A CRITICAL EVALUATION OF THE ROLE OF EMOTIONAL INTELLIGENCE IN LEADERSHIP

EFFECTIVENESS

David Rosete and Joseph Ciarrochi

University of Wollongong

Correspondence addressed to:

David Rosete
Organisational Psychologist
Department of Psychology, University of Wollongong
Wollongong, Australia, 2500.
Phone: 61-02-4223-2345
Fax: 61-02-4223-2404
Email: <[email protected]>

Word Count:

6,000

Classification:

Feature Article

Running Head:

Emotional Intelligence & Leadership Effectiveness

When referencing please use:

Rosete, D. & Ciarrochi, J. (2005). EI and its relationship to workplace performance

outcomes of leadership effectiveness. Leadership Organizational Development, 26, 388399

Abstract
Does Emotional Intelligence (EI) make someone a better leader? We utilized a
cross-sectional survey to examine the relationships between leadership effectiveness and
tests of ability EI and cognitive reasoning, and self-report measures of EI and personality.
The sample consisted of 117 senior executives (Males = 56; Females = 60; 1 unreported;
mean age = 40.54). Regression analyses indicated medium to large relationships between
ability EI scales and achieving business outcomes. The results held even after controlling
for reasoning ability, self-reported EI, and self-reported personality. Self-reported EI also
was linked to business outcomes, but was no longer significant when controlling for
personality. These results may have important implications for how we select and develop
executives.

Key Words: emotional intelligence, leadership effectiveness, ability measures, personality,

cognitive reasoning

A CRITICAL EVALUATION OF THE ROLE OF EMOTIONAL INTELLIGENCE IN LEADERSHIP

EFFECTIVENESS
Does Emotional Intelligence (EI) make someone a better leader? This question has
preoccupied writers and scientists for the past decade (Higgs, 2003). Many factors have
been identified as important to leadership, such as cognitive ability (Sternberg & Vroom,
2002) and conscientiousness (Barrick & Mount, 1991). With little evidence, people have
also proposed that emotional intelligence (EI) is an important predictor of leadership
performance, perhaps the most important predictor (Goleman, 1998; McKee, 2005). The
purpose of this paper is to critically evaluate this claim. Specifically, we examine the
extent to which different EI measures predict leadership performance, even after
controlling for traditional measures of personality and cognitive reasoning.

Emotional Intelligence
In 1990, Salovey and Mayer drew together much of the material available on
cognition (i.e., capacity to carry out abstract reasoning) and emotion (defined as signals
that convey regular, discernable meanings about relationships) and developed a theory of
EI. There more recent refinement of their definition see EI as the ability to recognise
the meanings of emotions and their relationships and to reason and problem-solve based
on them (Mayer, Caruso & Salovey, 1999, p. 267). EI is seen as the capacity to perceive
emotions, assimilate emotion-related feelings, understand the information of those
emotions, and manage them.
Between 1994 and 1997, EI was popularized by psychologist and journalist Daniel
Goleman (1995) in his book Emotional Intelligence. During this time, Goleman (1995)

changed the definition quite substantially with the new definition appearing to equate EI
with good social behavior. Since 1997 there has been an explosion of activity in a new
and now fuzzily-defined area. EI is now used popularly to mean various things, including
motivation, empathy, sociability, warmth, and optimism (Mayer, 2001).
Mayer and Salovey (1997) argue that there are two major EI models that drive
research: (1) ability models and (2) mixed models. Ability models conceptualise EI in a
similar way to cognitive intelligence (i.e., Intelligence Quotient - IQ). These models
suggest that EI should develop over time, be correlated with measures of IQ, and be
measurable with a test based on performance (Ciarrochi, Chan & Caputi 2000; Mayer,
Salovey, & Caruso 2004). In contrast, mixed models of EI incorporate both non-cognitive
models (e.g., BarOn 1997) and competency-based models (e.g., Goleman 1995). These
mixed models typically overlap or mix with traditional models of personality and tend to
utilise self-reports as their primary mode of assessment.
Importantly, while each approach conceptualises and measures EI from different
perspectives, the approaches themselves appear to complement rather than contradict
each other (Ciarrochi, Chan & Caputi, 2000). For example, both mixed and ability based
measures of EI have been shown to be only modestly related to each other and to relate
to important criteria such as social support, mental health, and social behavior (Ciarrochi,
Scott, Deane, and Heaven, 2003; Mayer, Salovey and Caruso, 2004).
A critical question is whether EI is distinguishable from traditional measures of
personality and IQ. The ability based measure of EI has been repeatedly shown to have
incremental value over traditional measures (Ciarrochi, Chan, & Caputi, 2000; Ciarrochi,
Dean & Anderson, 2002; Mayer in press; Mayer, Salovey, Caruso, & Sitarenios, 2003;

Mayer, Salovey & Caruso). In contrast, mixed model measures of EI correlate, sometimes
strongly, with measures of personality (Ciarrochi, Chan & Caputi, 2000; Ciarrochi, Chan,
Caputi, and Roberts, 2001; MacCann, Roberts, Mathews, and Zeidner, 2004; Rosete
2004). Some research has shown that these measures can show incremental value over
personality (Ciarrochi, Chan, & Bajgar, 2001; Ciarrochi, Scott, Deane, & Heaven, 2003;
Ciarrochi, & Scott, 2005; Schutte, Malouff, Hall, Haggerty, Cooper, Golden, & Dornhein,
1997), whereas others have shown that they add little incremental value (Davies, Stankov,
& Roberts, 1998).
Our study sought to establish the extent to which mixed and ability based measures
of EI were useful for predicting leadership effectiveness over and above traditional
personality and cognitive ability measures.

Leadership Effectiveness and Emotional Intelligence

People have theorized that EI contributes to peoples capacity to work effectively in
teams, manage stress, and/or lead others (Ashkanasy & Trevor-Roberts, 2000; Mayer,
Salovey, & Caruso, 2004; George, 2000; Goleman, 1998; Goleman, Boyatzis, & Mckee,
2002; Prati, Douglas, Ferris, Ammeter, & Buckley, 2003). For example, leaders who are
poor at perceiving their emotions may unknowingly miss important emotional signals from
their co-workers. Similarly, leaders who are poor at managing emotions may allow their
emotions to interfere with effective action. For instance, when they feel anxious, they
may avoid giving an important speech, or when they feel angry, they may inappropriately
lash out at a co-worker.

The transformational/transactional leadership model of Bass and Avolio (1990) has

provided the general framework for most of the research on emotional intelligence and
leadership effectiveness (see for example, Barling, Slater & Kelloway, 2000; Palmer, Walls,
Burgess & Stough, 2001; Gardner & Stough, 2002). Transformational leaders are seen as
those people that are able to create a vision, communicate this vision, build commitment
amongst subordinates to the vision, and model the vision within the workplace. In
contrast, transactional leaders are viewed more as managers that maintain the status quo.
It is argued that transformational leaders are able to deal with strategic matters more
efficiently and in turn are able to build commitment in employees, and are therefore more
likely to take an organization forward (Bass and Avolio, 1994; McShane & Von Glinow,
2000).
The available EI-Leadership research supports the hypothesis that self-reported EI is
linked to self-reported transformational leadership style (Barling, Slater & Kelloway, 2000,
Palmer, Walls, Burgess & Stough, 2001; Gardner & Stough, 2001). Barling, Slater and
Kelloway (2000) conducted an exploratory study on the relationship between emotional
intelligence and transformational leadership. Their results suggest that self-reported
emotional intelligence is associated with three aspects of transformational leadership,
namely idealised influence, inspirational motivation and individualised consideration. The
leaders who report exhibiting these behaviours were assumed to be more effective in the
workplace.
Palmer, Walls, Burgess and Stough (2001) administered a self-report EI measure to
43 managers in order to evaluate the link between EI and leadership style. They found
significant correlations with several components of the transformational leadership model.

Specifically, the inspirational, motivation and individualised consideration components of

transformational leadership correlated with self-reported ability to both monitor and
manage emotions.
While research based on self-reported managerial style is important, it does not
establish whether an ability-based test of EI or self-report measure of EI is related to
relatively objective (or non-self-report) measures of leadership effectiveness such as
supervisor ratings or measures of actual business unit performance. In one relevant
study, Lopes, Ct, Grewal, Kadis, Gall and Salovey (2003) examined the link between an
ability measure of EI and several indicators of job performance, including salary, merit
increases, and company rank. They also assessed interpersonal facilitation, affect and
attitudes at work, and leadership potential, using both peer and supervisor ratings in 44
analysts and clerical/administrative employees. The findings revealed that EI was related
to the percentage of pay increase, internal classification level, and better peer and
supervisor ratings. Importantly, these results held even after controlling for the effects of
cognitive ability and personality traits.
In another study, Rosete and Ciarrochi (in press) conducted a small exploratory
study of the relationship between an ability measure of emotional intelligence, personality,
cognitive intelligence and leadership effectiveness amongst senior executives. Leadership
effectiveness was assessed using both managerial performance ratings and a 360
assessment involving each leaders subordinates and direct manager ratings. Correlational
analyses revealed that higher emotional intelligence was associated with higher leadership
effectiveness

In the present study we sought to replicate and extend past research by utilising a
larger sample, and administering both self-report and ability based measures of EI. We
also administered a measure of personality and reasoning ability, to examine the
incremental value of the EI measures over measures that are commonly used in the
organizational setting.

The Present Study and Research Questions

Our primary measures of leadership effectiveness will be based on a structured
performance management assessment. The structured performance management system
assesses employees progress towards organisational goals (Corporate Leadership Council,
2002). This system is seen as a good indicator of an individuals leadership effectiveness
(Management Advisory Committee 2001; Corporate Leadership Council 2002; Hogan,
Curphy & Hogan, 1994). It establishes whether an individual meets business outcomes in
such a manner that they not only achieve results but also build effective working
relationships. Importantly, the performance management system is tied to concrete
outcomes for the individual: Leaders who score well on the system receive larger bonus
pay increases (up to 15% of their normal pay).
This study sought to address two major research questions. First, to what extent are
self-reported and ability based measures of EI related to leadership performance?
Second, to what extent do the EI measures predict variance in leadership performance
over and above that accounted for by traditional personality and cognitive reasoning
measures?

Method
Participants
The sample consisted of 122 executives from a large Australian Public Service
organisation. Executives who did not complete the main battery of measures or whose
performance ratings were not accessible were then excluded leaving a final sample of 117
executives (56 men, 60 women, 1 unreported; mean age 40.54, SD = 8.95). The final
sample consisted of both assistant directors (n = 64) and directors (n = 52) with 1
unreported. Multivariate analyses revealed that there were no differences between these
groups in terms of gender, age, tenure and qualifications.
Participants mean tenure within the organisation was approximately 15 years. The
level of qualifications amongst the sample was quite impressive with approximately 32%
reported to have completed a masters degree or higher; 22% reported to have completed
a postgraduate diploma or certificate; 33% reported to have completed an undergraduate
degree and 4% reported to have completed some form of associate diploma or certificate.

Procedure
Participants were administered a battery of psychological tests that assessed
personality, reasoning ability, and EI. In exchange for their participation, individuals were
provided with a confidential feedback report on their results on each of the instruments.

Materials

The Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT) v 2.0

The MSCEIT v.2.0 is an ability measure of EI requiring participants to complete a set
of tasks associated with either perceiving emotion, using emotion, understanding
emotional information or managing emotions (Mayer, Salovey & Caruso, 2002). It
contains 141 items which are broken down into eight tasks, which are further divided into
four branches of abilities including (a) perceiving emotion, (b) using emotion to facilitate
thought, (c) understanding emotion, and (d) managing emotions. The research to date
on the MSCEIT has shown good reliability and promising validity studies across a diverse
range of psychological constructs (Mayer, Caruso and Salovey, 2000; Mayer, Salovey,
Caruso and Sitarenios, 2003; Palmer, Gignac, Manocha and Stough, in press; Roberts,
Zeidner & Matthews, 2001).

The Swinburne University Emotional Intelligence Test (SUEIT)

Given space limitations in our survey, we could not include every self-report measure
of emotional intelligence that exists in the literature. We thus sought a measure that was
representative of self-report measures in general. The SUEIT was developed after an
extensive factor analytic study involving five of the predominant measures of EI (Palmer &
Stough 2001; Palmer 2003). Four of these were commonly used self-report measures of
EI including the Bar-On Emotional Quotient Inventory (Bar-On, 1997), the Trait MetaMood Scale (Salovey, Mayer, Goldman & Palfai, 1995), the twenty-item Toronto
Alexithymia Scale-II (TAS-20; Bagby, Taylor & Parker, 1994a,b), and the Emotional
Intelligence Scale (EIS, Schutte, Malouff, Hall, Haggerty, Cooper, Golden and Dornhein,

10

1997). In addition to representing a number of central EI dimensions, the SUEIT also had
the advantage of focusing on EI in the workplace, rather than general EI.
The SUEIT consists of 65 items that ask participants to indicate the extent to which a
particular statement is true of the way they typically think, feel and act at work (on a fivepoint scale: 1 = never, 2 = seldom, 3 = sometimes, 4 = usually, 5 = always). The SUEIT
report provides an overall EI score that indicates a participants general workplace EI, and
five sub-scale scores including (1) emotional recognition and expression (in oneself) which
is the ability to identify ones own feelings and emotional states, and the ability to express
those inner feelings to others (2) emotions direct cognition, which measure the extent to
which emotions and emotional knowledge is incorporated in decision making and/or
problem solving (3) understanding of emotions external, which is the ability to identify
and understand the emotions of others and those manifest in external stimuli (4)
emotional management, which is the ability to manage positive and negative emotions
both within oneself and others and (5) emotional control, which is how effectively
emotional states experienced at work such as anger, stress, anxiety and frustration are
controlled (Palmer and Stough, 2001, p. 5).
While a relatively new EI measure, there is growing research to support the SUEITs
reliability and predictive validity with leadership factors (Palmer and Stough, 2001; Palmer,
Walls, Burgess and Stough, 2001; Palmer, Gardner and Stough, 2003; Rosete, 2004).

Measuring Personality via the 16PF

Participants completed the well validated Sixteen Personality Factor (16PF)
questionnaire (Conn & Rieke, 1998). The 16PF was chosen as it is a widely used and

11

recognised personality test within the Australian Public Service with availability of
Australian norms. The total scale contains 185 items and 16 subscales each containing 10
to 15 items.

Measuring Cognitive Ability via Factor B Reasoning

The 16PF primary factor B examines what is described as Reasoning. Factor B is
generally used as a quick measure of general mental ability due to its brevity and
moderately high relationship with measures of intelligence (Conn & Rieke, 1988). The
scale measures three areas of general reasoning ability, namely verbal, numerical, and
logical reasoning in a 15-item scale. The scale has an overall coefficient alpha of .80 and
has been shown to correlate with other general ability measures such as the Information
Inventory (r = .61) and the Culture Fair Intelligence Test (r = .51).

Measuring Leadership Effectiveness via an Annual Appraisal

The annual appraisal is a formal feedback discussion between a manager and an
executive about work performance. The purpose of the annual appraisal is to review an
executives performance against their business plans, sometimes referred to as
performance and development agreement (PDA). There are two focuses placed on the
annual appraisal: (1) What has been achieved, and (2) How has it been achieved.
Examples of the what focus include whether occupational health premiums have been
decreased, whether external charter standards have been met, whether the specific area
has delivered on key government initiatives, or whether tax revenue targets have been

12

achieved. The What section is always expressed in terms of both deliverable and
measurable targets.
The How focus of the PDA is on how leadership behaviors were adopted to achieve
business outputs. A significant part of the How aspect is an individuals ability to
demonstrate core leadership behaviours as outlined by the Australian Public Service
Commission (APSC) Leadership Capability Framework. This framework has five core
leadership capabilities: (1) Shapes strategic thinking (i.e., someone who inspires a sense
of purpose and direction; focuses strategically; harnesses information and opportunities;
and shows judgment, intelligence and commonsense); (2) Achieves results (i.e., someone
who builds organisational capability and responsiveness; marshals professional expertise;
ensures closure and delivers on intended results; and steers and implements changes and
deals with uncertainty); (3) Cultivates productive working relationships (i.e., someone who
nurtures internal and external relationships; values individual differences and diversity;
guides, mentors and develops people; and facilitates co-operation and partnerships); (4)
Communicates with influence (i.e., someone who communicates clearly; listens,
understands and adapts to an audience; and negotiates persuasively) and (5) Exemplifies
personal drive and integrity (i.e., someone who engages with risk and shows personal
courage; commits to action; displays resilience; and demonstrates self-awareness and a
commitment to personal development).
Both the What and How are initially rated on a five point scale (1 to 5) by the
participants direct manager. The meaning of the five ratings are:- (5) Exceptional Performance well beyond expectations, breaking new ground, producing outcomes of
considerable value to the organisation, often quite unanticipated; (4) Superior -

13

Achievement has been consistently high on the range of indicators, behaviours,

capabilities and any leadership role throughout the financial year; (3) Fully Effective Good and meritorious achievement. Has achieved standard as detailed in performance
agreement for both business outputs and behaviours; (2) Borderline Performance - has
slipped below standard as detailed in performance agreement for either business outputs
and or behaviours; and (1) Unsatisfactory - Continued failure to achieve expected
standard.
An important integrity aspect of the appraisal process is the fact that the What and
How ratings go through a series of checks before they are finalized. An individual
executive will first collate evidence about his or her performance, which is further
supplemented by the executives direct manager who also collates evidence on their
performance. The aim of this is to identify if specific business targets have been met.
During the appraisal discussion, both the executive and direct manager will review
business objectives and outcomes, review the executives behavior against the leadership
framework and agree on interim ratings for both the What and How components.
The interim ratings are then moderated by a third party committee, and generally,
individuals displaying superior or exceptional performances are highly scrutinized by both
department and corporate assurance processes. The importance of the corporate
assurance process becomes particularly evident when one considers that executive
directors are entitled to performance pay should their individual performance ratings reach
a certain level. The higher the performance ratings, the higher the performance based
pay they are entitled to receive.

14

Consistent with the validity of this rating procedure, the What and How ratings have
been shown to correlate significantly with internal peer 360 performance ratings, r = .52
(Rosete 2004). In addition, past research has supported the validity of this sort of rating
system (McEvoy and Beatty, 1989; Hogan, Curphy and Hogan, 1994; Management
Advisory Committee 2001; Corporate Leadership Council, 2002). For example, McEvoy
and Beatty (1989) examined the predictive validity of performance evaluations compared
to assessment center ratings and concluded that performance ratings were as effective as
assessment center data in forecasting performance seven years later.

Results
Descriptive Statistics and Internal Reliabilities

The SUEIT
The means, standard deviations, and internal reliabilities of the SUEIT subscales
were (1) emotional recognition and expression (in oneself) (M = 37.93; SD = 6.19; r =
.74); (2) emotions direct cognition (M = 32.42; SD = 6.33; r = .78); (3) understanding of
emotions external (M = 74.86; SD = 8.83; r = .69); (4) emotional management (M =
41.15; SD = 5.82; r = .73); and (5) emotional control (M = 32.48; SD = 4.36; r = .76).
The sample mean for the overall SUEIT EI was 218.83 (SD = 22.28). These results are
similar to the normative sample (M = 226.75; SD = 17.25) as reported for the workplace
SUEIT (Palmer and Stough, 2001).
In four of the five sub-scales, and for the overall SUEIT EI score, the mean
percentile scores fell within the average range, whereas the Emotions Direct Cognition
(EDC) subscale fell within the low range. The EDC results indicate that generally,

15

executives within this organization may tend to exhibit a very analytical or technically
oriented decision-making style where decisions at work are predominantly made on facts
and technical information (Palmer and Stough 2001).
Further coefficient alpha analysis was carried out on the intercorrelations among the
five sub-scales and total SUEIT EI score. The sizes of these correlations are generally
higher than that reported for the workplace SUEIT (Palmer and Stough, 2001). In
particular, there is a substantial correlation between the Understanding of Emotions
External and Emotional Management (r = .60) scales suggesting that the subscales share
approximately 36% of the variance in this data set. The highest inter correlation was
found between Emotional Control and Emotional Management (r = .70) suggesting that
these scales share approximately 49% of the variance in this data set.

The MSCEIT
The means, standard deviations, and reliabilities of each of the subscales is as
follows (1) perceiving emotion (M = .51; r = .76), (2) using emotion to facilitate thought
(M = .49; = .70), (3) understanding emotion (M = .55; r = .78), and (4) managing
emotions (M = .41; r = .75). Overall, the raw MSCEIT sample mean of .49 (SD = .05)
scored similarly to the general scoring normative sample (M = .51; SD = .06) as reported
for the MSCEIT users manual (Mayer, et al, 2002). In all four of the branch level scores
and for the overall EI score, the mean scores fell within the average range.
Further correlational analyses were carried out on the four branch levels and total
MSCEIT EI score. Perceiving emotion was correlated with understanding emotion r = .48,
understanding emotions r = .16 and managing emotion r = .26. Using emotion with

16

correlated with understanding emotion r = .19, and managing emotion r = .34. Finally,
understanding emotion was correlated with managing emotion r = .38. The sizes of these
correlations are generally lower than that reported in the users manual (Mayer, et al,
2002).

Personality & Cognitive Ability

The Sixteen Personality Factor (16PF) is a comprehensive multi-dimensional measure
of personality based on extensive factor analytic research. It is comprised of sixteen
primary personality factor scales and five secondary or global bipolar factors. These
include introversion versus extroversion (M = 4.2, S.D. = 2.34), low versus high anxiety
(M = 5.94, S.D. = 1.92), tough-mindedness versus receptivity (M = 5.00, S.D. = 2.02),
independence versus accommodation (M = 4.85, S.D. = 1.93), and low versus high selfcontrol (M = 4.41, S.D. = 1.54). The five global factors are comparable to the Big Five
dimensions of personality (Conn & Rieke, 1998).
For both the primary scales and global factors, mean scores between 4 and 6
indicate normal average range of functioning. Overall, results for the sixteen primary
scales tended to be between 4 and 6 with minimum and maximum scores generally being
between 1 and 10. These results indicate that the sample was approximately normal, with
proportional numbers of executives exhibiting the breadth of personality dimensions one
would be expected in any normal population.
The exception was the Reasoning scale which had a mean of 7.5 (SD = 1.5). The
relatively high mean in this study suggests that as a group, the participants are generally

17

more intelligent, brighter, and have higher general mental capacity than the normal
population, operating at the 84 percentile.

Correlational Analyses

Leadership Effectiveness
Pearson correlation coefficients were used to examine any relationship that may exist
between the SUEIT EI measure and leadership effectiveness as measured through the
performance management ratings. No significant correlations were found between total
SUEIT EI scores and any of the performance management measures. As can be seen in
Table 1, only the sub-branch factors of emotional management and emotional control
correlated significantly with one aspect of the performance rating, namely the how rating.
In contrast, the total EI score of the MSCEIT correlated with both what and how
ratings (see Table 1), indicating that higher ability EI was associated with better
performance. In addition, Table 1 illustrates that all of the EI subscales correlated with
how performance ratings, and two subscales, perceiving and managing emotions
correlated with what performance ratings.
Pearson correlation coefficients were also used to examine relationships between
personality factors, reasoning and leadership effectiveness as measured through the
performance management ratings. There were no significant correlations between the
five global factors and reasoning with performance ratings. At the primary factor
subscales there were some correlations. In particular, low levels of vigilance, high levels
of abstractedness, low levels of privateness, and high levels of openness to change with

18

higher what performance ratings. Low levels of vigilance were also associate with high
how performance ratings.

Insert Table 1 about here

EI, Personality and Reasoning Ability

We next investigated the extent that the self-report and ability based EI measures were
distinctive from personality and reasoning ability. Concerning self-reported EI (the
SUEIT), table 2 shows moderate to strong correlations between all but one of the scales.
For the MSCEIT EI measure, weak to moderate correlations were found between total EI
score and anxiety. The self-report measure of EI appears to be more highly correlated
with personality than the ability measure. In contrast, reasoning seems to be more highly
correlated with the MSCEIT understanding emotion subscale than it is with the other
scales.

Insert Table 2 about here

We carried out a series of hierarchical regression analyses in order to determine if

ability and self report measures of emotional intelligence related to performance after
controlling for personality and reasoning ability. Reasoning ability was entered in Step 1,
the five global personality factors on Step 2 and either ability EI or self-reported EI were
entered in step 3.

19

The results of the first regression analysis with the What ratings showed that neither
reasoning ability (Step 1, r = .01, R2change = .00, FChange = .00), the five global personality
factors of the 16PF (Step 2, r = .29 (R2change = .08, FChange = 1.89), or the self-report
measure of EI (Step 3, r = .31 (R2change = .00, FChange = .01), had a significant effect. In
contrast, the ability measure of EI produced significant improvement in the model (step 3,
r .37, R2change = .05, FChange = 6.22) accounting for approximately 5% of the variance in the
What perspective of the performance management system. The ability measure of EI thus
related to variance over and above that involving reasoning and personality.
The How rating analyses produced similar results. Essentially, the self-reported EI
measure did not predict variance over and above personality and reasoning, p > .2. In
contrast, the ability measure of EI significantly improved the fit of the model (Step 3b, r =
53, R2change =.22, FChange = 31.99). Thus, the ability measure of EI was able to explain an
additional 22% of the variance above both reasoning ability and personality
characteristics.
In addition to controlling for the global dimensions of personality (i.e., the Big Five),
we conducted analyses that controlled for the individual personality subscales (see Table
1). We focused on the subscales of EI that had been identified as significant. Hierarchical
regression analysis was conducted for the What performance rating with reasoning ability
at Step 1, the significant 16PF primary factors at Step 2 and any significant EI measure at
the branch level at Step 3. No significant self-report EI effects were found at step 3,
suggesting these measures do not add incremental value over and above personality and
reasoning. In contrast, we found that even when controlling for reasoning in step 1 and
the four significant 16PF subscales in step 2 (see Table 1), the MSCEIT subscales

20

(perceiving r = .26, p < .01 and managing emotions r = .08, p > .4) significantly
improved the fit of the model (Step 3, r = .46, R2change = .08, FChange = 5.10) and
accounted for 8% of the variance.
Finally, we ran hierarchical regressions that focused on the How ratings, and again
found that the self-reported EI measures did not contribute significantly to the model at
Step 3. In contrast, the ability based model (perceiving emotion, r = .40, p < .01, using
emotion r = .06, p > .5, understanding emotion r = .20, p < .05, and managing emotion r
= .05, p > .6) contributed significantly (Step 3, r = .54, R2change = .30, FChange = 9.52),
indicating that it predicted 30% of the variance in the How variance even after controlling
for reasoning and personality.

Discussion
The findings suggest that the executives who achieve the best business outcomes
are the ones who 1) score higher on an EI ability test, 2) manifest certain personality
characteristics (e.g., high openness, low vigilance and privateness), and 3) have higher
self-reported ability to manage emotions. Importantly, the ability measure of emotional
intelligence was able to predict effective leadership over and above already well
established workplace measures such as reasoning ability and personality. In contrast,
self-reported measures of emotional intelligence had little to offer over and above wellestablished measures of personality and reasoning. These results may have important
implications on how we engage in performance management, and select and develop
executives.

21

Self-reported EI and personality

The SUEITs emotional control and emotion management subscale was related to
higher How ratings, explaining between 3 to 5 % of the variance. Personality also
explained approximately 3 to 6% of the variance. Specifically, the most effective
executives were those who showed low levels of vigilance and privateness, and high levels
in abstractness and openness to change.
The SUIET tended to correlate moderately with personality, especially with
extraversion, anxiety and independence. Importantly, the SUEIT failed to predict
leadership effectiveness over and above personality. These results are consistent with
evidence that suggests self-report measures of EI overlap with well-established personality
traits or behaviors (Ciarrochi, Chan & Caputi, 2000; Ciarrochi, Chan, Caputi, and Roberts,
2001; MacCann, Roberts, Mathews, and Zeidner, 2004).
On a practical level, these results suggest that self-report EI may not add
incremental value over personality in predicting leadership success. Thus, there appears
to be little empirical justification, to date, for the use of self-report EI measures for the
selection of executives (Meidner, Matthew & Roberts, 2004; Landy, 2005).
One major difficulty with the overlap between self-reported EI and personality is that
one can always question whether a result involving self-reported EI is really just a
replication of a previous result using a similar personality measure. For example, research
has clearly demonstrated a link between self-esteem and mental health (Ciarrochi, Scott,
Deane & Heaven, 2003). Thus if an EI measure correlates substantially with self-esteem
and mental health, is the EI test measuring something new, or is it replicating previous
self-esteem effects?

22

Our findings do not imply, however, that self-report EI measures have no practical
use. Both self-report EI and personality explained the same level of variance in leadership
success (about 4%). Thus, it may be that in some instances one might use the self-report
EI measure instead of a personality measure. Specifically, one might use the EI measure
because it provides more useable feedback to managers than personality measures. For
example, it might be more instructive to provide managers feedback on emotion
management skill (from the self-report EI measure) than on their level of privateness
(from the personality measure).
Future research is needed to evaluate whether self-report EI does have some added
practical value over personality measures. For example, do leaders improve their
performance more when provided with EI feedback compared to when they are provided
with personality feedback? Are self-report EI measures more useful for guiding
intervention programs?

Ability based measures of EI & Reasoning

We found that the executive were on average in the top 84 percentile in terms of
cognitive intelligence. Variations at this level did not predict performance. Perhaps once
people obtain a certain level of cognitive intelligence, having higher intelligence makes
little difference (Sternberg & Vroom, 2002).
In contrast, ability-based EI demonstrated a strong relationship with job
performance. An analysis of the individual subscales revealed that the most important
scales were perceiving emotions and managing emotions. These scales related to the
extent that executives achieved business outcomes, or the What of performance. In

23

relation to the How of performance, the capacity to perceive emotion, use emotions,
understand emotion and manage emotion all related to how effectively an executive
achieved their business outcomes.
George (2000) suggests that leadership involves the development of a collective
sense of goals, instilling in others both knowledge and appreciation of certain work
activities, and generating a sense of excitement, confidence, and trust. One could argue
that these elements all require a leader to use their emotions to enhance how they
communicate to subordinates, which in turn would assist them to effectively carry out the
role of a leader. Not surprisingly, we find all four elements are critical in enabling a leader
to meet business outcomes. The ability to perceive emotions provides an awareness of
emotions and the ability to accurately read others emotions, especially subordinates.
Using emotions provides a means to generate ideas within a team. Understanding
emotions offers insights into what motivates people and others points of view. Finally,
managing emotions may allow a leader to stay open to their emotions, extract valuable
information, and use them constructively to enhance team and individual performance
(Caruso & Salovey, 2004). All four elements may enhance a leaders ability to
communicate.
Of particular interest was the branch of perceiving emotion which in both the What
and How rating had the highest effect sizes. A leader who is skilled in perceiving emotion
would be described as someone who knows what people feel, reads people accurately, is
good at recognising their own feelings and can express their feelings appropriately
(Caruso & Salovey, 2004). These skills are particularly important as it allows a leader to

24

accurately capture important social data around them, in particular the ability to read
between the lines.
It is important to put the effect sizes we obtained here in context. Our effect size for
ability EI was in the order of .26 for the What outcomes and .50 for the How outcomes.
This effect level is considered to be large by many reviews (Barrick & Mount, 1991;
Schmidt & Hunter, 1998). For example, research has consistently shown that personality
traits such as conscientiousness are relate to most job performance criteria and typically
have an effect size in the order of .2 (Barrick & Mount, 1991). Schmidt and Hunter (1998)
have also shown that in analyzing 85 years of research in personnel selection that General
Mental Ability (GMA) tests have one of the highest effect sizes at .51.
The present results have potential implications for how we manage performance,
selection and development of executives. In terms of performance management, it is
important for an executive to both deliver outputs (What performance), and also deal
effectively with colleagues and staff (the How of performance - MAC 2001). It may be
common for technical leaders to have What but not How skills. For example, a technical
specialist may perform complex tasks tenaciously and manage to produce business
outputs, but may be ineffective at managing his or her subordinates, leading to issues of
staff turnover and underperformance. The results of this study show that emotional
intelligence may be useful in identifying who is and is not likely to deal effectively with
colleagues and staff. Furthermore, they also show which leaders are likely to achieve
business outcomes. Organizations who wish to maximize their ability to meet business
outcomes therefore have the choice to either recruit for these abilities or further develop
these abilities in their top executives.

25

In conclusion, these results make significant contributions to the study of emotional

intelligence in a work setting, but leave a number of important questions unanswered.
First, research is needed to evaluate why EI is related to performance. Does it relate to
interpersonal effectiveness on the job, as we suggested above? Or does it relate to other
aspects of management, such as inspiring enthusiasm, or managing stressful jobs?
Second, given the correlational nature of the present study, research is needed to
evaluate whether EI predicts future performance, or merely co-occurs with performance.
Third, research is needed to examine whether these results generalize across different
occupational settings? Is EI more important in interpersonal settings, compared to
settings that require little intrapersonal contact (e.g., some information technology jobs)?
Regardless of the answers to these and many yet unanswered questions, this study
suggests that ability based measures of EI have interesting applications within an applied
setting.

26

References
Ashkanasy, N. M., E. Trevor-Roberts, et al. (2000). Leadership attributes and cultural
values in Australia and New Zealand compared: An initial report based on globe data.

International Journal of Organisational Behaviour 2(3): 37-44.

Bagby, R. M., G. J. Taylor., J. D. A. Paker, (1994 a). The twenty-tem toronto alexithymia
scale-I. Item selection and cross-validation of the factor structure. Journal of

Psychosomatic Research 38(1): 23-32.

Bagby, R. M., G. J. Taylor., J. D. A. Paker, (1994 b). The twenty-item toronto alexithymia
scale-II. Convergent, discriminant, and concurrent validity. Journal of Psychosomatic

Research 38(1): 33-40.

Barling, J., Slater, F. and Kelloway E.K. (2000), Transformational leadership and emotional
intelligence: an exploratory study, Leadership & Organization Development Journal,
21(3), pp. 157-161.
Bar-On, R. (1997), Bar-On Emotional Quotient Inventory: Users Manual, Multi-Health
Systems, New York, NY.
Barrick, M. R. and M. K. Mount (1991). The Big Five personality dimensions and job
performance: A meta-analysis. Personnel Psychology 44(1): 1-26.
Bass, B.M. and Avolio, B.J. (1990), Transformational Leadership Development: Manual for

the Multifactor Leadership Questionnaire, Consulting Psychologist Press, Palo Alto,

CA.
Bass, B.M. and Avolio, B.J. (1994), Improving Organizational Effectiveness through

Transformational Leadership, Sage Publications, Thousand Oaks, CA.

Caruso, D. R. and P. Salovey (2004). The Emotionally Intelligent Manager. San Francisco,

27

Jossey-Bass.
Ciarrochi, J.V., Chan, A.Y.C. and Caputi P. (2000), A critical evaluation of the emotional
intelligence construct, Personality and Individual differences 28, pp. 539-561.
Ciarrochi, J.V., Chan, A.Y.C., Caputi P. and Roberts, R. (2001), Measuring Emotional
Intelligence. In Ciarrochi, Forgas and Mayer (Eds), Emotional Intelligence in

Everyday Life: A Scientific Inquiry, Psychology Press, Philadelpha, USA.

Ciarrochi, J., Chan, A. Y., & Bajgar, J. (2001). Measuring emotional intelligence in
adolescents. Personality & Individual Differences, 31(7), 1105-1119.
Ciarrochi, J., & Scott, G. (2005). Social and Emotional competence and well-being: a

longitudinal study. submitted for publication.

Ciarrochi, J., Dean, F. P., & Anderson, S. (2002). Emotional intelligence moderates the
relationship between stress and mental health. Personality & Individual Differences,

32(2), 197-209)
Ciarrochi, J., Scott, G., Deane, F.P., and Heaven, C.L. (2003), Relations between social
and emotional competence and mental health: a construct validation study,

Personality and Individual Differences, 35, pp, 1947-1963.

Con, S.R., and Rieke, M.L. (1998), The 16PF Fifth Edition Technical Manual, Institute for
Personality and Ability Testing, Inc, Champaign, Illinos.
Corporate Leadership Council (2002). Closing the Performance Gap: Driving Business

Results Through Performance Management. Corporate Executive Board, Washington.

Davies, M., Stankov, L., and Roberts, R. D. (1998) Emotional intelligence: In search of an
elusive construct. Journal of Personality & Social Psychology, 75(4), 989-1015.
Gardner, L. and Stough, C. (2002), Examining the relationship between leadership and

28

emotional intelligence in senior level managers, Leadership & Organization

Development Journal 23(2), pp. 68-78.

George, J. M. (2000). Emotions and leadership: The role of emotional intelligence. Human
Relations 53: 1027-1055.
Goleman, D. (1995), Emotional Intelligence, Bantam, New York.
Goleman, D. (1998), Working with Emotional Intelligence, Bloomsbury Publishing, London.
Goleman, D., Boyatzis, R., and McKee, A (2002). Primal Leadership: Realizing the power of

EI. Boston, Harvard Business School Press.

Higgs, M. (2003). How can we make sense of leadership in the 21st century? Leadership

and Organization Development Journal 24(5), 273-284.

Hogan, R., G. Curphy, Hogan. (1994). What we know about leadership: Effectiveness and
personality. American Psychologist 49(6): 493-504.
Landy, F. J. (2005). Some historical and scientific issues related to research on emotional
intelligence. Journal of Organizational Behavior 26, 411-424.
Lopes, Ct, Grewal, Kadis, Gall and Salovey (2003). Evidence that Emotional Intelligence
is Related to Job Performance, Interpersonal Facilitation, Affect and Attitudes at
Work, and Leadership Potential. Manuscript submit for publication.
MacCann, C., Roberts, R.D., Mathews, G. and Zeidner, M. (2004), Consensus scoring and
empirical option weighting of performance-based emotional intelligence tests,

Personality and Individual Differences.

Management Advisory Committee. (2001). Performance Management in the Australian

Public Service. Canberra, Commonwealth of Australia.

McEvoy, G.M., and Beatty, R.W. (1989). Assessment centers and subordinate appraisals of

29

managers. Personnel Psychology, 42(1), p 37.

McKee, A. (2005). Leadership, Passion, and Power: EI, Resonance and Renewal. 5th
Annual Emotional Intelligence Conference, Netherlands, 12 14 June.
Mayer, J.D., and Salovey, P. (1997), What is emotional intelligence? In Salovey, P. and
Sluyter, D.J. (Eds), Emotional Development and Emotional Intelligence: Implications

foe educators, Basic Books, New York, NY.

Mayer, J. D., Ed. (2001). A field guide to emotional intelligence. Emotional intelligence in
everyday life: A scientific inquiry. Philadelphia, Psychology Press.
Mayer, J. D. (in press). A new field guide to emotional intelligence. Emotional Intelligence
and Everyday Life. J. Ciarrochi, J. P. Forgas and J. D. Mayer. New York, Psychology
Press.
Mayer, J. D., Caruso, D. R., and Salovey, P. (1999). Emotional intelligence meets
traditional standards for an intelligence, Intelligence, 27, 267-298.
Mayer, Caruso and Salovey, 2000. Selecting a measure of EI: The case of ability scales. In
R. BarOn and J.D.A Parker (Eds). The handbook of EI (pp 320 342). San Francisco:
Jossey Bass.
Mayer, J.D., Salovey, P. and Caruso, D.R. (2002), Mayer-Salovey-Caruso Emotional

Intelligence Test, Multi-Health Systems, Toronto.

Mayer, J.D., Salovey, P., Caruso, D.R., and Sitarenios, G. (2003), Measuring emotional
intelligence with the MSCEIT V2.0, Emotion 3(1), pp. 97 - 105.
Mayer, J.D., Salovey P and Caruso, D.R. (2004). Emotional intelligence: Theory, Findings
and Implications. Psychology Inquiry, 60, 197-215.
Meyer, G. J., S. E. Finn, et al. (2001). Psychological testing and psychological assessment:

30

A review of evidence and issues. American Psychologist 56(2): 128-165.

Meidner, M., Matthew, G., and R. D. Roberts, (2004). EI in the workplace: A critical
review, Applied Psychology: An International Review, 53(3), 371-399.
McShane, S. L., and Von Glinow, M.A. (2000), Organizational Behavior, McGraw-Hill Inc,
USA.
Palmer, B. R. (2003). An Analysis of the Relationships Between Various Models and
Measures of Emotional Intelligence. Centre for Neuropsychology, School of
Biophysical Sciences and Electrical Engineering. Victoria, Swinburne Unversity of
Technology.
Palmer, B., Walls, M., Burgess, Z. and Stough, C. (2001), Emotional intelligence and
effective leadership, Leadership & Organization Development Journal 22(1), pp. 5 10.
Palmer, B. and Stough, C. (2001), Workplace SUEIT: Swinburne University Emotional

Intelligence Test Manual, Organizational psychology Research Unit, Swinburne

University, AU.
Palmer, B.R., Gardner, L and Stough, C. (2003), The relationship between emotional

intelligence, personality and effective leadership, 5th Australian Industrial &

Organizational Psychology Conference, Melbourne.
Palmer, B. R., G. Gignac, R. Manocha and C. Stough (in press). A psycometric evaluation
of the Mayer-Salovey-Caruso Emotional Intelligence test version 2.0. Intelligence.
Prati, L. M., C. Douglas, et al. (2003). Emotional intelligence, leadership effectiveness, and
team outcomes. International Journal of Organizational Analysis 11(1): 21-40.
Roberts, R., Zeidner, M., and Matthews, G. M. 2001. Does emotional intelligence meet

31

traditional standards for an intelligence? Some new data and conclusions. Emotion,
1, pp 196 231.
Rosete, D. (2004). A Leaders profile - What attributes make an effective leader? 4th
International Test Users' Conference, Melbourne, Australia.
Rosete, D. and J. Ciarrochi (in press). Emotional intelligence and its relationship to
workplace performance outcomes of leadership effectiveness. Leadership and

Organization Development Journal.

Salovey, P and Mayer, J.D. (1990). Emotional Intelligence. Imagination, Cognition and

Personality, 9, 185-211.
Salovey, P., Mayer, J., Goldman, S., Turvey, C. & Palfai, T. (1995). Emotional attention,
clarity and repair: Exploring emotional intelligence using the Trait Meta-Mood Scale.
In J.W. Pennebaker, Emotion, disclosure, and health (pp. 125-154) Washington, DC:
American Psychological Association.
Schutte, N. S., J. M. Malouff, Hall, Haggerty, Cooper, Golden and Dornhein, (1997).
Development and validation of a measure of emotional intelligence. Personality and

Individual Differences 25: 167-177.

Shmidt, F. L. and J. E. Hunter (1998). The validity and utility of selection methods in
personnel psychology: Practical and theoretical implications of 85 years of research
findings. Psychological Bulletin 124(2): 262-274.
Sternberg, R. J., & Vroom, V. H. (2002). The person versus the situation in leadership.

Leadership Quarterly, 13, 301-323

32

Table 1 Correlations Coefficients between the MSCEIT, SUEIT, Personality and Reasoning Scales
with Performance Management Ratings
Performance

Performance How

What Rating

Rating

Total EI Score

.26**

.50**

Perceiving Emotion

.27*

.48**

Using Emotion

.18

.30**

Understanding Emotion

.11

.29**

Managing Emotions

.23*

.26**

SUEIT EI

.09

.15

Emotional Recognition &

.06

.02

Understanding Others Emotions

-.01

-.02

Emotions Direct Cognition

.00

.08

Emotional Management

.09

.23*

Emotional Control

.17*

.21*

Vigilance

-.21*

-.18*

Abstractness

.22*

.08

Privateness

-.18*

-.14

Openness to change

.25*

.18

Reasoning

-.00

.12

Ability Based EI: MSCEIT

Self-reported EI

Expression

Personality (16PF)

p < .05; ** p < .01

33

Table 2
Personality

Correlations Coefficients between the MSCEIT, SUEIT, and Personality

SUEIT

ER&E

EDC

UEE

EM

EC

EI

MSCEIT

PE

UFE

UnE

ME

EI

Extraversion

0.52**

0.57**

0.34**

0.27**

0.43**

0.21*

0.15

0.06

0.17

0.10

0.22*

Anxiety

0.35**

-0.18*

0.00

-0.16

0.45**

0.44**

-0.28**

-0.22*

-0.24

-0.06

0.28**

Tough Mindedness

0.29**

0.34**

0.36**

-0.18*

-0.05

-0.08

-0.07

0.05

-0.04

-0.17

-0.14

Independence

0.43**

0.42**

0.24**

0.29

0.32**

0.24**

0.08

-0.07

0.10

0.08

0.18

Self Control

-0.06

0.22**

-0.22*

-0.04

0.13

0.14

0.00

0.07

-0.03

-0.20*

0.00

Reasoning

0.02

0.10

-0.02

-0.07

-0.09

0.15

0.11

0.14

-0.07

0.31*

-0.03

p < .05; ** p < .01

ER&E = emotional recognition and expression; EDC = emotions direct cognition; UEE =
understanding of emotions external; EM = emotional management; EC = emotional control;
PE = perceiving emotion; UE = using emotion; UnE = understanding emotion and ME =
managing emotion

34