994. Psychometric theory (3rd ed.). New York:MacMillan.
O'Leary, A. M., Duffy, M. K., & Griffin, R. W. 2000. Construct confusion in the studyof antisocial work behavior, Research in Personnel and Human ResourcesManagement, Vol. 18: 257-303: JAI Press, Inc.
Ones, D. S., Viswesvaran, C., & Schmidt, F. L. 1993. Comprehensive meta-analysisof integrity test validities: Findings and implications for personnel selectionand theories of job performance. Journal of Applied Psychology, 78(4): 679-703.
Ones, D. S. & Viswesvaran, C. 1997. Personality determinants in the prediction ofaspects of expatriate job success. In Z. Aycan (Ed.), New approaches toemployee management, Vol. 4: Expatriate management: Theory and research:63-92. Stamford, CT: JAI Press, Inc.
Organ, D. W. 1988. Organizational citizenship behavior: The good soldier syndrome.Lexington, MA: Lexington Books.
Organ, D. W. 1997. Organizational citizenship behavior: It's construct clean-up time.Human Performance, 10: 85-97.
O'Shaughnessy, J. 1999. Review of 'Working with Emotional Intelligence'. Journal ofMacromarketing(December): 178-179.
Ostroff, C., Kinicki, A. J., & Clark, M. A. 2002. Substantive and Operational Issues ofResponse Bias Across Levels of Analysis: An Example of Climate-SatisfactionRelationships. Journal of Applied Psychology, 87(2): 355-368.
Phares, E. J. & Chaplin, W. F. 1997. Introduction to personality (4th ed.). New York:Longman.
Podsakoff, P. M. & Organ, D. W. 1986. Self-reports in organizational research:Problems and prospects. Journal of Management, 12: 531-544.
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y, & Paodsakoff, N. P. 2003. CommonMethod Biases in Behavioral Research: A critical Review of the Literature andRecommended Remedies. Journal of Applied Psychology, 88(5).
Pugh, S. D. 2001. Service with a smile: Emotional contagion in the service encounter.Academy of Management Journal, 44(5): 1018-1027.
142
Riemann, R., Angleitner, A., & Strelau, J. 1997. Genetic and environmental influenceson personality: A study of twins reared together using the self- and peer reportNEO-FFI scales. Journal of Personality, 65: 449-475.
Robbins, S. 2000. Managing Today! (2nd ed.). Upper Saddle River, NJ: Prentice Hall.
Robinson, S. L. & Bennett, R. J. 1995. A typology of deviant workplace behaviors: Amultidimensional scaling method. Academy of Management Journal, 36: 555-572.
Rosenthal, R., Hall, J. A., DiMatteo, M. R., Rogers, P. L., & Archer, D. 1979.Sensitivity to nonverbal communication: The PONS Test. Baltimore, MD:Johns Hopkins University Press.
Rozell, E. J., Pettijohn, C. E., & Parker, J. D. A. 2001. An empirical evaluation ofemotional intelligence: The impact on management development. Journal ofManagement Development, 21(4): 272-289.
Saarni, C. 1997. Emotional competence and self-regulation in childhood. In P.Salovey & D. J. Sluyter (Eds.), Emotional development and emotionalintelligence: 35-66. New York: Basic Books.
Sackett, P. R., Gruys, M. L., & Ellingson, J. E. 1998. Ability-personality interactionswhen predicting job performance. Journal of Applied Psychology, 83(4): 545-556.
Salgado, J. F. & Rumbo, A. 1997. Personality and job performance in financialservices manager. International Journal of Selection and Assessment, 5: 91-99.
Salovey, P. & Mayer, J. D. 1990. Emotional Intelligence. Imagination, Cognition, andPersonality, 9: 185-211.
Sapolsky, R. M. 1998. Open season: When do we lose our taste for the new?, NewYorker, Vol. March 30: 57-58,71-72.
Schaefer, D. & Dillman, D. 1998. Development of a standard e-mail methodology:Results of an experiment. Public Opinion Quarterly, 62: 378-397.
Schein, E. H. 1980. Organizational Psychology. Englewood Cliffs, NJ: Prentice-Hall.
Schmidt, F. L. & Hunter, J. E. 1992. Development of a causal model of processesdetermining job performance. Current Directions in Psychological Science, 1:89-92.
Schmidt, F. L. & Hunter, J. E. 1996. Measurement error in psychological research:Lessons from 26 research scenarios. Psychological Methods, 1: 199-223.
Schmidt, F. L. & Hunter, J. E. 1998. The validity and utility of selection methods inpersonnel psychology: Practical and theoretical implications of 85 years ofresearch findings. Psychological Bulletin, 124: 262-274.
143
Schmitt, N. W., Gooding, R. Z., Noe, R. A., & Kirsch, M. 1984. Meta-analyses ofvalidity studies published between 1964 and 1982 and the investigation ofstudy characteristics. Personnel Psychology, 37: 407-422.
Schneider, K. C. & Johnson, K. C. 1995. Stimulating response to market surveys ofbusiness professionals. Industrial Marketing Management, 24: 265-276.
Scholl, R. W., Cooper, E. A., & McKenna, J. F. 1987. Referent selection indetermining equity perceptions: Differential effects on behavioral andattitudinal outcomes. Personnel Psychology, 40(1): 113-124.
Schutte, N. S., Malouff, J. M., Hall, L. E., Haggerty, D. J., Cooper, J. T., Golden, C. J.,& Dornheim, L. 1998. Development and validation of a measure of emotionalintelligence. Personality and Individual Differences, 25(167-177).
Sekaran, U. 2000. Research Methods for Business: A Skill-building Approach. NewYork: John Wiley & Sons, Inc.
Sin, H. P., Harrison, D. A., Shaffer, M. A., & Lau, V. 2004. Breaking ties:Relationship disruptive behaviors at work. Paper presented at the Academy ofManagement Meetings, New Orleans, LA.
Smith, H. 1991. The World's Religions. New York: HaperCollins.
Spearman, C. 1904. "General intelligence" objectively determined and measured.American Journal of Psychology, 15(201-293).
Spector, P. E. & Jex, S. M. 1998. Development of four self-report measures of jobstressors and strain: Interpersonal conflict at work scale, organizationalconstraints scale, quantitative workload inventory, and physical symptomsinventory. Journal of Occupational Health Psychology, 3: 356-367.
Spencer, L. M. 1986. Calculating human resource costs and benefits. New York, NY:Wiley.
Spencer, L. M. & Spencer, S. M. 1993. Competence at work: Models for superiorperformance. New York, NY: Wiley.
Spencer, L. M. 2001. The Economic Value of Emotional Intelligence Competenciesand ETC-Based HR Programs. In C. Cherniss & D. Goleman (Eds.), TheEmotionally Intelligent Workplace: 45-82. San Francisco, CA: Jossey-Bass.
Stanton, J. M. 1998. An empirical assessment of data collection using the Internet.Personnel Psychology, 51: 709-724.
Stein, S. & Book, H. 2001. The EQ Edge: Emotional Intelligence and Your Success.Toronto: Stoddart Publishing Co.
Steiner, C. 1997. Achieving Emotional Literacy. London: Bloomsbury Publishing.
Sternberg, R. J. 1999. Working with Emotional Intelligence. Personnel Psychology,52(3): 780-783.
144
Sullivan, H. S. 1953. The interpersonal theory of psychiatry. New York, NY: Norton.
Tett, R. P., Jackson, D. N., & Rothstein, M. 1991. Personality measures as predictorsof job performance: A meta-analytic review. Personnel Psychology, 44(4):703-742.
Tett, R. P., Jackson, D. N., Rothstein, M., & Reddon, J. R. 1999. "Meta-analysis ofbidirectional relations in personality-job performance research": Errata.Human Performance, 12(2): 177-181.
Tett, R. P. & Guterman, H. A. 2000. Situation trait relevance, trait expression, andcross-situational consistency: Testing a principle of trait activation. Journal ofResearch in Personality, 34: 397-423.
Tett, R. P. & Burnett, D. D. 2003. A Personality Trait-Based Interactionist Model ofJob Performance. Journal of Applied Psychology, 88(3): 500-517.
Thompson, L. F., Surface, E. A., Martin, D. L., & Sanders, M. G 2003. From paper topixels: Moving personnel surveys to the Web. Personnel Psychology, 56: 197-227.
Thorndike, E. L. 1920a. A constant error in psychological ratings. Journal of AppliedPsychology, 11(3): 25-29.
Thorndike, R. L. 1920b. Intelligence and its uses. Harper's Magazine: 140,227-235.
Toms, M. 1998. Emotional intelligence in the workplace: An interview with DanielGoleman., The Inner Edge: 14-17.
Tsai, W. C. 2001. Determinants and consequences of employee displayed positiveemotions. Journal of Management, 27: 497-512.
Van Scotter, J., Motowidlo, S. J., & Cross, T. C. 2000. Effects of task performanceand contextual performance on systemic rewards. Journal of AppliedPsychology, 85(4): 526-535.
Van Scotter, J. R. & Motowidlo, S. J. 1996. Interpersonal facilitation and jobdedication as separate facets of contextual performance. Journal of AppliedPsychology, 81(5): 525-531.
Vinchur, A. J., Schippmann, J. S., Switzer, F. S., III, & Roth, P. L. 1998. A meta-analytic review of predictors of job performance for salespeople. Journal ofApplied Psychology, 83(4): 586-597.
Vleeming, R. G 1979. Machiavellianism: A Preliminary Review. PsychologicalReports(February): 295-310.
Warr, P. 1999. Logical and judgmental moderators of the criterion-related validity ofpersonality scales. Journal of Occupational and Organizational Psychology,72: 187-204.
145
Wechsler, D. 1940. Nonintellective factors in general intelligence. PsychologicalBulletin, 37: 444-445.
Wechsler, D. 1958. The Measurement and Appraisal of Adult Intelligence (4th ed.):Williams and Wilkins.
Weiss, H. M. & Cropanzano, R. 1996. Affective Events Theory. In B. M. Staw & L. L.Cummings (Eds.), Research in Organizational Behavior, Vol. 18: 17-19.Greenwich, CT: JAI Press.
Welbourne, T. M., Johnson, D. E., & Erez, A. 1998. The role-based performancescale: Validity analysis of a theory-based measure. Academy of ManagementJournal, 41: 540-555.
Wiggins, J. S. 1979. A psychological taxonomy of trait-descriptive terms: Theinterpersonal domain. Journal of Personality and Social Psychology, 37: 395-412.
Wolfe, C. & Caruso, D. R. 2002. MSCEIT Certification Workshop August 6-10, 2002.Mashantucket, Connecticut.
Wong, C. S. & Law, K., S. 2002. The Effect of leaders' and Followers' EmotionalIntelligence on Performance and Attitudes: An Exploratory Study. LeadershipQuarterly, 13(3): 243.
Yost, P. R. & Homer, L. E. 1998. Electronic versus paper surveys: Does the mediumaffect the response? Dallas, TX: Paper presented at the 13th AnnualConference of the Society of Industrial and Organizational Psychology.
Young, S. A., Daum, D. L., Robie, C., & Macey, W. H. 2000. Paper versus web surveyadministration: Do different methods yield different results? In M. Sederberg& S. R. (Chairs) (Eds.), Improving the survey effort: Methodological questionsand answers. New Orleans, LA: Symposium conducted at the 15th AnnualConference of the Society for Industrial and Organizational Psychology.