analytic methodology employed by Bar-On (1997a), and his interpretation of the results that render theSince Salovey and Mayers conceptualisation of emotional intelligence (EI; Salovey & Mayer,dimensional structure of the EQ-i unclear. In contrast to claims by Bar-On, in the present study a series ofexploratory and conrmatory factor analyses found evidence for a general factor of emotional intelligenceand six primary factors. Dierences between the results reported by Bar-On (1997a) and those of the cur-rent study are attributed largely to the more appropriate factor analytic methodology employed. Implica-tions and directions for future research are discussed.# 2003 Elsevier Ltd. All rights reserved.

Keywords: Emotional Intelligence; Emotional Competencies; Bar-On EQ-i; Factor structure

1. Introduction

1.1. Models and measures of emotional intelligencegeneral population sample

Benjamin R. Palmera, Ramesh Manochab, Gilles Gignaca, Con Stougha,*aOrganisational Psychology Research Unit, Neuropsychology Laboratory, School of Biophysical Sciences and Electrical

Engineering, Swinburne University of Technology, PO Box, 218, Hawthorn, Victoria, 3122, AustraliabUniversity of New South Wales, Australia

Received 10 October 2001; received in revised form 16 September 2002; accepted 29 October 2002

Abstract

It has been claimed that the dimensional structure of the Bar-On Emotional Quotient Inventory (EQ-i)(Bar-On, 1997a) represents a hierarchical model of emotional and social intelligence describing a generalfactor, ve-second order factors and 15 primary factors. However, there are several anomalies in the factorExamining the factor structure of the Bar-OnEmotional Quotient Inventory with an Australian

Personality and Individual Differences 35 (2003) 11911210

www.elsevier.com/locate/paid0191-8869/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.doi:10.1016/S0191-8869(02)00328-8

* Corresponding author. Tel.: +61-3-9214-8167; fax: +61-3-9214-5230.

E-mail address: cstough@swin.edu.au (C. Stough).1990), a number of dierent models and measures have been developed (e.g. Bar-On, 1997a;

Cooper & Sawaf, 1997; Goleman, 1995; Mayer & Salovey, 1997). These alternative models andmeasures have been compared according to their theoretical structure (Mayer, Salovey, & Car-uso, 2000; McCrae, 2000), and according to the way they measure EI (Mayer, Caruso, & Salovey,2000; Petrides & Furnham, 2000). Models of EI have been placed into two general categories,ability and mixed (or personality) models of EI. Ability models have been identied as thosethat dene EI as intelligence in the traditional sense (e.g. Mayer & Salovey, 1997). That is, as aconceptually related set of mental abilities to do with emotions and the processing of emotionalinformation, that are apart of, and contribute to, logical thought and intelligence in general. Incomparison, mixed models of EI (e.g. Bar-On, 1997a) have been identied as those that dene EIas a mixture of emotion-related competencies, personality traits and dispositions. Measures of EIsimilarly fall into two categories, self-report measures of EI, and performance-based (objective)measures. Self-report measures pertaining to ability models of EI (e.g. Trait-Meta Mood Scale,TMMS; Salovey, Mayer, Goldman, Turvey, & Palfai, 1995) are purported to assess individualsbeliefs about emotional abilities rather than their actual capacity (Mayer, Caruso et al., 2000).Self-report measures pertaining to mixed models (e.g. the Bar-On Emotional Quotient Inventory,Bar-On EQ-i; Bar-On, 1997a) have been described as embedded within the personality frame-work, and to assess cross-situational consistencies in behaviour (Petrides & Furnham, 2000).Performance-based measures of EI pertain to ability models (e.g. Mayer, Salovey, Caruso Emo-tional Intelligence Test, MSCEIT; Mayer, Salovey, & Caruso, 1999), and involve a series ofemotion-related questions for which there are more and less correct answers according to con-sensual responses. The present paper examines the dimensional structure of one of the pre-dominant mixed model self-report measures of EI, the Bar-On EQ-i, (Bar-On, 1997a), in anAustralian general population sample.

1.2. Bar-Ons (1997a) model of emotional intelligence

Bar-Ons model of EI (1997a) involves an array of personal, emotional, and social abilities andskills thus constituting a mixed model. While Bar-On places this model under the banner of EI, itis a somewhat broader construct that he more generically refers to as . . .emotional and socialintelligence (Bar-On 2000, p. 363). Bar-On purports to have identied 15 determinants of suc-cessful emotional functioning and positive psychological well-being from a review of the mentalhealth literature, which have been operationally dened and conceptualised as the 15 componentsof his model. These components include: Emotional Self-Awareness (ES), the ability to recogniseand to understand ones feelings; Assertiveness (AS), the ability to express feelings, beliefs andthoughts, and to defend ones rights in a non-destructive manner; Self-Regard (SR), the ability torespect and accept oneself; Self-Actualisation (SA), the ability to realise ones potential capacities;Independence (IN), the ability to be self-directed and self-controlled in ones thinking and actionsand to be free of emotional dependency; Empathy (EM), the ability to be aware of, to under-stand, and to appreciate the feelings of others; Interpersonal Relationship (IR), the ability toestablish and maintain mutually satisfying relationships; Social Responsibility (RE), the ability todemonstrate oneself as a cooperative contributing, and constructive member of ones socialgroup; Problem Solving (PS), the ability to identify and dene problems as well as to generateand implement potentially eective solutions; Reality Testing (RT), the ability to assess the cor-

1192 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210respondence between what is experienced and what objectively exists; Flexibility (FL), the ability

to adjust ones emotions, thoughts, and behaviour to changing situations and conditions; StressTolerance (ST), the ability to withstand adverse events and stressful situations; Impulse Control(IC), the ability to resist or delay an impulse, drive or temptation to act; Happiness (HA), theability to feel satised with ones life, to enjoy oneself and others, and to have fun; Optimism(OP), the ability to look at the brighter side of life and to maintain a positive attitude. These 15components of Bar-Ons model are described in greater detail in the EQ-i Technical Manual (Bar-On, 1997a).Within Bar-Ons (1997a) model, the 15 components are theoretically arranged into ve broader

or major conceptual components. These include; Intrapersonal Emotional intelligence (RAeq),representing abilities, capabilities, competencies and skills pertaining to the inner self, i.e. the ES,AS, SR, SA and IN components; Interpersonal Emotional intelligence (EReq), representinginterpersonal skills and functioning i.e. EM, IR, RE; Adaptability Emotional Intelligence(ADeq), representing how successfully one is able to cope with environmental demands by eec-tively sizing-up and dealing with problematic situations, comprising PS, RT and FL; StressManagement Emotional Intelligence (SMeq), representing the ability to manage and cope eec-tively with stress comprising the ST and IC components; and General Mood Emotional Intelli-gence (GMeq), representing the ability to enjoy life and maintain a positive disposition whichcomprises the HA and OP components. These ve major components of EI are theoreticallyrelated to a general factor of EI thus constituting a hierarchical model comprising overall EI, ve-composite components, and 15 specic components at the bottom of the hierarchy. Within thismodel, EI is dened as . . .an array of non-cognitive capabilities, competencies, and skills thatinuence ones ability to succeed in coping with environmental demands and pressures (Bar-On,1997a, p. 16). The 15 components of the model are described as non-cognitive variables that. . .resemble personality factors (Bar-On, 1997b, p. 6). It is also proposed that the componentsof the model develop over time (with age), change throughout life, and can be improved throughtraining and remedial programs.

1.3. The Bar-On Emotional Quotient Inventory

The Bar-On EQ-i (1997a) has been designed to assess Bar-Ons (1997a) model of EI. Consistentwith Bar-Ons proposed theoretical structure of EI, the EQ-i comprises 15 sub-scales pertainingto the 15 components of the model, which render 15 sub-scale scores, ve EQ composite scalescores and an overall or total EQ score. As with other self-report measures of EI, the EQ-i isdescribed to provide an index of cross-situational consistencies in (emotionally and socially)competent behaviour and as such, provides an estimate of an individuals EI (Bar-On, 2000).Psychometric analyses of the EQ-i reported in the technical manual (Bar-On, 1997a), indicate

that it has good internal reliability and testretest reliability. Across seven population samples, the 15sub-scales are reported to have average-to-high internal consistency coecients with Cronbachalphas ranging from =0.69 for RE to =0.86 for SR. Similarly, with a South African sample(n=44) the average stability coecient of the 15 sub-scales after a 1-month period was found to ber=0.85, and with a smaller sub-set of this sample (n=27) was found to be r=0.75 after a4-month period. A large number of correlation studies are also reported in the technical manual(Bar-On, 1997a), in support of the validity of the EQ-i as a measure of the ability to succeed in

B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210 1193coping with environmental demands and pressures, and psychological well-being. For example;

total EQ-i scores have been shown to correlate positively with measures of emotional stability(e.g. r=0.72 with the Emotional Stability factor of the Sixteen Personality Factor Questionnaire,16PF; Cattell, Cattell, & Cattell, 1993); overall life satisfaction (e.g. r=0.41 with the KirkcaldyQuality of Life Questionnaire, Kirkcaldy 1995); acculturation (e.g. r=0.34 with the ShortAcculturation Scale, SAS; Marin, Sabogal, Marin, Otero-Sabogal, & Perex-Stable, 1987); attri-bution style (e.g. r=0.37, with the Attributional Style Questionnaire, ASQ; Peterson, Semmel,von Baeyer, Abramson, Metalsky, & Seligman, 1982); and proxy measures of job performanceand satisfaction (e.g. r=0.51 with the Sense of Competence Questionnaire, SCQ; Wagner &Morese, 1975). Furthermore, total EQ-i scores correlate negatively with measures of poor emo-tional health (e.g. r=0.85 with the Ninety Symptom Check List, SCL-90; Derogatis, 1973); anddepression (e.g. r=0.56 with the Beck Depression Inventory, BDI; Beck & Steer, 1987). Overallthese results provide preliminary evidence for the construct validity of the EQ-i, however, as withmost existing measures of EI, the validity of the EQ-i needs to be further established by inde-pendent research and with larger samples (Bar-On 2000).One question that is currently relevant to the EQ-i is its discriminant validity from per-

sonality traits such as neuroticism and general aect (Mayer, Caruso et al. 2000; Newsome,Day, & Catano, 2000). A recent study by Newsome et al. (2000) reports correlations betweenthe ve personality factors of the 16PF (Cattell et al., 1993) and the ve EQ-i compositescores ranging from r=0.1 to r=0.77. The highest correlation in this study was between thetotal EQ scale score of the EQ-i and the Anxiety factor of the 16PF (r=0.77) leading theseresearchers to conclude . . .that the EQ-i is largely a measure of neuroticism. . . (Newsome et al.,2000; p. 1014). Dawda and Hart (2000) have also demonstrated considerable overlap between theEQ-i and personality. These authors found the total EQ scale score of the EQ-i correlated withthe NEO FFI (Costa & McCrae, 1992): for males Neuroticism (N) (r=0.62), Extraversion (E)(r=0.52), Openness (O) (r=0.12), Agreeableness (A) (r=0.43), and Conscientiousness (C)(r=0.51); and for females N (r=0.72), E (r=0.56), O (r=0.17), A (r=0.43), and C (r=0.33).Given this overlap with personality, it is possible that the EQ-i may be predicting theoreticallyrelated life criteria in preliminary validity studies (e.g. life satisfaction, workplace performance,psychological well-being) because the EQ-i is measuring personality traits and dispositionsknown to account for these important human values. The discriminant validity of the EQ-ifrom personality traits and dispositions, and indeed whether it accounts for variance in lifesatisfaction, job performance and psychological well-being not accounted for by well establishedpersonality traits has not yet been empirically substantiated and needs be addressed by futureresearch.While there are issues concerning the discriminant and predictive validity of the EQ-i, this scale

is one of the most advanced self-report measures of EI currently available. The EQ-i has beentranslated into 22 languages and normative data has been established in more than 15 countries(Bar-On, 2000). Furthermore, the EQ-i shows a meaningful pattern of convergent validity withmeasures of psychological well-being and alexithymia (Dawda & Hart, 2000). In addition, theEQ-i is a well-constructed measure, involving indices of social response bias, and response valid-ity indicators (i.e. omission rate and an inconsistency index), that are purported to increase theaccuracy of test scores. Finally, the EQ-i is a relatively brief and easy scale to administer andinterpret. Results are computer generated, interpreted by the test publisher Multi-Heath Systems

1194 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210(MHS), and are represented in layperson terms numerically, graphically and textually in feedback

reports. Given the potential utility of the EQ-i, further validation with more diverse and largerpopulation samples is essential (Bar-On, 2000).

1.4. Dimensional structure of the Bar-On EQ-i

While validation of the discriminant and predictive validity of the EQ-i is needed, it could beargued that further validation of its dimensional structure is also required. To-date, Bar-On(1997a, 2000) is the only researcher to have examined the factor structure of the EQ-i. Empiricalsupport is claimed, with a large representative population sample (n=3831) through bothexploratory and conrmatory factor analyses, that the hierarchical factor structure of Bar-Onsmodel of EI is measured by the 15 component sub-scales of the EQ-i. However, there are someanomalies in the interpretation of the results and factor solutions chosen to represent the datathat need to be addressed.Bar-On (1997a), performed a principal components factor analysis of the EQ-i and reported

using the eigen value-greater-than-one, and the scree criteria to determine the correct number offactors in the data set. Neither the number of factors with eigen values greater than one, nor thenumber of factors suggested by the scree plot were reported. Rather, Bar-On reported that therewere 13 factors in the data set and that the variance revealed by each of the factors was, in order,23.1, 4.8, 3.6, 2.8, 2.6, 2.0, 1.7, 1.5, 1.3, 1.3, 1.2, 1.1 and 1. With large sample sizes of 300 or more,factor loadings as small as .15 (indicating that 2.25% of the variance is accounted for by thefactor) are typically signicant at the P

Bar-On (1997a) labelled the 13-factors that emerged from the data set by examining the highestloading items on each factor (factor loadings of 0.4 and higher across and within factors). Therewere some dierences between Bar-Ons (1997a) theoretical structure of EI, and that whichemerged from the EQ-i as a result of the exploratory analysis. Firstly, two factors, rather thanone single factor, emerged from the items pertaining to the Impulse Control sub-scale of the EQ-i.Secondly, the items pertaining to the Self-Regard, Self-Actualisation, Optimism and Happinesssub-scales loaded on two factors (factor 1 & 10), rather than four single factors. Similarly, theitems from the Assertiveness and Independence sub-scales loaded on one factor (factor 6) ratherthan two single factors. Finally, Bar-On (1997a) reported that the Empathy and Self-Regard sub-scales emerged as separable factors in the data set, however, they were highly correlated r =0.80,a surprising result given that the solution was rotated orthogonally. In summary, Bar-On con-cluded that the major discrepancy between the theoretical 15-factor structure of the EQ-i and the13-factor structure that emerged from factor analysis was due to the fact that ve factors emergedfrom seven original sub-scales.A series of conrmatory analyses by way of structural equation modelling were performed by

Bar-On (1997a), to determine whether sub-scales that loaded on single factors could be treated asseparable factors. It was reported that the results of these analyses conrmed that Self-Regard,Self-Actualisation, Optimism and Happiness could be treated as separable factors as could theAssertiveness and Independence sub-scales. Conrmatory analyses were also applied to assesswhether a one-factor second order model involving a general factor of emotional intelligence andthe hypothesised ve composite factors tted the data. Multiple t statistics supporting thismodel were reported (i.e. GFI=0.971, Adjusted GFI=0.892, NFI=0.956, and CFI=0.982), andit was concluded that there was empirical support that the EQ-i accounted for the 1515 totalEQcomposite scalessubscales) hierarchical structure of Bar-Ons (1997a) model. Despite theseconrmatory results supporting the dimensional structure of the EQ-i as a match with thetheoretical structure of Bar-Ons (1997a) model, other possible structural equation models of thedata were not reported. Moreover Bar-Ons conrmatory models have not yet been replicated,and need to be by independent research with similar population samples before it can beconcluded that the EQ-i provides an index of the 1515 hierarchical model of emotional andsocial intelligence.In summary it could be argued that the dimensional structure of the EQ-i is unclear. While Bar-

On (1997a) claims to have found empirical support through both exploratory and conrmatoryanalyses that the dimensional structure of the EQ-i represents the 1515 hierarchical model ofemotional and social intelligence, the results of these analyses are not substantive. The con-rmatory analyses have not been replicated, and a wider range of alternative structural models ofthe EQ-i need to be assessed. Moreover, there are a number of issues concerning the interpreta-tion of the exploratory results that need to be addressed, such as, the factor solution that bestrepresents the dimensional structure of the EQ-i.

1.5. The current study

Given that the factor structure of the EQ-i is unclear, in the current study we employedexploratory factor analyses at the outset. Principal axis factoring was performed and joint Par-

1196 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210allel Analysis (Horn, 1965)/Scree Test (Cattell, 1966) criteria were used to determine the factor

On reports no gender dierences in overall EI, consistently (i.e. over a number of populationsamples reported in the EQ-i Technical Manual), females tend to have signicantly higher Inter-

personal EQ than males, however, males tend to have signicantly higher Intrapersonal EQ,Adaptability EQ, and Stress Management EQ than females. Finally, the EQ-i reports severalstudies comparing the EI of clinical groups and matched normative groups showing that those whosuer from mental illness tend to exhibit lower EI than normal populations. We hypothesisedthat there would be a positive correlation between Age and EI; that females would score higheron EI than males; and that individuals reporting no history of mental illness would have asignicantly higher EI than those reporting a history of, and/or current mental illness.

2. Method

2.1. Participants

The sample comprised 377 participants (270 females, 103 males, four unreported ranging in agefrom 15 to 79 with a mean age 39.44 years S.D.=13.83) drawn from the general population viaadvertisements detailing the research. Participants received a small stipend for completing theEQ-i. The ethnic composition comprised; 71%(260) White Caucasian Australians; 19%(70)White Caucasian Emigrants; 9%(34) Asian/Pacic Islanders; Other/Not Reported 1%(9). Incomparison to the Australian population, the sample was slightly above average in education;2.7%(10) had completed primary school education only; 22%(82) had completed secondaryschool education only; 20%(73) had completed a tertiary certicate/diploma; 28%(105) hadcompleted an undergraduate degree; 22%(81) had completed a postgraduate degree; [6%(22) notreported]. Of the 377 participants 58 reported having a history of mental illness, 10 reportedhaving a current mental illness whilst the majority (288) reported having no known history ofmental illness. While there was a gender imbalance in the sample, the age, ethnic, and educationalsolution that best represented the data. There is an emerging consensus amongst the factor ana-lytic literature that these two criteria (Parallel Analysis/Scree Test) are the most accurate fordetermining the number of factors in a data set (Barrett and Kline, 1992). We then performedconrmatory factor analyses via structural equation modelling to compare the goodness of t ofthe exploratory factor solution identied in the current study and the second order dimensionalstructure of the EQ-i proposed by Bar-On (1997a) comprising Overall EI (or a general factor) andthe ve composite components (Intrapersonal EI; Interpersonal EI; Adaptability EI; StressManagement EI; General Mood EI). Finally we examined whether the factor solution identiedin the current study and the second order dimensional structure of the EQ-i proposed by Bar-Ondiered in terms of their relationships with age, gender and two criterion groups; (1) individualsreporting no history of mental illness; and (2) individuals reporting a history of, and/or currentmental illness.Although the eects are small in magnitude, Bar-On (1997a), reports signicant dierences in

EI according to age and gender. Bar-On (1997a) has shown that individuals aged 4049 havesignicantly higher overall EI than individuals aged 2029 (M=102.7 and M=96.8 respectivelyF=0.46. 3, P

aTh mo nt Ba n,

true of the way they typically think, feel or act on a ve-point scale (1=Very Seldom or Not trueof me, 5=Very often true of me or True of me). The items of the EQ-i are summed to yield

Omission Rate is higher than 6% for a given participant, their EQ-i results are considered invalid(Bar-On, 1997a).

1198 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 119112102.3. Procedure

Participants were administered the EQ-i according to the instructions outlined in the EQ-itechnical manual (Bar-On, 1997a).

3. Results

Prior to conducting the factor analyses, the validity of participants responses were examined.Four of the participants responses were found to be invalid as per the Omission Rate criterion,and 15 participants responses were found to be invalid as per the inconsistency index criterion.These participants responses were removed from subsequent analyses. In order to facilitatecomparisons among population samples the test publisher (MHS) converts raw scores to stan-dard scores such that each scale score has a mean of 100 and a standard deviation of 15.Accordingly standardised means and standard deviations of the present sample are provided inTable 1.

Table 1Means and standard deviations for the Bar-On EQ-i Scales

Intrapersonal Interpersonal Adaptability Stress management General mood Total EQ

Mean 94.80 96.48 94.46 94.56 93.92 93.70S.D. 16.17 15.54 15.33 15.58 17.04 15.93scores on 15 lower-order sub-scales, ve higher-order composite scores, and an overall emotionalintelligence score. Of the 133 items that comprise the EQ-i, eight items comprise a PositiveImpression Scale, and seven items comprise a Negative Impression Scale. These two scales aredesigned to determine whether a participant is responding in an overly positive or overly negativefashion. In addition, there is an Inconsistency Index, calculated by summing the dierences inscores between responses on ten pairs of similar items designed to assess random responding.Finally, item 133 I responded openly and honestly to the above sentences, also assesses randomresponding. If a participants response to this item is either 2 or 1 (Seldom true of me orVery seldom or Not true of me), their results are considered invalid. Participants responses arealso considered invalid if a certain percentage of items are not answered (Omission Rate). If theThe EQ-i is a self-report inventory comprising 133 (items) declarative statements phrased in therst-person singular. Participants are required to indicate the degree to which each statement is2.2.1. e Bar-On E tional Quotie Inventory ( r-On EQ-i; Bar-O 1997a)2.2. M terials

As shown in Table 1, the present sample scored slightly lower than the North American nor-mative sample (M=100; S.D.=15) for the EQ Total score as well as the EQ composite scaleshowever, the dierence in average performance for all composite scales fall within one standarddeviation and are therefore considered to be within the normal range according to the EQ-iTechnical Manual.

3.1. Exploratory factor analyses

Principal axis factoring was applied to the items pertaining to the 15 sub-scales of the EQ-i(validity items omitted). The scree test suggested that six factors should be extracted fromthe data set a result conrmed by a parallel analysis using the procedure provided byOConnor (2000). In total, the six factors accounted for 40.3% of the variance (22.8, 5.0,4.3, 3.1, 2.8, and 2.3%, respectively) in the data set. In the un-rotated solution most itemsloaded on a single factor providing evidence for a general factor of EI, a nding consistent withthe interrelatedness of the EQ-i sub-scales reported by Bar-On (1997a). According to the screetest and parallel analysis, six factor (orthogonal- and oblique-rotated) solutions were examined inorder to further interpret the factor structure of the EQ-i. Examination of the factor solu-tions revealed that the six-factor oblique (Direct Oblimin with Kaiser Normalisation) rotatedsolution best represented the present data. Both the orthogonal and oblique factor solutionsproduced highly similar results, however, the oblique-rotated solution was clearer, involving amore even spread of items loading on each factor. The resulting factor loadings are pre-sented in Table 2.We used the process employed by Bar-On (1997a) to identify and label the factors shown

in Table 2, that is, items loading 0.4 or higher. Consistent with Bar-Ons exploratory factoranalyses of the EQ-i, the rst factor that emerged in the analysis accounted for more thanhalf of the variance (22.8%) in the data set, and had items loading .4 and higher primarilyfrom the Self-Regard and Happiness sub-scales with some items from the Optimism, Self-Actualisation and Stress Tolerance sub-scales. Bar-On (1997a, pp. 50,51) describes peoplewho score high on the Self-Regard sub-scale as those who . . .have a good sense of self-esteem,feel positive about themselves, and know who they are. Similarly, people who score high on theHappiness sub-scale of the EQ-i are described by Bar-On as those who have . . .a happydisposition and are pleasant to be with. Although Bar-On labelled this factor Self-Con-tentment because it measures emotionality in general, we labelled this factor EmotionalDisposition.Items loading 0.4 or higher on the second factor that emerged in the present analysis were from

the Interpersonal Relationship, Social Responsibility and Empathy sub-scales of the EQ-i.According to Bar-On (1997a) these three sub-scales all measure aspects of interpersonal skills andfunctioning thus we labelled this factor Interpersonal EQ. The third factor that emerged wasalmost an exact replication of the Impulse Control sub-scale of the EQ-i with one item loading0.4 and higher from the Reality Testing sub-scale, as such, this factor was labelled Impulse Con-trol. Items loading 0.4 and higher on the fourth factor were predominantly from the ProblemSolving sub-scale of the EQ-i, with one item from each of the Self Actualisation, Stress Toleranceand Optimism sub-scales, thus this factor was labelled Problem Solving. The fth factor that

B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210 1199emerged in the present analyses had items loading 0.4 and higher primarily from the Emotional

dings pe the Ba (direct-

1200 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 1191121Q61 0.123 0.333 0.132 0.214 0.144Q82 0.167 0.317 0.119 0.162 0.230 0.175Q62Q690.3230.1800.3570.3360.134 0.2230.1750.2290.235Q55

Q120 0.3520.3890.379 0.3120.178 0.2170.143Q39 0.311 0.390 0.134 0.320 0.141

Q113 0.246 0.407 0.155 0.251 0.228

Q105 0.347 0.417 0.178

Q16 0.421 0.121

Q72 0.440 0.229 0.152 0.102

Q44 0.200 0.449 0.162 0.115 0.265 0.200

Q84 0.478 0.124 0.211 0.165

Q124 0.483 0.163 0.203 0.172

Q90 0.151 0.501 0.173 0.118

Q98 0.149 0.531 0.226 0.181

Q99 0.195 0.650 0.121 0.148

Q36 0.301 0.153 0.101 0.173

Q125 0.303 0.112 0.277 0.276

Q17 0.312 0.294 0.236 0.210 0.174

Q38 0.345 0.258 0.125 0.318

Q132 0.352 0.106 0.215 0.268Q95 0.356 0.300 0.241 0.348 0.101

Q53 0.366 0.351 0.112

Q11 0.390 0.160 0.149 0.270Q127 0.393 0.183 0.210 0.226 0.147

Q106 0.448 0.388 0.182

Q110 0.455 0.138 0.164 0.190 0.122

Q100 0.460 0.133

Q24 0.465 0.187 0.319 0.244

Q26 0.476 0.196 0.239 0.114

Q54 0.478 0.301 0.232 0.153

Q114 0.491 0.109 0.164

Q122 0.509 0.183 0.133 0.122 0.227

Q64 0.515 0.119 0.252 0.259Q31 0.516 0.393 0.115

Q51 0.542 0.104 0.101

Q40 0.561 0.191 0.190Q129 0.563 0.159 0.112

Q77 0.592 0.273

Q85 0.599 0.162 0.105

Q47 0.622 0.147 0.119 0.206

Q70 0.676 0.169 0.175 0.131

Q2 0.678 0.153 0.125

Q91 0.712 0.157Q56 0.725 0.167Item No. Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6Factor loa rtaining to r-On EQ-i oblimin)a

Table 20(continued on next page)

ontinued

0.146 0.384

B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 119112Q21 0.242Q18 0.139 0.179 0.318 0.383 0.102Q68 0.287 0.111 0.116 0.106 0.406Q22Q350.2010.201 0.2820.108 0.4870.4390.166

Q63 0.183 0.238 0.488

Q10 0.129 0.194 0.212 0.498 0.162

Q9 0.105 0.224 0.104 0.569 0.112

Q116 0.609 0.160

Q52 0.131 0.732

Q23 0.107 0.757

Q7 0.181 0.128 0.772

Q112 0.151 0.190 0.212 0.202 0.114Q37 0.225 0.309 0.287

Q75 0.199 0.266 0.138 0.313 0.171 0.190

Q88 0.296 0.335 0.151Q32 0.244 0.362 0.263 0.324

Q4 0.130 0.133 0.375 0.201 0.210

Q8 0.254 0.135 0.196 0.375 0.117

Q67 0.195 0.379 0.204 0.250

Q48 0.312 0.384 0.189 0.273

Q78 0.132 0.238 0.394 0.174 0.349

Q108 0.201 0.101 0.396 0.249

Q6 0.167 0.411

Q20 0.153 0.419 0.262

Q1 0.104 0.108 0.465Q81 0.322 0.154 0.242 0.477

Q29 0.128 0.561 0.126

Q89 0.581Q15 0.609

Q45 0.110 0.160 0.617

Q80 0.145 0.618

Q60 0.145 0.111 0.623

Q107 0.129 0.182 0.281 0.238 0.281

Q58 0.314

Q27 0.121 0.335 0.141 0.118

Q83 0.303 0.250 0.350 0.180 0.136 0.112

Q66 0.315 0.102 0.363 0.167 0.194

Q97 0.182 0.413 0.283 0.123Q13 0.221 0.313 0.548 0.148 0.125

Q86 0.122 0.565 0.115

Q130 0.163 0.282 0.586 0.118

Q102 0.116 0.605 0.196Q42 0.606 0.217

Q73 0.638 0.124 0.158

Q117 0.138 0.193 0.647 0.129

Q119 0.309 0.187

Item No. Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6Table 2 (c )10 1201(continued on next page)

reness le of th ith on rom ea e Inter Relatiness a ty Tes scales Q-i, th actor w led Emreness ading ixth fac prima the F and Inb-scale EQ-i w item e the S lerance f-Regalexibili ported s how ndividu n their s andepend urport sess ho eliant i ls are, indepin their g and a s such lled th Chara exiblepende dent). lations these rs rang 0.01 tomary, ults of lorato r anal ed to the 13of the evious ed by B 1997a) deed th tical 15f emot d socia ence. R he resu e curr y suggprises al facto and six y facto

rmato r analy

analys amine the six

r loadin en sorted g. Items 0:4j j ar ace, and s loadinomittedTable 2 (continued)

Item No. Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6

Q96 0.137 0.259 0.215 0.372 0.214Q111 0.218 0.370 0.233Q126 0.288 0.261 0.111 0.368 0.230Q30 0.276 0.215 0.296 0.320 0.123Q46 0.198 0.219 0.301Q128 0.228 0.108 0.106 0.295 0.132Q131 0.155 0.105 0.571Q92 0.102 0.194 0.167 0.224 0.508

Q104 0.151 0.136 0.135 0.382 0.487Q93 0.135 0.172 0.125 0.437Q103 0.194 0.148 0.125 0.426

Q19 0.106 0.208 0.214 0.266 0.409Q87 0.191 0.255 0.137 0.407Q59 0.245 0.189 0.108 0.403Q76 0.134 0.267 0.249 0.379Q28 0.182 0.149 0.369

1202 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210Next we conducted a series of conrmatory factor es to ex whether -factor3.2. Con ry facto sesEQ-i com a gener r of EI primar rs.

model o ional an l intellig ather, t lts of th ent stud est that

structure EQ-i pr ly report ar-On ( , and in e theore -factor

In sum the res the exp ry facto ysis fail support -factorible, inde nt/depen Intecorre between six facto ed from 0.37.

they are thinkin ctions. A , we labe is factor cter (i.e. /inex-

while ind ence is p ed to as w self-r ndividua and how endent

scales. F ty is pur to asses exible i als are i thought actions,

dence su s of the ith one ach from tress To and Sel rd sub-

Self-Awa . Items lo on the s tor were rily from lexibility depen-

Assertive nd Reali ting sub- of the E us this f as label otional

Self-Awa sub-sca e EQ-i w e item f ch of th personal onship,have been .Facto gs have be ascendin loading > e in bold f those item g < 0:1j ja

Q121 0.150 0.188 0.203 0.145 0.243

Q118 0.169 0.140 0.186 0.181 0.266

Q74 0.177 0.134 0.200 0.285

Q43 0.140 0.218 0.242 0.286

Q3 0.110 0.127 0.249 0.115 0.287

Q14 0.194 0.336Q33 0.280 0.157 0.245 0.357Q49 0.253 0.194 0.195 0.339model determined by the current study better represented the dimensional structure of the EQ-i in

equation modelling (AMOS 4.01; Arbuckle, 1999). Bar-Ons ve-factor model, as illustrated in

Fig. 1, was found to provide a reasonably good t with the present data set (CFI=0.97,RFI=0.95, RMSEA=0.15). Similarly, the 6-factor model derived from the exploratory factoranalysis, as illustrated in Fig. 2, also had a good t (CFI=0.98, RFI=0.96, RMSEA=0.12).To test statistically which model provided a better overall t of the data the dierence in the

chi-square values were determined (Breckler, 1990). The ve-factor model yielded a chi-squarevalue of 742.92 with 85 degrees of freedom. In contrast, the six-factor model from the exploratoryfactor analysis yielded a chi-square value of 664.76 with 113 degrees of freedom. As such, the six-factor model determined by the current study provided a signicantly better t than Bar-Onsoriginal ve-factor model (chi-square dierence was 78.16, df. 28, P

1204 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210Fig. 1. Bar-Ons (1997a) second order ve-factor model of EI.

B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210 1205Fig. 2. Exploratory six-factor model of the Bar-On EQ-i determined by the current study.

i 7 r n n ry s a W ma a 2 t a s e o f i

a o m C l x h a t sin n t t u a P i

e re o

4. Discussion

the current study failed to support previous claims by Bar-On (1997a) tcture of the EQ-i comprises 13 factors that closely match the theoreticalsite scalessubscales) model of emotional and social intelligence. In contrnd that the dimensional structure of the EQ-i comprised a general factofactors.ued that the ndings of the current study oer a more conclusive interpr

Table 4Dierences in emotional intelligence as a function of age, sex, mental illness

Interb=Interpersonal; Imp=Impulse Control; Prob=Problem Solving; Emo self=Emotional Self-Awareness;Five-factor model Six-factor model g

Intra Inter Adapt Stress Mood Emo Disp Interb Imp Prob Emo Self Char

Age 0.09 0.11 0.16 0.20 0.10 0.13 0.12 0.22 0.15 0.02 0.04 0.13

Sex 0.22 0.86 0.13 0.04 0.16 0.08 0.87 0.04 0.09 0.50 0.04 0.33MI 0.42 0.35 0.44 0.47 0.63 0.68 0.34 0.04 0.46 0.28 0.22 0.50Eect sizes in bold are P

in the EQ-i using joint scree test/parallel analysis criteria which have been shown to be the mostaccurate methods for determining the correct number of factors in a correlation matrix (Fava,Velicer, & Eaton, 2000; Zwick & Velicer, 1986). In addition, Bar-On (1997a, 1997b) used anorthogonal (varimax) rotation procedure to interpret the 13-factor solution found in his analysesof the EQ-i. When correlations of r=0.3 or higher exist between variables it has been argued thatoblique-rotated solutions oer a clearer interpretation than orthogonal rotated solutions from anempirical standpoint, providing superior simple structure, superior factor replicability, andcorrelations between the factors which provide insightful information about the construct beinganalysed (Reise, Waller, & Comrey, 2000). In the current study an oblique rotation procedurewas used to interpret the dimensional structure of the EQ-i. Indeed three of the correlationsbetween the six factors identied in the current study were greater than r=0.3 suggesting that theoblique-rotated solution oered the most meaningful interpretation of the current data.The rst factor that emerged in the exploratory analyses of the current study, Emotional Dis-

position, had items loading primarily from the Self-Regard and Happiness sub-scales of the EQ-iwith some items from the Optimism, Self-Actualisation and Stress Tolerance sub-scales. Thesecond factor that emerged, Interpersonal EQ, had items loading from the Interpersonal Rela-tionship, Social Responsibility and Empathy sub-scales of the EQ-i. The Impulse Control, Prob-lem Solving and Emotion Self-Awareness sub-scales of the EQ-i emerged relatively clearly asseparable dimensions (factors 3, 4 and 5), whilst the sixth factor that emerged in the current studyloaded items from the Flexibility and Independence sub-scales of the EQ-i. This six-factor model isconsiderably dierent from Bar-Ons (1997a) second-order ve-factor model of EI that describeshow the 15 sub-scales of the EQ-i cluster to provide broader denitive dimensions (IntrapersonalEQ; Interpersonal EQ; Adaptability EQ; Stress Management EQ; and General Mood EQ). Theonly subscales of the EQ-i that were found in the current study to cluster together according toBar-Ons (1997a) ve-factor model were Interpersonal Relationship, Social Responsibility andEmpathy that form the broader Interpersonal EQ dimension. None of the other sub-scales of theEQ-i were found to cluster according to Bar-Ons second order ve-factor model.Conrmatory factor analyses using structural equation modelling were conducted to compare

the goodness of t of the six-factor model determined in the exploratory analyses of the currentstudy, and Bar-Ons (1997a) second order ve-factor model of EI. Both models were found tohave adequate t, although a statistical comparison of the two models demonstrated the six-fac-tor model determined by the current study was superior. It should be emphasized, however, thatthere is likely a large number of models that would have t the correlation matrix well, as hasbeen demonstrated to be a common characteristic of SEM (MacCallum, Wegener, Uchino, &Fabrigar, 1993). In this case, the dominance of a general factor would likely have supported alarge number of models with a higher-order general factor. The strength of the general factor canbe appreciated by the PCA that found all ve facets to load positively on one factor thataccounted for 72% of the variance. The problem of equivalent models (MacCallum et al., 1993) islikely not solvable using comparative t model t statistics. Rather, the dierential validity of anymodel must be tested using external criterion variables.A series of discriminant function analyses did not nd either Bar-Ons (1997a) second order

ve-factor model or the six-factor model determined by the current study to be more predictive ofgender or mental illness. However, there were dierences in the pattern of eects. For example, in

B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210 1207the current study Bar-Ons (1997a) second order ve-factor model only had the Interpersonal EQ

the six-factor solution identied by the current study. However, the ndings of the current studydo suggest that researchers should factor analyse the scale and use total scores and factor scores

1208 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210given that the items of the EQ-i may not comprise either the 15 or 5 composite components ofBar-Ons (1997a) model of EI. Perhaps the most important implication of the current studyconcerns psychologists, psychiatrists, human resources professionals and organisational devel-opment consultants. Such practitioners have been informed (by both the Technical Manual of theEQ-i and in accreditation courses run by the test publisher MHS), that the EQ-i can clearly mapout individual dierences in emotional and social competencies (e.g. the EQ-i can be used as. . .an aid in the more indepth psychodiagnostic assessment process to clearly map out areas thatneed to be explored and managed in therapy., Bar-On, 1997a, p. 152), when many of its com-ponent parts do not clearly emerge as separable dimensions in factor analyses, (e.g. Assertiveness,Self-Actualisation, Reality Testing, Flexibility, Stress Tolerance, Optimism, and Happiness).Given the EQ-i meaningful pattern of convergent validity with measures of psychological well-being however, as outlined in the introduction of this paper, the EQ-i does appear to provide ageneral index of individual dierences in psychological health, and perhaps therefore, individualscapacity to succeed in coping with environmental demands and pressures.At a broader level the ndings of the current study highlight the need for the psychometric

properties of existing measures of EI to be substantiated by independent research. The dimen-sional structure of tests is one of the most rudimentary of these psychometric elements (Kline,2000), and there is little research to-date that has examined the dimensional structure of existingmeasures of EI.

References

Arbuckle, J. L. (1999). Amos 4.01. SmallWaters Corp.Bar-On, R. (1997a). Emotional Quotient Inventory: technical manual. Toronto: Multi-Health Systems.

Bar-On, R. (1997b). Development of the Bar-On EQ-i: a measure of emotional and social intelligence. Paper presented atthe 105th Annual Convention of the American Psychological Association, Chicago, USA.

Bar-On, R. (2000). Emotional and social intelligence: insights from the Emotional Quotient Inventory. In R. Bar-On,& J. D. A. Parker (Eds.), The handbook of emotional intelligence (pp. 363388). San Francisco: Jossey-Bass.

Bar-On, R., Brown, J. M., Kirkcaldy, B. D., & Thome, E. P. (2000). Emotional expression and implications for occu-pational stress; an application of the Emotional Quotient Inventory (1997a, 2000).

Barrett, P., & Kline, P. (1992). Factor extraction: an examination of three methods. Journal of Personality and Group

Behaviour, 2, 9498.dimension as evidencing a signicant dierence betweenmales and females. In contrast, the six-factormodel identied by the current study had both the Interpersonal and Emotional Self-Awarenessdimensions as evidencing signicant dierences between males and females. The items comprisingEmotional Self-Awareness were in the Intrapersonal facet of the ve-factor model; however, theeect was reduced to non-signicance, because of the other items grouped into Intrapersonal (e.g.Assertiveness, and Self-Actualisation). Thus, although neither model oered more predictive valid-ity, the six-factor model oered a more insightful interpretation when analysing gender dierences.It should be noted that the six-factor solution of the EQ-i identied by the current study needs

to be replicated in order to substantiate whether this solution oers a more accurate descriptionof the EQ-i dimensional structure. Indeed, there is a considerable portion of error variance with

B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210 1209Beck, A. T., & Steer, R. A. (1987). Beck Depression Inventory manual. San Antonio, TX: The Psychological Corpora-

tion.Breckler, S. J. (1990). Applications of covariance structure modeling in psychology: cause for concern? PsychologicalBulletin, 107, 260273.

Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioural Research, 1, 141161.Cattell, R. B. (1978). The scientic use of factor analysis. New York: Plenum.Cattell, R. B., Cattell, A. K., & Cattell, H. E. (1993). Sixteen personality factor questionnaire (5th ed.). Champaign, IL:

Institute for Personality and Ability Testing.Cattell, R. B., & Kline, P. (1977). The scientic analysis of personality and motivation. London: Academic Press.Cooper, R. K., & Sawaf, A. (1997). Executive EQ: emotional intelligence in leadership and organizations. New York:Grosset/Putnum.

Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PIR) and NEO Five-Factor Inventoryprofessional manual. Odessa, FL: Psychological Assessment Resources.

Dawda, D., & Hart, S. D. (2000). Assessing emotional intelligence: reliability and validity of the Bar-On Emotional

Quotient Inventory (1997A; 2000) in university students. Personality and Individual Dierences, 28, 797812.Derogatis, L. R. (1973). SCL-90: Administration, scoring and procedures. Baltimore, MD: John Hopkins UniversitySchool of Medicine.

Fava, J. L., Velicier, W. F., & Eaton, C. A. (2000). Construct explication through factor or component analysis: areview and evaluation of alternative procedures for determining the number of factors or components. InR. D. Gon, & E. Helmes (Eds.), Problems and solutions in human assessment: honoring Douglas N. Jackson at

seventy (pp. 4171). New York, NY, US: Kluwer Academic/Plenum Publishers.Goleman, D. (1995). Emotional intelligence. New York: Bantam Books.Horn, J. L. (1965). A rationale and technique for estimating the number of factors in factor analysis. Psychometrika,30, 179185.

Kirkcaldy, B. (1995). The Kirkcaldy Quality of Life Questionnaire. Unpublished test.Kline, P. (1994). An easy guide to factor analysis. London: Routledge.Kline, P. (2000). Handbook of psychological testing (2nd ed.). London: Routledge.

MacCallum, R. C., Wegener, D. T., Uchino, B. N., & Fabrigar, L. R. (1993). The problem of equivalent models inapplications of covariance structure analysis. Psychological Bulletin, 114, 185199.

Marin, G., Sabogal, F., Marin, B., Otero-Sabogal, R., & Perex-Stable, E. (1987). The development of a Short

Acculturation Scale for Hispanics. Hispanic Journal of Behavioural Sciences, 9, 183205.Mayer, J., Caruso, D. R., & Salovey, P. (2000). Selecting a measure of emotional intelligence: the case for ability scales. InR. Bar-On, & J. D. A. Parker (Eds.),The handbook of emotional intelligence (pp. 320342). San Francisco: Jossey-Bass Inc.

Mayer, J., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey, & D. Sluyter (Eds.), Emotionaldevelopment and emotional intelligence: implications for educators (pp. 331). New York: Basicbooks.

Mayer, J. D., Salovey, P., & Caruso, D. R. (1999). Working manual for the MSCEIT research version 1.1. Toronto,Canada: Multi-Health Systems.

Mayer, J. D., Salovey, P., & Caruso (2000). Models of emotional intelligence. In R. J. Sternberg (Ed.), Handbook ofintelligence (pp. 396420). New York: Cambridge.

McCrae, R. R. (2000). Emotional intelligence from the perspective of the ve-factor model of personality. In

R. Bar-On, & J. D. A. Parker (Eds.), The handbook of emotional intelligence (pp. 263276). San Francisco: Jossey-Bass.

Newsome, S., Day, A. L., & Catano, V. M. (2000). Assessing the predictive validity of emotional intelligence. Person-

ality and Individual Dierences, 29, 10051016.OConnor, P. (2000). SPSS and SAS programs for determining the number of components using parallel analysis andVeliciers MAP test. Behaviour Research Methods, Instruments & Computers, 32(3), 396402.

Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., & Seligman, M. E. P. (1982). The Attri-

bution Style Questionnaire. Cognitive Therapy Research, 16, 287300.Petrides, K., & Furnham, A. (2000). On the dimensional structure of emotional intelligence. Personality and IndividualDierences, 29, 313320.

Reise, S. P., Waller, N. G., & Comrey, A. L. (2000). Factor analysis and scale revision. Psychological Assessment,

12(3), 287297.Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition and Personality, 9, 185211.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 (Ed.), Emotion, disclosure, and health(pp. 125154). Washington, DC: American Psychological Association.

Wagner, F. R., & Morese, J. J. (1975). A measure of individual sense of competence. Psychological Reports, 36, 451

459.Zwick, W. R., & Velicer, W. F. (1986). Comparison of ve rules for determining the number of components to retain.Psychological Bulletin, 17, 253269.

1210 B.R. Palmer et al. / Personality and Individual Dierences 35 (2003) 11911210

Examining the factor structure of the Bar-On Emotional Quotient Inventory with an Australian general population sampleIntroductionModels and measures of emotional intelligenceBar-Ons (1997a) model of emotional intelligenceThe Bar-On Emotional Quotient InventoryDimensional structure of the Bar-On EQ-iThe current study

MethodParticipantsMaterialsThe Bar-On Emotional Quotient Inventory (Bar-On EQ-i; Bar-On, 1997a)

Procedure

ResultsExploratory factor analysesConfirmatory factor analysesDemographic differences

DiscussionReferences