Requests for copies should be addressed to: GP de Bruin, Department of Psychology, Uni- versity of Stellenbosch, Private Bag X1, Matieland,7602 THE RELATIONSHIP BETWEEN PERSONALITY TRAITS AND VOCATIONAL INTERESTS GIDEON P DE BRUIN Department of Psychology University of Stellenbosch ABSTRACT This study examined the relationship between vocational interests and basic personality traits.The interest ¢elds of the 19-Field-Interest Inventory were related to the second order factors of the 16 Personality Factor Question- naire by means of a factor extension analysis. The results showed that extroverts tend to be interested in ¢elds related to social contact and the in£uencing of other people. Emotionally sensitive individuals tend to be interes- ted in the arts and languages. Independent individuals tend to be interested in creative thinking.The implications of the ¢ndings for career counselling are discussed. OPSOMMING Hierdie studie het ondersoek ingestel na die verband tussen beroepsbelangstellings en basiese persoonlikheids- trekke. Die 19 belangstellingsvelde van die 19-Veld-belangstellingsvraelys is aan die hand van ’n faktorverlen- gingsontleding met die tweede orde faktore van die 16-Persoonlikheidsfaktorvraelys in verband gebring. Die resultate dui daarop dat ekstroverte geneig is om invelde wat sosiale kontaken die be|«nvloeding van mense behels, belang te stel. Emosioneel sensitiewe individue is geneig om in kunssinnge en taal verwante velde belang te stel. Onafhanklike individue is geneig om in kreatiewe denke belang te stel. Die implikasies van die resultate vir loop- baanvoorligting word bespreek. The aim of this study was to investigate the relationship be- tween vocational interests, as measured by the 19-Field-Inte- rest Inventory (19FII), and personality traits, as measured by the 16 Personality Factor Questionnaire (16PF). Ackerman (1997) proposed two reasons why the studyof the relationships between interests and personality traits are important. In the ¢rst place it is a common scienti¢c goal to investigate the relationships between di¡erent constructs. This strategy may lead to important new scienti¢c discoveries.The second reason is that the studyof the relationships between interests and perso- nality traits will reveal what they have in common and what unique information they can provide.This information is po- tentially useful in the career counselling context. Recently, several studies have investigated the overlap between vocational interests and personality traits (Blake & Sackett, 1999; Carless,1999; De Fruyt & Mervielde,1997; Gottfredson, Jones & Holland, 1993; Hogan & Blake, 1999; Katz, Joyner & Seaman,1999;Tokar & Swanson,1995).The results of these stu- dies reveal a consistent but relatively weak pattern of relation- ships between interests and personality traits. Most of the recent studies made use of Holland’s (1997) hexagonal model of vocational interest types and the ¢ve-factor model of per- sonality traits (Wiggins & Trapnell, 1997). Holland proposes six broad and fundamental interest or vocational personality types: (a) Realistic, (b) Investigative, (c) Artistic, (d) Social, (e) Enterprising, and (f) Conventional.The ¢ve-factor model of personality traits proposes ¢ve broad and fundamental personality traits: (a) Extroversion, (b) Neuroticism, (c) Agree- ableness, (d) Conscientiousness and (e) Openness to Expe- rience (see Digman, 1990; John, 1990; and Wiggins & Trapnell, 1997 for full discussions of these factors). The results of studies investigating the correspondence between Holland’s interest types and the ¢ve-factor model generally show the following: Extroversion is related to Enterprising and Social interests, Openness to Experience is related to Investigative and Artistic interests, and Conscientiousness is related to Con- ventional interests (Blake & Sackett, 1999; de Fruyt & Mer- vielde, 1997; Gottfredson et al, 1993; Hogan & Blake, 1999; Holland,1999;Tokar & Swanson,1995). These ¢ndings are consistent with Holland’s (1997) theory and provide support for the construct validity of Holland’s model of vocational interests and the ¢ve-factor model of personality. From a scienti¢c perspective, the overlap between the two models provides a parsimonious view of how interests and personality traits are related. However, from a practical per- spective, predictive and explanatory power is lost with broad constructs such as Holland’s interest types and the broad traits of the ¢ve-factor model (Mershon & Gorsuch, 1989; Ashton, Jackson, Paunonen, Helmes & Rothstein, 1995). In the career counselling context it often is desirable to have knowledge about an individual’s speci¢c interests. From this perspective it may be useful to know how such speci¢c and narrow ¢elds of interest relate to personality traits. The19FII and the16PF are two commonly used instruments for career counselling purposes in South Africa. Each instrument is thought to provide useful information in the career counselling process. Career counsellors usually hope to ¢nd a logically con- sistent pattern of vocational interests and personality traits and it is hoped that an individual’s vocational interests will be compati- ble with his or her personality traits. For instance, on theoretical grounds career counsellors would probably expect extroverted individuals to be interested in occupations that involve opportu- nities for social contact. The 19FII provides measures for 19 speci¢c ¢elds of interest that are narrower than the six types of Holland’s model. However, conceptual and factor analyses have shown that the correlations between the 19 ¢elds of the 19FII can be explained by six broad factors that appear to correspond with Holland’s interest types (De Bruin & duToit,1995; Langley,1989; Malan, 1987).The relationships between the 19FII ¢elds and Holland’s types are summarised inTable 1.The table is intended to serve as a heuristic summary of the relationship between the 19FII ¢elds and Holland’s interest types. It should be emphasized, however, that the empirical relationship between the 19FII scales and Holland’s types are more complex than what is pres- ented inTable 1. For instance, several of the 19FII ¢elds are re- lated to more than one Holland type. However, in the table each 19FII ¢eld is linked only with the Holland type that it probably is most strongly associated with.There are further as- pects of Table 1 that need to be clari¢ed. Firstly, it may appear theoretically inconsistent that the Practical-Male ¢eld is asso- ciated with the Realistic type, while the Practical-Female ¢eld is associated with the Artistic type. Inspection of the items of the Practical-Male ¢eld reveals that they re£ect stereotypically male activities, such as repairing broken appliances and work- ing with hand tools.These activities are clearlyconsistent with the activities associated with the Realistic type. The items of the Practical-Female scale re£ect stereotypical female activi- ties, such as preparing meals and making clothes or soft furni- ture, cooking, and needlework. These activities involve an aesthetic element and in this respect the Practical-Female scale is associated with the Artistic type. SAJournal of Industrial Psychology, 2002, 28(1),49-52 SATydskrif vir Bedryfsielkunde, 2002, 28(1),49-52 49 TABLE 1 THE RELATIONSHIPS BETWEEN THE 19FII AND HOLLAND’S INTEREST TYPES Holland ¢elds 19FII ¢elds Realistic Practical-Male Nature Sport Investigative Science CreativeThinking Numerical Artistic Fine Arts Performing Arts Language Practical-Female Social Sociability Welfare Travel Enterprising Law Public Speaking Business Conventional Clerical Service Roos (1988) investigated the relationships between the ¢elds of the 19FII and the 14 personality traits of the High School Per- sonality Questionnaire (HSPQ). Roos (1988) performed a principal factor analysis on the intercorrelations of the interest and personality scales. He extracted and rotated seven factors. The ¢rst two factors were characterised by high loadings of the HSPQ scales and the remaining ¢ve factors were charac- terised by high loadings of the 19FII scales. The most signi¢- cant ¢ndings of the study were (a) that the extroversion factor is related to two interest scales, namely Sociability and Sport, and (b) that emotional sensitivity is related to artistically rela- ted interest ¢elds, namely Performing Arts, Fine Arts, Language and Practical-Female. The analytic strategy followed in the present study was to sub- ject the 16PF scales to a factor analysis. The resulting second order factors were then extended to the scales of the 19FII by means of Gorsuch’s (1997) factor extension analysis procedure. With this procedure a subset of a larger set of variables is sub- jected to a factor analysis.These variables are referred to as core variables. In the present study the 16PF scales served as the core variables.The resulting factors are then extended to the varia- bles that were not included in the factor analysis.These varia- bles are referred to as extension variables. In the present study the 19FII ¢elds served as the extension variables. Gorsuch (1997) explains that the overlap between the extension variables and the factors can only be in£uenced by the overlap between the core variables and the extension variables.The use of the factor extension analysis will allow for an examination of how voca- tional interests ¢t into the space de¢ned by the factors of the 16PF. METHOD Participants The participants were 1502 ¢rst year university students at a South African university.The mean age of the students was 18 years and there were approximately equal numbers of males and females. The students represented all the faculties at the university. Measuring instruments 16 Personality Factor Questionnaire (16PF) The 16PF (Form A) was used to obtain measures of the parti- cipants’ personality traits. The 16PF £owed from Cattell’s ex- tensive factor analytic research program into the structure of human personality. He identi¢ed about 20 basic normal perso- nality traits, of which 16 are measured by the 16PF. One of the 16PF scales, namely Scale B, measures intelligence rather than personality and this scale was not included in the present analysis. Previous factor analytic studies have shown that the correlations among the15 remaining scales can be summarised and explained in terms of ¢ve second-order factors, namely Extroversion, Anxiety,Tough Poise, Independence and Con- trol (Hofer et al.,1997; Krug & Johns,1986).The second-order factors are broader than the ¢rst-order factors and explain a wider range of behaviour. The second-order factors are also more reliable and replicable (in a factor analytic sense) than the ¢rst-order factors (Hofer et al.,1997; Krug & Johns,1986). 19-Field-Interest Inventory (19FII) The 19FII was constructed by Fouche¤ and Alberts (1971) and provides normative scores for 19 ¢elds of interest.The 19FII is intended for use with Grade 10 to Grade 12 pupils, but the in- strument is also used in higher education institutions.The ma- nual reports split-half reliability coe⁄cients ranging between .88 and .98 for high school boys and girls. Overall, the reliabi- lity of the 19FII for this group can be described as satisfactory. The participants in the present study are similar in age and educational level to the Grade 12 normative group and it is as- sumed that the present 19FII scores are also reliable. Nicol (1978) demonstrated that the 19FII e¡ectively discriminates between university students in di¡erent study ¢elds. RESULTS AND DISCUSSION On the basis of previous factor analytic studies, it was expected that the correlations between the 15 16PF scales would be ade- quately and meaningfully explained by ¢ve second order fac- tors (Hofer et al., 1997; Krug & Johns,1986). Accordingly, ¢ve second order factors were extracted by means of the principal axis method with iterated communalities.The factors were ro- tated to simple structure according to the promax criterion (k = 4). The resulting factor pattern matrix and the cor- relations between the factors are reported inTables 2 and 3. TABLE 2 PROMAX ROTATED FACTOR PATTERN MATRIX OF THE 16PF SCALES 16PF Scale 16PF second-order factors I II III IV V A (Warmth) .59 .33 .18 -.12 .15 C (Emotional Stability) .25 -.01 .38 .18 -.37 E (Dominance) .54 -.07 -.17 .66 .01 F (Liveliness) .76 -.03 -.03 .31 .05 G (Rule-consciousness) .06 -.11 .73 -.16 .11 H (Social Boldness) .81 .08 .03 .29 -.18 I (Sensitivity) .05 .69 -.10 -.10 .00 L (Vigilance) .06 -.12 -.02 .37 .44 M (Abstractedness) -.03 .54 -.03 .17 .13 N (Privateness) -.08 -.11 .40 .20 .15 O (Apprehension) -.05 .09 .11 -.10 .72 Q1 (Openness to Change) .04 .11 .05 .41 -.07 Q2 (Self-reliance) -.62 .18 .02 .18 -.04 Q3 (Perfectionism) -.15 .05 .57 -.09 -.19 Q4 (Tension) -.04 .04 -.10 .07 .73 Note. All factor pattern coe⁄cients are rounded to two decimal places. All absolute fac- tor pattern coe⁄cients of .30 and higher are underlined. I = Extroversion; II = Tough Poise; III = Control; IV = Independence; V = Anxiety. Inspection of Table 2 reveals that the ¢ve factors correspond closely with the well-known16PF second-order factors repor- ted by Hofer et al. (1997) and Krugand Johns (1986). Following KrugandJohns (1986) these factors are labeled as Extroversion, Tough Poise, Control, Independence and Anxiety. TABLE 3 INTERCORRELATIONS OF THE 16PF SECOND-ORDER FACTORS E TP C I A Extroversion (E) 1.00 Tough Poise (TP) -.03 1.00 Control (C) -.02 .19 1.00 Independence (I) -.11 .14 .32 1.00 Anxiety (A) .00 .27 -.37 .03 1.00 Note. All correlations are rounded to two decimal places. DE BRUIN50 1 Because of the large sample size and the resulting high statistical power of even very small correlations between the interest ¢elds and personality traits are statistically signi¢cant.Therefore, Cohen’s (1988) guidelines for the interpreta- tion of correlation coe⁄cients in psychological research were followed. Accor- ding to Cohen correlations of .20 can be described as moderate in strength. The second-order factors of the 16PF were extended to the 19 scales of the 19FII by means of Gorsuch’s (1997) factor exten- sion analysis procedure (see Table 4). In the following para- graphs the relationships between the personality factors and the interest scales are discussed. TABLE 4 CORRELATIONS BETWEEN 19FII FIELDS AND 16PF SECOND-ORDER FAC- TORS Interest ¢eld 16PF second-order factors I II III IV V Fine Arts .02 .15 -.02 .14 .04 Clerical -.06 .06 .06 -.10 .10 Welfare .15 .33 .06 -.11 .06 Performing Arts .13 .26 -.01 .07 .07 Science -.15 -.13 .10 .17 -.13 Historical -.07 .14 .00 .12 .08 Public Speaking .38 .14 .03 .18 .05 Numerical -.12 -.25 .11 .11 -.07 Sociability .57 .07 .01 .03 .07 Creative Thinking .04 -.05 .14 .24 -.05 Travel .25 .08 -.02 .07 .11 Practical-Female .03 .16 -.01 -.02 .05 Law .22 .06 .01 .16 .10 Sport .23 -.15 .01 .06 -.07 Language .07 .33 -.07 .09 .14 Service .20 .14 -.03 -.06 .12 Practical-Male -.13 -.22 .07 .14 -.13 Business .09 -.20 .02 .15 .02 Note. All correlations are rounded to two decimal places. All absolute correlations of .20 and higher are underlined. I = Extroversion; II = Tough Poise; III = Control; IV = In- dependence; V = Anxiety. Factor 1 (Extroversion) The following19FII ¢elds were meaningfullycorrelated (r �.20)1 withthe Extroversion factor: Sociability, .57; Public Speaking, .38; Travel, .25; Sport, .23; Law, .22; and Service .20.These ¢elds relate to contact with other people on mainly two levels: (a) recreation (Sociability,Travel and Sport), and (b) in£uencing of other people (Law and Public Speaking). The high correlations of Sociability and Public Speaking with the Extroversion factor clearly demon- strate extroverts’need for contact with other people. Factor 2 (Tough Poise) Low scores for theTough Poise factor indicate tough-minded- ness and high scores indicate emotional sensitivity.This factor is similar in meaning to the Openness to Experience factor of the ¢ve-factor model (Cattell,1995).The following19FII ¢elds were meaningfully correlated with the Tough Poise factor: Welfare, .33; Language, .33; Performing Arts, .26; Numerical, -.25; Practical-Male, -.22; and Business, -.20. It is interesting that three of the 19FII ¢elds correlated positively (Welfare, Language and Performing Arts) and three correlated negative- ly (Numerical, Practical-Male and Business) with the Tough Poise factor.The three ¢elds that correlated positively all relate to contact with people and communication, whereas the three ¢elds that correlated negatively are related to ‘‘things’’and ab- stract concepts (numbers).The positive association of Langua- ge and Performing Arts withTough Poise provide support for the link between Holland’s (1997) Artistic ¢eld and the Open- ness to Experience factor of the ¢ve-factor model.The remai- ning correlations indicated that emotionally sensitive individuals are interested in the welfare of others and are not interested in ¢elds that mostly relate to’’things’’ (i.e. Business, Numerical and Practical-Male) rather than people. Factor 3 (Control) High scores for the Control factor are associated with perfec- tionism and adherence to societal norms and rules. Low scores are associated with a lack of personal control and a disregard for societal norms and rules.This factor is similar to the Con- scientiousness factor of the ¢ve-factor model (Noller, Law & Comrey, 1987). No 19FII ¢eld had a meaningful correlation with the Control factor. However, previous studies have sug- gested a link between Conscientiousness (or Control) and the Conventional interest type (Blake & Sackett, 1999; Hogan & Blake,1999).The lack of a similar ¢nding in the present study suggests that the19FII possiblydoes not provide adequate mea- surement of Holland’s Conventional type. Factor 4 (Independence) High scores for the Independence factor are associated with a critical and domineering interpersonal style. Low scores are associated with group dependence and submissiveness. This factor is similar to the Agreeableness factor of the ¢ve-factor model. Only one 19FII ¢eld had a positive and meaningful correlation with the Independence factor, namely Creative Thinking (r = .24), suggesting that independent individuals like to generate alternative and new ideas. Factor 5 (Anxiety) High scores for the Anxiety factor indicate emotional instabi- lity and a lack of self-con¢dence. Low scores indicate emotio- nal control and high self-con¢dence. This factor is similar to the Neuroticism factor of the ¢ve-factor model. No19FII ¢eld had a meaningful correlation with Anxiety. Previous studies also failed to demonstrate a link between Anxiety and voca- tional interests (de Fruyt & Mervielde, 1997; Gottfredson et al,1993;Tokar & Swanson,1995). In summary, the ¢ndings show that three second-order perso- nality factors, namely Extroversion,Tough Poise and Indepen- dence have meaningful relationships with vocational interests. Individuals with high scores for Extroversion are interested in ¢elds that relate to contact with other people. Individuals with low scores for Tough Poise are interested in the performing arts and languages, but tend to move away from business, numbers and stereotypical practical-male activities. Indivi- duals with high scores for the Independence factor are drawn towards creative thinking. It should be noted, however, that the correlations between the personality factors and interest scales are relatively weak. Thus, although there are psychologically meaningful rela- tionships between the two sets of variables, personality traits and vocational interests appear to represent two clearly dif- ferent domains of behaviour. These ¢ndings are not consis- tent with Holland’s (1999) strong assertion that interest inventories are also personality inventories. Hogan and Blake (1999, p. 54) explain that interests re£ect a person’s ‘‘goals, values and aspirations’’, while personality traits re£ect a person’s typical behaviour. In this sense interest question- naires and inventories directly tap into an individual’s iden- tity, while personality questionnaires and inventories directly tap into an individual’s characteristic behaviour and only indirectly into his or her identity. Hogan and Blake (1999, p. 54) further contend that while ‘‘...interest measures tell us how much an individual will like an occupation, per- sonality measures tell us about the social skills and drives ne- cessary to succeed once in the occupation.’’ Thus, although the personality traits of Anxiety and Control appear to be unrelated to vocational interests, they may be important in the career counselling context because they provide infor- mation about how the individual is likely to behave in the work place. It is concluded that the personality trait measures, such as the 16PF, and vocational interest measures, such as the 19FII, can each contribute potentially useful information to the career counselling process. By measuring personality traits and voca- tional interests the career counsellor is likely to obtain an inte- grated and comprehensive picture or understanding of his or her client. Similarly, the client is likely to gain a more compre- hensive understanding of him or herself if personality traits and vocational interests form part of the psychological assess- ment process. 51PERSONALITY TRAITS AND VOCATIONAL INTERESTS REFERENCES Ackerman, P.L. (1997). Personality, self-concept, interests, and intelligence:Which construct doesn’t ¢t? Journal of Persona- lity, 65,171-204. Ashton, M.C., Jackson, D.N. Paunonen, S.V., Helmes, E. & Rothstein, M.G. (1995).The criterionvalidityof broad fac- tor scales vs speci¢c factor scales. Journal of Research in Perso- nality, 29, 432-442. Blake, R.J., & Sackett, S.A. (1999). Holland’s typology and the ¢ve factor model: A rational-empirical analysis. Journal of CareerAssessment,7, 249-279. Carless, S.A. (1999). Career assessment: Holland’s vocational interests, personality characteristics, and abilities. Journal of CareerAssessment, 7,125-144. Cattell, H.E.P. (1995). Some comments on a factor analysis of the 16PF and the NEO Personality Inventory-Revised. Psychological Reports, 77,1307-1311. Cattell, R.B., Eber, H.W. & Tatsuoka, M. (1970). Handbook for the16PF. Champagne, Il: IPAT. Cohen, J. (1988). Statistical poweranalysis for the behavioral sciences. (2nd ed.). Hillsdale, NJ: Erlbaum. De Bruin, G.P. & duToit, K. (1995). Die faktorstruktuur van die 19-Veldbelangstellingsvraelys en Holland se struktuur van belangstellings. Journal of Industrial Psychology, 21(3), 24-28. De Fruyt, F., & Mervielde, I. (1997). The ¢ve factor model of personality and Holland’s RIASEC interest types. Persona- lity and Individual Di¡erences, 23, 87-103. Digman, J.M. (1990). Personality structure: Emergence of the ¢ve-factor model. Annual Review of Psychology, 41, 417-440. Fouche¤ , F.A., & Alberts, N.F. (1971). Manual for the 19-Field-Interest Inventory(19-FII).Pretoria:HumanSciencesResearch Council. Gorsuch, R.L. (1997). New procedure for extension analysis in exploratory factor analysis. Educational and Psychological Measurement, 57, 725-740. Gottfredson, G.D., Jones, E.M., & Holland, J.L. (1993). Perso- nality and vocational interests: The relation of Holland’s six interest dimensions to ¢ve robust dimensions of perso- nality. Journal of Counseling Psychology, 40, 518-524. Hofer, S.M., Horn, J.L., & Eber, H.W. (1997). A robust ¢ve- factor structure of the 16PF: Strong evidence from inde- pendent rotation and con¢rmatory factorial invariance procedures. Personalityand IndividualDi¡erences, 23, 247-269. Hogan, R., & Blake, R. (1999). John Holland’s vocational typology and personality theory. Journal ofVocational Beha- viour, 55, 41-56. Holland, J.L. (1997). Making vocational choices. Odessa, Fl: Psy- chological Assessment Resources. Holland, J.L. (1999).Why interest inventories are also persona- lity inventories. In M.L. Savickas & A.R. Spokane (Eds.), Vocational interests: Meaning, measurement and counseling use (pp. 87-101). Palo Alto, CA: Davies-Black. John, O.P. (1990). The‘‘Big Five’’factor taxonomy: Dimensions of personali- ty in the natural language and in questionnaires. In L.A. Pervin (Ed), Handbook of personality:Theory and research (pp. 66-100). NewYork, NY: Guilford. Katz, L., Joyner, J.W. & Seaman, N. (1999). E¡ects of joint in- terpretation of the Strong Interest Inventory and the Myers-Briggs Type Indicator in career choice. Journal of CareerAssessment, 7, 281-298. Krug, S.E. & Johns, J.F. (1986). A large scale cross-validation of second-order personality structure de¢ned by the 16PF. Psychological Reports, 59, 683-693. Langley, R. (1989). Gerekenariseerde loopbaanvoorligting: ’n Evalue- ring van die DISCOVER-stelsel. (Computerised career counseling: An evaluation of the DISCOVER system). Unpublished doctoral dissertation. Johannesburg, Rand Afrikaans University. Mershon, B. & Gorsuch, R.L. (1988). Number of factors in the personality sphere. Journal of Personalityand Social Psychology, 55, 675-680. Nicol, J. (1978). Die di¡erensie« le waarde van die 19-Veld-Belangstel- lingsvraelys. (The di¡erential value of the 19-Field-Interest Inventory). Unpublished master’s thesis. Pretoria, Univer- sity of Pretoria. Noller, P., Law, H., & Comrey, A.L. (1987). Cattell, Comrey, and Eysenck personality factors compared: More evidence for the ¢ve robust factors? Journal of Personality and Social Psychology, 53, 775-782 Roos,W.L. (1988). Die verband tussen persoonlikheid en belangstel- ling: ’n eksploratiewe ondersoek. (The relationship between personality and interests: An exploratory investigation.) Pretoria: Human Sciences Research Council. Tokar, D.M., & Swanson, J.L. (1995). Evaluation of the corres- pondence between Holland’s vocational personality typo- logy and the ¢ve factor model of personality. Journal of Vocational Behavior, 46, 89-108. Wiggins, J.S. & Trapnell, P.D. (1997). Personality structure:The return of the big ¢ve. In R. Hogan, J. A. Johnson, & S.R. Briggs (Eds.), Handbook of personality psychology (pp. 737- 758). San Diego, CA: Academic. DE BRUIN52