Requests for copies should beaddressed to:TGroenewald,Technikon SA, Private Bag X6, Florida,1710 PERSONALITYAND COGNITIVE ABILITYAS PREDICTORS OF THE JOB PERFORMANCE OF INSURANCE SALES PEOPLE L LA GRANGE G ROODT Department of Human Resource Management, Rand Afrikaans University ABSTRACT The purpose of this study was to determine whether personality and a measure of cognitive ability (’verbal reaso- ning ability’) would signi¢cantly predict the job performance (’managerial ratings’) of sales people in a large South African insurance company.The Customer Contact Styles Questionnaire (CCSQ 5.2) and theVerbal Eva- luationTest (VCC 3) were administered to170 broker consultants, and their managers rated their job performance on the Customer Contact Competency Inventory (CCCI). By making use of multiple regression analysis it was found that certain personalitydimensions signi¢cantly predict job performance, and that’verbal reasoningability’ did not have any signi¢cant predictive power.These ¢ndings, the implications thereof and suggestions for possible further research are discussed. OPSOMMING Die doel van hierdie studie was om te bepaal of persoonlikheid en ’n meting van kognitiewe vermoe« (’verbale redeneervermoe« ’) beduidende voorspellings van die werksprestasie (’bestuursbeoordelings’) van verkoopsmense in ’n groot Suid-Afrikaanse versekeringsmaatskappy, kan maak. Die ’Customer Contact Styles Questionnaire’ (CCSQ 5.2) en die ’Verbal Evaluation Test’ (VCC 3) was op 170 makelaarskonsultante afgeneem, en is deur hul bestuurders, met behulpvan die’Customer Contact Competency Inventory’ (CCCI), beoordeel. Deur gebruik te maak van meervoudige regressie analises is daar bevind dat werksprestasie beduidend deur sekere persoonlik- heidsdimensies voorspel word. Ook is bevind dat’verbale redeneervermoe« ’nie werksprestasie beduidend voorspel nie. Hierdie bevindinge, die implikasies daarvan en voorstelle vir moontlike verdere navorsing, word bespreek. The business environment inwhich organisations have to ope- rate has become increasingly complex. Business organisations are faced with ever increasing uncertainty, turbulence and changes in the external environment. These changes are due to, amongst others, increased international and local com- petition, technological advances and increased stakeholder- and customer expectations. To survive in this ever-changing external environment management must use its awareness of these forces to improve its internal business operations (DeVil- liers & Slabbert,1996). To provide quality customer service is one way to improve in- ternal business operations. According to Connellan and Zem- ke (1993) this factor is the only true way that an organisation can di¡erentiate itself from its competition. Product inno- vations are being duplicated within weeks by competitors’ and the margins between the quality and functionality of pro- ducts, within the same industry, are constantly reducing. Sales people can no longer only focus on selling the product. Over and above delivering reliable, quality-orientated customer ser- vice, they also have to ensure a sense of customer satisfaction that gives customers a sense of surprise and delight. These changes in the roles of sales people re£ect the increased importance of customer focused behaviour. By employing the right type of sales person with the necessary skills, behaviours and abilities an organisation can not only increase productivi- ty, and hence its bottom-line, but also maintain a competitive advantage (Connellan & Zemke, 1993;Vinchur, Schippmann, Switzer III & Roth,1998). These factors, amongst others, have caused researchers in the past to search for the inherent behaviours and abilities that would be able to be used to predict sales success. According toVinchur et al. (1998) a wide range of predictors have, in the past, been used for these purposes. Predictors such as bio-data, personality and cognitive abilities, and also more uncon- ventional approaches (e.g. handwriting analysis) have been researched. The focus of this study will be on the more con- ventional predictors i.e. personality and cognitive abilities.To the lay person it might appear self-evident that personality factors and a cognitive ability such as verbal reasoning would play an important part in the performance of sales people. However, psychological literature and research ¢ndings on these matters are equivocal. In the next section, relevant litera- ture and research ¢ndings pertaining to the topics of per- sonality, cognitive abilities and job performance will be discussed. The aim of this study is to determine whether personality con- structs and cognitive ability can be used to predict the job per- formance of sales people in a large South African insurance company. Personality Allport (1937, p.48) de¢ned personality as: ’Personality is the dynamic organisation within the individual of those psycho- physical systems that determine his unique adjustment to his environment.’ According to Ivancevich and Matteson (1993), one of the most complex matters to understand in organisational settings, is the relationship between work behaviour and personality. Per- sonality is in£uenced mainly by cultural, social, family re- lationship and hereditary forces.They provided the following de¢nition of personality: An individual’s personality is a relatively stable set of characteristics, tendencies, and temperaments that have been signi¢cantly formed by inheritance and by social, cul- tural, and environmental forces.This set of variables deter- mines the commonalities and di¡erences in the behavior of the individual (p. 98). An issue of interest to behavioural scientists and researchers is whether personality factors such as those measured by ques- tionnaires or inventories can predict behaviour or perfor- mance in organisations. According to Barrick and Mount (1991), and Hogan and Ni- cholson (1988), researchers in the ¢eld of personality have re- cently advanced more compelling arguments than in the past that (a) personality constructs, while abstractions of be- haviour, can be measured with reasonable reliability; (b) there is stability to personality measures over time and occasions; (c) Journal of Industrial Psychology, 2001, 27(3), 35-43 Tydskrif vir Bedryfsielkunde, 2001, 27(3), 35-43 35 personality measures are signi¢cantly related to some nontest criterion measures of performance; and (d) personality mea- sures are useful in predicting performance of candidate em- ployees in certain settings. Many other authors have argued that certain personality di- mensions, or patterns of dimensions, can be used in the pre- diction of successful sales people.There are however, di¡ering viewpoints on the role of personality in the prediction of per- formance, and also on which dimensions of personality can be used for prediction. Prior to the1990’s most researchers did not view personality as a valid predictor of job performance. Guion and Gottier (1965) concluded that there is no generalisable evidence that perso- nality measures could be recommended, or used, in most sit- uations as a basis for making employment decisions. Churchill, Ford, HartleyandWalker (1985) in their meta-study demonstrated that personality measures only accounted for 4% of the variance in outcome-based sales performance. Another meta-analysis of validation studies of personality measures showed that the average validity coe⁄cient across all situations and studies was a modest r = 0,21for performance rating criteria.The authors commented that while assessment centres, work samples and measures of cognitive ability sho- wed good validity, this could not be said for personality mea- sures. This conclusion was, however, arrived at on aggregated results for all criteria used.They found that personality measu- res predicted some criteria relatively well (Schmitt, Gooding, Noe & Kirsch,1984). Ford,Walker, Churchill and Hartley (1986) conducted an ana- lysis of all studies appearing between 1918 and 1982 in which the relationship between biographical characteristics, psycho- logical characteristics and sales performance were examined. Only a few personality variables were weakly related to suc- cess; most showed no relationship. They also found that as a class, personality factors were less predictive of sales perfor- mance than were biographical, cognitive or skill factors.Anot- her study by Gomer and Dubinsky (1985) also found that personality characteristics, amongst others, such as experien- ce/background factors and physical characteristics were poor predictors of sales performance. More recently (post 1990), however, a substantial body of evi- dence has emerged suggesting that personality traits can be used to predict job performance (Arneson, Milikin-Davies & Hogan, 1993; Barrick & Mount, 1991; Barrick, Mount & Strauss,1993; Dale,1995; Hogan, Hogan & Gregory,1992; Mu- chinsky,1993; Nell,1994; Piedmont & Weinstein,1994; Robe- rtson & Kinder,1993; Salgado,1997;Tett, Jackson & Rothstein, 1991;Verbeke,1994;Vinchur, et al.,1998). Barrick and Mount (1991) conducted and published a meta- analysis investigating the relation of the ‘‘big ¢ve’’ persona- lity dimensions (‘Neuroticism’, ‘Extraversion’, ‘Openness to Experience’,‘Agreeableness’and‘Conscientiousness’) to three job-performance criteria (job pro¢ciency, training pro- ¢ciency, and personnel data) across ¢ve occupational groups, namely professionals, police, managers, sales, and skilled/semi-skilled. A major ¢nding of this study was that one dimension of personality,’Conscientiousness’, proved to be a valid predictor (r > 0,20) of all job-related criteria for all occupational groups studied. Additionally, ‘Extraversion’ was found to be a valid predictor of two occupations (across all criterion types), namely managers and sales. Other re- sults indicate that the validity of conscientiousness as a pre- dictor of sales perfomance is generalisable across organisations (Barrick et al.,1993). According to Mount and Barrick (1998) this study was a major shift from the thinking at that time about non-cogni- tive predictors of performance. They commented that their study grew out of their belief that people have long-term, dispositional traits that in£uence their behaviour in work settings, even though most studies about the usefulness of personality measures in personnel selection were quite pessimistic. It was one of the ¢rst studies to introduce the ‘‘big ¢ve’’framework of personality (¢ve broad, stable traits that describe normal adult personality) into the ¢eld of in- dustrial-organisational psychology. Although this taxono- my was well known in the ¢eld of personality psychology, it was not universally accepted at the time of the study. Another meta-analytic study conducted by Tett et al. (1991) provided some optimism about the use of personality mea- sures for the prediction of performance. They assessed the overall validity of personality measures (eight dimensions, in- cluding the ‘‘big ¢ve’’) as predictors of job performance and investigated various moderator variables. Among the major ¢ndings were that, studies using con¢rmatory strategies pro- duced a corrected mean personality scale validity (0,29) that was larger than that based on exploratory strategies (0,12). Re- sults for the ‘‘big ¢ve’’ predictor constructs revealed corrected mean validities ranging from 0,16 for ’Extraversion’to 0,33 for ‘Agreeableness’. There were, however some inconsistencies between the results of these two studies. In two articles (Ones, Mount, Barrick & Hunter, 1994; Tett, Jackson, Rothstein & Reddon, 1994) the authors tried to explain these discrepant results. According to Mount and Barrick (1998) the result of the above-mentioned debate stimulated additional personality research and contri- buted to the recent strides made in understanding the role of personality measures in predicting job performance. A substantial body of research emerged supporting the‘‘big ¢- ve’’, and other models of personality, in predicting job perfor- mance. Salgado (1997), in a meta-analysis investigating the ‘‘big ¢ve’’model of personality in relation to job performance conducted in the European Community, found that ‘Con- scientiousness’and ‘Emotional Stability’are valid predictors of job performance across all criteria and occupational groups in- vestigated. Additionally, ‘Extraversion’ predicted job perfor- mance in jobs where interpersonal characteristics were likely to be important factors. By using 20 validation studies involving the Occupational Personality Questionnaire, and employing meta-analytic techniques, Robertson and Kinder (1993) explored the cri- terion-related validity of some personality variables. On ave- rage, results indicated mean sample size-weighted validity coe⁄cients of around 0,20 for personality variables. Higher values (to r = 0,33) were found for some criteria, i.e. ‘Crea- tivity’, ‘Judgement’ and ‘Analysis’. Salgado (1996), after com- menting on this study and indicating that some errors were made, reanalysed the same data. These conclusions also con- ¢rmed the criterion-related validity of the personality va- riables analysed. Barrick and Mount (1993), although investigating the mo- derating e¡ect of ’Autonomy’ between the ‘‘big ¢ve’’per-son- ality dimensions and job performance, also found that the dimensions of ’Conscientiousness’ (r = 0,25) and‘Extraversion’ (r = 0,14) were signi¢cantly related to job performance. Some researchers have investigated how sales performance could be predicted by making use of personality question- naires (Arneson et al., 1993; Barrick et al., 1993; Hogan et al., 1992; Piedmont & Weinstein, 1994;Vinchur et al, 1998). Pied- mont andWeinstein (1994) used the NEO Personality Inven- tory, an instrument speci¢cally designed to measure the ¢ve- factor model, to evaluate the prediction of supervisors’ratings of performance. The subjects were 52 women and 159 men who were engaged in a wide range of occupations, including customer services, sales, management, and ¢nance.The majo- rity of the sample was drawn from sales and customer services jobs (73%). The results indicated that ‘Conscientiousness’ predicted high ratings in all the performance areas across all occupational groups used. ‘Neuroticism’ and high ‘Extraversion’ scores also predicted high perfor- mance. These ¢ndings add to the emerging consensus LA GRANGE, ROODT36 that personality can make a substantive contribution to the prediction of job success. Considering that the majo- rity of the sample was from the sales and customer servi- ces areas, the ¢ndings could also indicate sales success. Other researchers have argued that it is possible to identify one or more underlying personality traits that can predict sa- les success. Seligman and Schulman (1986) found that a trait they call ‘Attributional style’ was related to sales performance and turnover in a sample of life insurance agents. According to the model of ‘Learned Helplessness’ (Abramson, Seligman & Teasdale, 1978; Seligman, Abramson, Semmel & von Baeyer, 1979), individuals with an optimistic ‘Attributional Style’ are more resilient when confronted by unfavourable events than are individuals with a pessimistic ’Attributional Style’. In a concurrent validation study (N = 94) of life insu- rance agents, those with an optimistic style sold 37% more insurance during the ¢rst two years of their careers than those with a more pessimistic style. A second prospective validation study (N = 104) showed that newly hired agents with an op- timistic style remained in their jobs twice as long, and sold more insurance, than those with a pessimistic style. Corr and Gray (1996) found that, in a sample of 130 experienced sales people, a positive ’Attributional Style’ positively correlated with sales (de¢ned in monetary terms) and performance ran- king within the sales force. Research onType A behaviour indicates that, in a study invol- ving life insurance sales people, various forms of ‘Achievement Orientation’ predict work performance (number of policies sold) and job satisfaction (Bluen, Barling & Burns, 1990). Re- searchers investigating‘Locus of Control’ have found that this trait has a signi¢cant role in the prediction of the work perfor- mance of sales people (Bothma & Schepers, 1997, Coetzer & Schepers, 1997). Other authors have proposed that the ability to elicit information from others, to self-monitor during con- versations, and to adapt during conversations, are good predic- tors of sales performance (Verbeke, 1994). Still others have postulated that for sales people to obtain results they need enough empathy, su⁄cient ego drive, as well as a strong servi- ce motivation (Greenberg & Amabile, 1996). Hogan, Hogan and Busch (1984) and Parasuraman, Berry and Zeithaml (1991) emphasize the construct of ‘Service Orientation’ (attitu- des and behaviours that a¡ect the qualityof interaction betwe- en sta¡ and customers) as one that underpins both sales and customer service roles. It therefore appears possible to identify one or more dimen- sions of behaviour that underlie particular types of perfor- mance on the job. It seems more likely, though, that di¡erent dimensions of underlyingbehaviour, or personality, will relate to di¡erent aspects of job performance. Adetailed pro¢le of an individual’s personality, with his or her cognitive ability, should provide more detailed information on how well that individual would perform in a sales role. Cognitive Ability There is much evidence to indicate that tests of cognitive ability are strong predictors of job performance in virtually every job studied (Wigdor & Garner, 1982; Hunter & Hunter, 1984; Schmidt, Hunter & Outerbridge, 1986; Nathan & Alexander, 1988; Ree & Earles, 1992; McHenry, Hough,Toquam, Hanson & Ashworth, 1990; Ree & Earles, 1994; Robertson & Kinder, 1993; Gottfredson,1997; Hakstian, Scratchley, Mcleod,Tweed & Siddarth,1997). In studies conducted by Hunter and Hunter (1984) and Ree and Earles (1992) the average validity coe⁄cient increased to the r = 0,50 level when observed validity coe⁄cients were corrected for measurement artifacts, such as restriction of range and mea- surement error. According toWigdor and Garner (1982), theave- rage correlation between cognitive ability tests scores and job performance ranges from r = 0,20 to r = 0,30. Thurstone (1938) proposed seven components that comprise intelligence or cognitive ability. These components are the ‘‘primary mental abilities‘‘ (see Thurstone, 1938) of an indivi- dual, namely ‘Verbal Comprehension’,’Verbal Fluency’,‘Num- ber Ability’,‘Spatial Visualisation’,‘Memory’,‘Form Perception’, and‘Inductive Reasoning’. There is a major school of thought which supports a view that mental ability tests provide single, global measures of intel- ligence (Caretta & Ree, 1996, Ree & Earles, 1992). Several re- cent studies have demonstrated that General Intelligence (psychometric ’g’), which is generally referred to as the com- mon variance in a battery of cognitive ability tests, accounts for the majority of variance in performance prediction. It was also shown that the remaining variance (often referred to as ‘‘speci¢c abilities’’) accounts for little or no additional variance in the criterion (Larson & Wolfe,1995; Ree, Earles & Teachout, 1994). Furthermore, research by Olea and Ree (1994) and Ree et al. (1994) suggests that speci¢c abilities account for some- what more variance when the criterion is job performance than when it is training performance. The‘‘speci¢c abilities’’school of thought has been critical of the General Intelligence approach. They are researchers seeking a ¢ner delineation of mental abilities by making use of multiple factor analysis. These individual abilities underlie the more modern multiple factor theory of intelligence (Guion & Gib- son, 1988; Landy, Shankster & Kohler, 1994). According to Schepers (1999) the predictive validity of multiple factors is al- ways better than that of ‘g’alone. Murphy (1996) argues that it is just as valid to enter speci¢c abilities ¢rst and then say psy- chometric ‘g’doesn’t contribute beyond the prediction found with‘‘speci¢c abilities’’alone. In this regard Muchinsky (1993) found, in a sample of manufacturing jobs, that ‘Mechanical Ability’ was the single best predictor of job performance, and General Intelligence had no incremental validity beyond the ’Mechanical Ability’test alone. Job Performance The domain of job performance is multifaceted and complex in nature. It has also been the most widely used criterion in the ¢eld of Applied Psychology (Adler,1996). Job performance can be measured in many ways, from one-dimensional to multi- dimensional conceptualisations.Wagner (1997) comments that a trend in recent studies has been to use more complex concep- tualisations of this criterion. Campbell (1994) provides a good example of such a multi- dimensional taxonomyof job performance. Performance is di- vided into eight basic components, namely: ‘Job-speci¢c Task Pro¢ciency’;‘Non-job-speci¢cTask Pro¢ciency’; ‘Written and Oral Communication Task Pro¢ciency’; ’Demonstration of E¡ort’; ’Maintenance of Personal Discipline’; ‘Facilitation of Peer and Team Performance’; ‘Supervision ^ Leadership’ and ’Management ^ Administration’. Not all jobs would necessari- ly contain all eight components. In some instances, the num- ber of relevant components could increase depending upon the nature of the job (Wagner,1997). The method of assessment of job performance is also an im- portant factor. A particularly salient distinction between crite- ria is the objective method and the subjective method of measuring job performance (Vinchur et al.,1998). Gottfredson (1991) noted that the vast majorityof validation studies for pre- dicting job performance have made use of the more subjective, ‘supervisor ratings’rather than more direct, objective measures. The di¡erent methods mentioned also result in di¡erences in research results. Barrick and Mount (1991) showed that, on average, personality variables are correlated more strongly with subjective appraisals of job performance (average validi- ty; r = 0,26) than with objective criteria (average validity; r = 0,14). For the role of a sales person objective criteria focus more on outcomes-based e¡ectiveness (‘sales volumes’), while the sub- jective ratings focus more on the controllable parts of a sales person’s job, such as ‘organisational citizenship behaviours’ (Campbell, McCloy, Oppler & Sager, 1993). Vinchur et al. 37PREDICTORS OF THE JOB PERFORMANCE (1998), in conducting a meta-analysis of the predictors of job performance for sales people, found that a very small number of studies used performance criteria other than objective ’sales volume’and’managerial ratings’of sales person performance. Studies involving Personality and Cognitive predictors of Job performance in Sales In a validation study, conducted by Hogan et al. (1992),127 sa- les representatives were asked to rate themselves on the Hogan Personality Inventory (HPI). This instrument measures the personal and social competencies shown to be of considerable importance (indicated through job analysis) in the role of a sa- les person. A second predictor, the Short Employment Test (SET) ^ a test of ‘Verbal Pro¢ciency’ (an indicator of’General Cognitive Ability’), was also used.‘Verbal Pro¢ciency’was hy- pothesized as being important for predicting those aspects of the sales representative’s job that concern maintaining sales knowledge and verbal communication skills. The criterion measures included two subjective ratings (‘managerial ratings’ and ‘categorisation’), and one objective measure (’sales revenue produced’). The results indicated that the test of ‘Verbal Pro¢ciency’ gene- rally had low correlations with the criterion measures, with only one of 18 correlations being signi¢cant at the p = 0,05 le- vel. It was found that all correlations of the personality varia- bles with the criteria were signi¢cant, ranging from r = 0,19 to r = 0,53.The strongest correlation found was with the total‘sa- les revenue produced’criterion. Another study that used personality and cognitive measures to predict job performance in the insurance industry, is that conducted by Arneson et al. (1993). The sample consisted of 50 insurance claims examiners. Job Analyses suggested that both cognitive and personality measures were necessary for successful performance. Criterion data included ‘supervisory ratings’, ‘employee nominations’, ‘average percent of per- formance’,‘absences’,‘disciplinary actions’, and ‘sick leave’.This was a concurrent validity study employing two measures of personality and four cognitive ability tests. Scores from the personality measures and three of the four cognitive tests correlated signi¢cantly with the average percentage of per- formance, and six of the ‘supervisory ratings’. In addition to the cognitive measures, the personality measures contributed signi¢cantly to the prediction of percentage of performance achieved with r = 0,64. In a study by Barrick et al. (1993) results showed that ‘Con- scientiousness’ (one of the‘‘big ¢ve’’dimensions) is directly rela- ted to the criterion of ’supervisor ratings’ (r = 0,23). ‘General Mental Ability’ was also related to ’supervisory ratings’ (r = 0,34) and to a lesser extent to’sales volumes’ (r = 0,16). In a recent meta-analytic review of predictors of job perfor- mance for sales people,Vinchur et al. (1998) found that certain personality variables predicted sales performance well. ‘Po- tency’ (a sub-dimension of ‘Extraversion’) predicted ‘super- visors’ ratings’ of performance (r = 0,28) and objective mea- sures of ‘sales’ (r = 0,26). ‘Achievement’ (a component of‘Con- scientiousness’) predicted‘supervisor’s ratings’ (r = 0,25) and’sa- les’ (r = 0,41). Furthermore, ‘General Cognitive Ability’ correlated with ’supervisor’s ratings’ (r = 0,40) but only r = 0,04 with’sales’measures. On the basis of the literature reviewed, the following hypo- theses are proposed: H1: Personality measures can predict job performance. H2: Measures of cognitive ability can predict job per- formance. METHOD Sample This research, using a concurrent validity strategy, was con- ducted in a large life insurance company in South Africa.The population consists of 199 broker consultants (third-party selling) that is geographically spread throughout SouthAfri- ca representing all major regions.The sample consisted of170 broker consultants who were available for testing. The sample consistsof131 (77,1%) males and 39 (22.9%) females. The language spoken most often is more evenly distributed: 96 (56,5%) are Afrikaans speaking, 73 (42,9%) English speaking, and one (0,6%) German Speaking.The level of education ranged from standard eight (grade 10) to post-graduate quali¢cations, with job experience (within four seniority levels) averaging 50,78 months (roughly speaking, just over 4 years). The ages of the consultants ranged from 23 to 62, with a mean age of 32,61 years. The racial distribution; 92,9% ^ White, 5,3% ^ Asian, 0,6% ^ Coloured, and 1,2% ^ African, does not represent the cultural diversity of the South African Business community. There was a form of pre-selection applied to this sample con- sisting of interviews conducted after passing a subjective, bio- graphically orientated, screening method taking into consideration aspects such as: ¢nancial situation, marital status, age, length of tenure in previous job, consequences of failure, ¢nancial liabilities, social and sports involvement, etc. Measuring Instruments Given the hypotheses the following instruments were chosen for the operationalisation of the variables. Predictors The Customer Contact Styles Questionnaire (CCSQ 5.2) of Saville & Holdsworth Limited (SHL) was used as the person- ality predictor. This questionnaire focuses on 16 dimensions of personality that are considered important for non-super- visory sales or customer services roles.The CCSQ 5.2 version is a normative measurement and therefore lends itself more to correlational studies. This version has 136 questions, and they are answered by making use of a ¢ve-point Likert scale, ranging from‘‘strongly agree’’to‘‘strongly disagree’’. In addi- tion, a ’Social Desirability’ scale is included as an accuracy check. The reported reliabilities range from r = 0,69 to r = 0,88 with a median value of r = 0,81. Several validity studies are reported in the User’s Manualwith positive results (SHL,1997, chapter 7, pp. 8-9). TheVerbal Evaluation test (VCC3 ^ SHL), a measure of cog- nitive ability, was used as the other predictor. This test mea- sures the ability to understand and evaluate the logic of more complex written arguments. It consists of 15 passages, follo- wed by four statements related to the information therein. After reading each passage, individuals are required to eva- luate each statement in terms of whether it follows logically from the passage, or not, or whether there is insu⁄cient in- formation to make such a judgement.There is a time limit of 30 minutes to the test.The content of the test re£ects a gene- ric customer contact focus. The level of di⁄culty of the test was suited to the individuals in the sample. Reported reliabi- lities (Cronbach Alpha) for a sample of 700 are reported in the User’s manual as being in the order of 0,85 (SHL, 1997, chapter 7, pp. 6). Criterion The Customer Contact Competency Inventory (CCCI ^ SHL) was used as the criterion measure.The CCCI provides a direct rating of an individual’s performance based on 16 custo- mer-orientated competencies (see Table 1). The instrument al- lows one to make use of 360-degree ratings, but in the current study, only ’managerial ratings’ were used as a subjective mea- sure of sales performance. An electronic version of the CCCI was used to gather crite- rion ratings.The item format for this instrument is ‘‘nipsative’’ (a combination of normative and ipsative item formats). Re- spondents completing the questionnaire rate an individual on 32 sets of four statements. Each set of four statements is indivi- dually rated by making use of a ¢ve-point Likert scale ^ LA GRANGE, ROODT38 TABLE 1 CUSTOMER CONTACT COMPETENCY INVENTORY (CCCI) -COMPETENCIES Number Label Competency 1. CP___01 Relating to Customers 2. CP___02 Convincing 3. CP___03 Communicating Orally 4. CP___04 Communicating In Writing 5. CP___05 Team Working 6. CI___01 Fact Finding 7. CI___02 Problem Solving 8. CI___03 Business Awareness 9. CI___04 Specialist Knowledge 10. CD___01 Quality Orientation 11. CD___02 Organisation 12. CD___03 Reliability 13. CE___01 Customer Focus 14. CE___02 Resilient 15. CE___03 Results Driven 16. CE___04 Using Initiative TABLE 2 ROTATED FACTOR MATRIX OF THE 3 FACTORS OF THE CCCI (DIRECT OBLIMIN WITH KAISER NORMALISATION) Factor Description I II III CI_04 CI_01 CI_02 CI_03 CP_02 CE_04 CE_03 CP_04 CP_03 CP_01 CP_05 CE_01 CE_02 CD_03 CD_02 CD_01 Specialist Knowledge Fact Finding Problem solving Business awareness Convincing Using initiative Results driven Communicating in writing Communicating orally Relating to Customers Team Working Customer Focus Resilience Reliability Organisation Quality orientation 0,911 0,896 0,875 0,858 0,752 0,739 0,535 0,526 0,515 -0,135 0,170 0,264 0,380 0,474 -0,239 0,259 0,143 0,356 0,288 0,817 0,760 0,568 0,378 0,375 0,234 0,162 0,136 -0,342 0,164 0,160 -0,329 0,220 0,133 0,713 0,653 0,496 Values smaller than 0,10 were omitted ranging from‘‘hardly ever’’ to‘‘always’’.Thereafter,‘‘most’’and ‘‘least’’rankings (ipsative format) are given for each set of four statements. For the purposes of this study, however, only the normative data were used. Reliabilities reported in the User’s Manual for managers (N = 365) who rated individuals, range from r = 0,76 to r = 0,92 (SHL ,1997, chapter 7, pp.13). Procedure Job analysis was used to identify the competencies required for success as a broker consultant. A countrywide sample, con- sisting of all levels of consultants and including managers, was used. Four methods of job analysis were used, namely: the Re- pertory Grid Technique, the Critical Incidents Method, the Work Pro¢ling System (WPS) from SHL, andVisionary Inter- views. Fromthis analysis competencies were developed for the role of a broker consultant.These competencies resemble most of the competencies identi¢ed in the CCCI model (Refer toTable 1). The competencies were ranked into groups of di¡ering im- portance, i.e. Essential, Important, and Relevant. From the job analysis the CCSQ 5.2 was chosen as a predictor for the job competencies. Following the view of di¡erential aptitudes or ’speci¢c abilities’, rather than that of ’General Reasoning Ability’,’Verbal Reasoning’ was identi¢ed as an essential attri- bute for this customer contact role.To measure this,VCC3 was chosen as the other predictor. To administer these instruments SHL accredited test admini- strators (TA’s) were used. The predictors were administered in controlled test-room conditions using standardised proce- dures. All respondents completed a biographical question- naire. All managers and their consultants were briefed about the purpose of the project beforehand and the subsequent in- volvement of managers in providing ratings of the consultants reporting to them. It was also pointed out that these ratings were to be used solely for research purposes (to validate a test battery), and that it would not have anye¡ect on their remune- ration, promotion or careers. Before actually rating their con- sultants, managers were coached by the TA’s in the use of the electronic version of the CCCI. The data were gathered form theTA’s, and statistically analysed by researchers at SHL South Africa under a license agreement with the insurance company. The data were then statistically analysed by the Statistical Consultation Service of the Rand Afrikaans University, for the purposes of this study. RESULTS In order to determine the structure of the CCCI it was decided to subject the 128 normative items to FactorAnalysis. In order to obviate the e¡ects of di¡erential skewness of items, the fol- lowing procedure was followed:The128 normative items were inter-correlated and the eigenvalues of the unreduced inter- correlation matrix were calculated. Based on the Kaiser (1961) criterion (number of eigenvalues greater than unity),16 simpli¢ed factor scores (SFS) were pos- tulated. Accordingly 16 factors were extracted, using the Prin- ciple Axis Factoring technique, and they were rotated to a simple structure by means of theVarimax rotation. Subsequently, SFS’s were calculated for each of the 16 factors that was extracted byadding the scoresofthe itemswith high loadings on each factor. Finally, the16 factors were inter-correlated, subjec- ted to a PrincipleAxisFactoring procedure and rotated toa simple structure by means of the Direct Obliminrotationwiththe Kaiser Normalisation. Three eigenvalues larger than unity were extra- cted and were 8,341,1,772 and1,438 respectively. The rotated factor matrix (with descriptions of the16 SFS’s in- cluded) of the factors obtained are shown inTable 2. Table 2 shows that factor I is well determined by the following competencies: ‘Specialist Knowledge’,‘Fact Finding’,‘Problem solving’, ‘Business awareness’, ‘Convincing’, ‘Using initiative’, ‘Results driven’,‘Communicating inwriting’,‘Communicating orally’. This factor is identi¢ed as Business and Sales Acumen. Factor II, Relating to Customers, consists of the competencies of ‘Relating to customers’,‘Team working’,‘Customer focus’, and ‘Resilience’. Factor III, consisting of ‘Reliability’, ‘Organisa- tion’ and ‘Quality orientation’, is identi¢ed as Dependability. The three obtained factors correlate moderately with each ot- her and vary between 0,482 and 0,201. Reliabilities (Cronbach coe⁄cient alpha) for the three cri- terion factors are r = 0,977, r = 0,946 and r = 0,950 respectively, with an average reliability of r = 0,957. The correlation matrix (18X3) below (Table 3) shows the cor- relations between various predictor dimensions and the three criterion scales. FromTable 3 it is evident that the correlations vary between low positive and negative values, with only a few that are statistically signi¢cant. In order to examine the two hypotheses that were postulated earlier, a Stepwise Linear Regression Analysis was conducted. First, Factor I (Business and Sales Acumen) was included as the dependant variable with the 16 personality dimensions of the Customer Contact Styles questionnaire and ‘Verbal Reason- ing Ability’as independent variables. From an inspection of Table 4, it is evident that three dimen- sions of the Customer Contact Styles Questionnaire 5.2 (CCSQ 5.2), namely ‘Competitive’, ‘Sociable’and ‘Participative’, explain 12,7 % of the variance of Factor I (‘Business and Sales Acumen’) of the Customer Contact Competency Inventory 39PREDICTORS OF THE JOB PERFORMANCE TABLE 3 CORRELATION MATRIX (18X3) OF PREDICTOR DIMENSIONS (CCSQ 5.2 AND VCC3) WITH THE CRITERIA (3 FACTORS OF THE CCCI) Factor 1 Factor 2 Factor 3 1. Persuasive. 0,127 -0,066 -0,237** 2. Self-control. -0,063 0,015 0,111 3. Empathic -0,036 0,017 -0,036 4. Modest -0,071 -0,091 0,139 5. Participative -0,164* 0,073 -0,063 6. Sociable 0,247** 0,179* -0,093 7. Analytical 0,124 0,025 0,077 8. Innovative 0,178* 0,064 -0,109 9. Flexible 0,107 0,033 0,015 10. Structured 0,079 0,020 0,130 11. Detail Conscious. 0,034 0,097 0,134 12. Conscientious 0,065 0,099 0,093 13. Resilience 0,081 -0,005 -0,043 14. Competitive 0,257** 0,060 -0,115 15. Results Orientated 0,217** 0,126 -0,086 16. Energetic 0,192* 0,158 -0,025 17. Social Desirability -0,100 0,045 -0,083 18. VCC 3 ^ Verbal Evaluation 0,159 0,101 0,114 ** Correlation is signi¢cant at the 0,01 level (2 tailed). * Correlation is signi¢cant at the 0,05 level (2 tailed). TABLE 4 REGRESSION OF PERSONALITYAND VERBAL REASONING ABILITY ON FACTOR I ^ BUSINESS AND SCALES ACUMEN (DEPARTMENT VARIABLE) Analysis of Variance Multiple correlation 0,381 Source of Variance Degrees of freedom Sum of squares Mean Square R square 0,145 Regression 3 27304,9 9101,628 Adjusted R square 0,127 Residual 142 160601 1130,994 Standard Error of the Estimate 33,630 F = 8,047; p = 0,000 Variables in the Equation Independent variables B Std error of B Beta t-value p Constant 211,694 14,318 14,785 0,000 Competitive 3,614 1,608 0,182 2,248 0,026 Sociable 4,978 1,654 0,252 3,010 0,003 Participative -3,735 1,362 -0,220 -2,741 0,007 TABLE 5 REGRESSION OF PERSONALITYAND VERBAL REASONING ABILITY ON FACTOR II ^ RELATING TO CUSTOMERS (DEPENDENT VARIABLES) Analysis of Variance Multiple correlation 0,341 Source of Variance Degrees of freedom Sum of squares Mean Square R square 0,099 Regression 3 3691,496 1230,499 Adjusted R square 0,080 Residual 142 33635,6 236,871 Standard Error of the Estimate 15,391 F = 5,195; p < 0,002 Variables in the Equation Independent variables B Std error of B Beta t-value p Constant 111,670 5,654 19,751 0,000 Sociable 2,315 0,843 0,262 2,747 0,007 Persuasive -3,296 1,036 -0,349 -3,182 0,002 Results Driven 1,828 0,835 0,228 2,190 0,030 TABLE 6 REGRESSION OF PERSONALITYAND VERBAL REASONING ABILITY ON FACTOR III ^ DEPENDABILITY (DEPENDENT VARIABLES) Analysis of Variance Multiple correlation 0,303 Source of Variance Degrees of freedom Sum of squares Mean Square R square 0,092 Regression 2 2606,511 1303,256 Adjusted R square 0,079 Residual 143 25871,9 180,923 Standard Error of the Estimate 13,45 F = 7,203; p < 0,001 Variables in the Equation Independent variables B Std error of B Beta t-value p Constant 92,503 4,987 18,548 0,000 Persuasive -2,309 0,674 -0,281 -3,425 0,001 Structured 1,321 0,560 0,193 2,357 0,020 (CCCI).The multiple correlation of R = 0,381obtained is sta- tistically signi¢cant: as shown in the analysis of variance F (df = 3; 142) = 8,047; p (F) = 0,000.The three predictors are statis- tically signi¢cant in the regression equation.The following re- gression equation was computed: Y’ = (3,614) Competitive + (4,978) Sociable + (- 3,735) Partici- pative + (211,694) The same procedure was carried out for criterion Factors II (Relating to Customers) and III (Dependability).The results are gi- ven inTables 5 and 6, respectively. Table 5 indicates that one of the same dimensions used in the above regression, namely ‘Sociable’and two other dimensions of the CCSQ 5.2, namely ‘Persuasive’and ‘Results Orientated’ explain 8% of the variance of Factor II (‘Relating to Cus- tomers’) of the CCCI. The multiple correlation of r = 0,341 obtained is statistically signi¢cant: F (df = 3;142) = 5,195; p (F) < 0,002.The three predictors are statistically signi¢cant in the regression equation. The following regression equation was computed: LA GRANGE, ROODT40 Y’ = (2,315) Sociable + (- 3,296) Persuasive + (1,828) Results orientated + (111,67) Table 6 shows that‘Persuasive’and‘Structured’ (CCSQ 5.2) ex- plain 7,9% of the variance of Factor III (Dependability) of the CCCI.The multiple correlation of 0,303 is also statistically sig- ni¢cant: F (df = 2; 143) = 7,203; p(F) < 0,001.These three pre- dictors were also statistically signi¢cant in the regression equation.The following regression equation was computed: Y’ = (-2,309) Persuasive + (1,321) Structured + (92,503) One should, however, note that the sample consists ofapre-selec- ted group of sales people, which will have a restrictive e¡ect on therange ofobtained scores. Obtained R2 valueswouldprobably be in£ated if adjusted for the restriction of range. DISCUSSION Based on the results of this study, the ¢rst hypothesis, ‘‘Perso- nality measures predict job performance’’, is not rejected. For the Business and Sales Acumen criterion ‘Competitive’, ‘So- ciable’and low ’Participative’ (negative regression coe⁄cient) personality attributes explain 12,7% of the variance. Accord- ing to the scale descriptions in the CCSQ 5.2 user’s manual, ‘Competitive’ is concerned with how much individuals feel they need to win at all costs. Individuals may want to seek out competition and may put in much e¡ort to beat others. ‘Sociable’describes how con¢dent, extraverted and lively indi- viduals are, and how comfortable they feel in a range of social situations.’Participative’deals with the degree to which indivi- duals enjoy teamwork and co-operative activities.Typical low scores would indicate that individuals enjoy working alone and that they are very much self-su⁄cient (SHL ,1997, chapter 2, pp.17-32). ‘Competitiveness’ is a personality trait that one would expect to have a prominent in£uence in sales performance, especially in the area of direct sales.This is con¢rmed by validity studies referred to in the SHL User’s Manual for the Customer Con- tact series of products. In these studies ‘Competitiveness’ pre- dicted the competencies of ‘Convincing’, ‘Problem Solving’ and ‘Results Driven’, which are all included in the criterion of Business and Sales Acumen. Inter-correlations between‘Com- petitiveness’ (CCSQ 5.2) and ’Achieving’ (Occupational Per- sonality Questionnaire ^ OPQCM 5.2) are reported to be in the order of r = 0,38 (at the p = 0,05 level) (SHL ,1997, appen- dix D, pp. 4). Vinchur et al. (1998) also showed that the ‘‘big ¢ve’’sub-dimension of ’Achievement’ (r = 0,25) predict mana- gerial ratings of sales success. The fact that ‘Sociable’ is signi¢cantly related to sales perfor- mance is in line with many other ¢ndings about the dimen- sion of ‘Extraversion’ (Barrick & Mount, 1991; Barrick & Mount,1993; Piedmont & Weinstein,1994; Salgado,1997;Tett et al., 1991). In a more recent meta-analytic review,Vinchur et al. (1998) found that ‘Extraversion’ predicted ratings of sales success (r = 0,18).‘Potency’ (assertiveness and intensity of inter- personal interactions) was a particularly strong predictor of sa- les success (r = 0,28). The negative correlation coe⁄cient for‘Participative’ (seeTable 3) was not expected because job analysis showed that ‘Team Working’ is an essential competency in the role of a broker consultant. This result shows that the absence of participative behaviour (individualism) predicts job performance.This was con¢rmed byVinchur et al. (1998) that ‘rugged individualism’ could be a predictor of sales success. The CCSQ 5.2 dimensions‘Sociable’,‘Persuasive’and ‘Results Orientated’ explain 8% of the variance of the Relating to Customers criterion.‘Persuasive’concerns with the extent to which individuals enjoy selling, negotiating, in£uencing and convincing.‘Results orientated’ indicates the extent to which individuals set high personal targets, how much they are sti- mulated by challenging goals, and how keen they are to im- prove their performance (SHL,1997, chapter 2, pp.17-32). The fact that ‘Sociable’ and ‘Results Orientated’ predicts Re- lating to customers, makes sense. Evidence for the role of ‘So- ciable’ in predicting performance has already been provided (see above).‘Results orientated’has the same meaningas‘Achie- vement’ (Vinchur et al., 1998) or ‘Ambition’ (Hogan et al., 1992). These studies provide signi¢cant validity coe⁄cients ^ ‘Achievement’ (r = 0,25 and r = 0,41) and ‘Ambition’ (r = 0,15 and r = 0,25). The CCSQ 5.2 dimensions‘Persuasive’and‘Structured’explain 7,9% of the variance in the criterion of Dependability. ‘Struc- tured’refers to the extent to which individuals plan ahead and how far they prepare, prioritise and structure their work (SHL,1997, chapter 2, pp.17-32). The role of ‘Persuasive’ in the criteria of Relating to Customers and Dependability, in this study could not be con¢rmed by the ¢ndings of other research. Relating to Customers has a small non-signi¢cant negative correlation coe⁄cient (see Table 3). In the case of Dependability this dimension has a signi¢cant (at the 1% level) negative correlation (seeTable 3). In both ca- ses, one would have expected signi¢cant positive correlations, indicating‘Persuasiveness’to be an important personality trait in the job success of a sales person. A possible explanation, for the Relating to Customers criterion, could be that the role of the broker consultant is mainly focussed on third-party selling, as opposed to direct sales. In third-party sales the consultants mainly deal with customers who are well known to them. Also, there is not always a direct link between the actual sales (in monetary terms) and the persuasive e¡orts of the con- sultant. Often there are other factors that in£uence the sales e¡ort. It appears that for a sales consultant to be viewed as De- pendable, highly ‘Persuasive’ behaviour is not seen as appro- priate. The fact that ‘Structured’ behaviour predicts the Dependability criterion makes intuitive sense. There is a substantial body of evidence indicating that the ‘‘big ¢ve’’ personality dimension of ‘Conscientiousness’, of which ‘Structured’ is a construct, predicts job performance (Barrick & Mount, 1991; Barrick et al., 1993; Piedmont & Weinstein, 1994; Salgado, 1997; Tett et al.,1991). Other studies speci¢cally involving sales people con- ¢rm the above (Hogan et al.,1992;Vinchur, et al.,1998). It was expected that Verbal reasoning ability would signi- ¢cantly predict the performance of sales people, as the verbal communication component of a sales role is very important. This was also con¢rmed in the ¢ndings of the Job Analysis performed prior to selecting this performance predictor. The second hypothesis of this study is, however, rejected based on the results of this study: In other words, measures of cogni- tive ability do not predict job performance.The cognitive abi- lity predictor (‘Verbal Reasoning Ability’) did not enter anyof the Multiple Regression analyses performed. This ¢nding stands in contrast to studies discussed earlier (Gott- fredson, 1997; Hakstian et al., 1997; Hunter & Hunter, 1984; McHenry et al.,1990; Nathan & Alexander,1988; Ree & Earles, 1992; Ree & Earles,1994; Robertson & Kinder,1993; Schmidt et al.,1986;Vinchur et al,1998;Wigdor & Garner,1982). The rejection of the second hypothesis seems to con¢rm the results of Sackett, Gruys and Ellingson (1998) and Hogan et al. (1992). Although the percentage of variance explained by the dif- ferent personality dimensions in job performance is relatively low, (7-12%), the ¢ndings of this study support the general dictum, ‘‘Behaviour is a function of the individual and his/ her environment’’. One should therefore bear in mind that numerous contextual factors contribute to variance in job be- haviour. In view of the small proportion of variance that is 41PREDICTORS OF THE JOB PERFORMANCE accounted for, further research is needed to improve both cri- terion and predictor variables. There are limitations to this study that should be highlighted. The sample size is relatively small and the study is limited to insurance sales people (broker consultants involved in third- party sales) in one speci¢c organisation.The ¢ndings can con- sequently not be generalised to other industries, or to other types of sales roles, e.g. insurance agents involved in direct sa- les. Another limitation could be the use of subjective‘anagerial ratings’of sales performance to the exclusion of objective ’sales measures’ (actual sales volumes). Future research should study the di¡erences in performance prediction between sales people in third-party selling roles, and sales people in direct-selling roles. Another area of re- search that should be considered is the di¡erent criteria that one uses to measure sales performance. Currently the criteria seem to be limited to ‘managerial ratings’ (subjective criteria) and’sales performance’ (objective criteria). Finally, it has to be borne in mind that attempts should be ma- de to avoid the likely methodological rami¢cations of mono- method bias in future research in this area. It is obvious that complying with this methodological exigency will involve a larger, more involved, and consequently more expensive, re- search undertaking. REFERENCES Abramson, L.Y., Seligman, M.E.P. & Teasdale, J. (1978). Learn- ed helplessness in humans: critique and reformulation.Jour- nal of Abnormal Psychology,87, 49-74. Adler, S. (1996). Personality and work behaviour: exploring the linkages. Applied Psychology: An International Review, 45(3), 207-224. Allport, G. W. (1937). Personality: a psychological interpretation. NewYork: Holt. Arneson,S., Millikin-Davies, M. & Hogan, J. (1993).Validation of personality and cognitive measures for insurance claims examiners. Journal of Business and Psychology, 7(4), 459-473. Barrick, M.R. & Mount, M.K. (1991).The big ¢ve personality dimensions and job performance: a meta analysis. Personnel Psychology, 35, 281-322. Barrick, M.R. & Mount, M.K. (1993). Autonomy as a mode- rator of the relationships between the big ¢ve personality dimensions and job performance. Journal of Applied Psycho- logy, 78(1),111-118. Barrick, M.R., Mount, M.K. & Strauss, J.P. (1993). Conscien- tiousness and performance of sales representatives: test of the mediating e¡ects of goal setting. Journal of Applied Psy- chology, 78(5), 715-722. Bluen, S.D., Barling, J. & Burns,W. (1990). Predicting sales per- fomance, job satifaction and depression by using the achie- vement strivings and impatience-irritability dimensions of type A behaviour. Journal of Applied Psychology, 75(2), 212-216. Bothma, A.C. & Schepers, J.M. (1997). The role of locus of control and achievement motivation in the work perfor- mance of black managers. Journal of Industrial Psychology, 23(3), 44-52. Campbell, J.P. (1994). Alternative models of job performance and their implications for selection and classi¢cation. In M.G. Rumsey, C.B.Walker & J.H. Harris (Eds), Personnel selection and classi¢cation (pp. 33 ^ 52). Hillsdale, NJ: Erlbaum. Campbell, J.P., McCloy, R.A., Oppler, S.H., & Sager, C.E. (1993). Atheory of performance. In N. Schmidt,W.C. Bor- man, Personnel selection in organizations (pp. 35 ^ 70). San Francisco: Jossey-Bass. Caretta,T.R. & Ree, M.J. (1996). US Air Force pilot selection tests: what is measured and what is predictive? Aviation, Space and Environmental Medicine,67(3), 279-283. Churchill, G., Ford, N., Hartley, S. & Walker, O. (1985). The determinants of salesperson performance: a meta analysis. Journal of Marketing Research,103-118. Coetzer, E.L. & Schepers, J.M. (1997). Die verband tussen lo- kus van beheer en die werksprestasie van swart bemarkers in die lewensversekeringsbedryf. Journal of Industrial Psycho- logy, 23(1), 34-41. Connellan,T.K. & Zemke, R. (1993). Sustaining knock your socks o¡ service. NewYork: Amacom. Corr, P.J. & Gray, A.G. (1996). Attributional style as a per- sonality factor in insurance sales performance in the UK. Journal of Occupational and Organisational Psychology,69, 83-87. Dale, G.J. (1995). Mental alertness, personality traits and work perfor- mance in the selection of supervisors. Unpublished master’s dis- sertation, University of South Africa, Pretoria. DeVilliers, A.S. & Slabbert, J.S. (1996).The South African organi- sational environment. Unpublished manuscript, Rand Afri- kaans University, Johannesburg. Ford, N.M., Walker, O.C., Jr., Churchill, G.A.,Jr., & Hartley, S.W. (1986). Selecting successful salespeople: A meta-analysis of bio- graphical and psychological selection criteria. Graduate School of BusinessWorking Paper, University of Wisconsin. Gomer, J.M., & Dubinsky, A.J. (1985). Managing the successful sales force. Lexington, MA: Lexington Books. Gottfredson, L.S. (1991).The evaluation of alternative measures of job performance. In A.K.Wigdor & B.F. Green, Jr. (Eds), Performance assessment in the workplace (Vol. 1, pp. 75 ^ 126). Washington, DC: National Academy Press. Gottfredson, L.S. (1997). Why g matters: the complexity of everyday life. Intelligence, 24(1), 79-132. Greenberg, H.M. & Amabile, D.T. (1996).The personality of a top salesperson. Agency Sales Magazine, 26(11), 40-41. Guion, R. & Gottier, R. (1965).Validity of personality measures in personnel selection. Personnel Psychology,18,155-164. Guion, R.M. & Gibson,W.M. (1988) Personnel selection and placement. Annual Review of Psychology, 39, 349-74. Hakstian, A.R., Scratchley, L.S., Mcleod, A.A.,Tweed, R.G. & Siddarth, S. (1997). Selection of telemarketing employ- ees by standardized assessment procedures. Psychology and Marketing,14(7), 703-726. Hogan, J., Hogan, R. & Gregory, S. (1992).Validation of a sales representative selection inventory. Journal of Business and Psychology, 6(2),161-171. Hogan, J.C., Hogan, R. & Busch, C.M. (1984). How to measu- re service orientation. Journal of Applied Psychology, 69(1), 167-173. Hogan, R. & Nicholson R. (1988).The meaningof personality test scores. American Psychologist,43, 621-626. Hunter, J.E. & Hunter, R.F. (1984).Validity and utility of alter- native predictors of job performance. Psychological Bulletin, 96, 72-98. Ivancevich, J.M. & Matteson, M.T. (1993). Organizational beha- viorand management. Homewood,IL: Irwin. Kaiser, H.F. (1961). A note on Guttman’s lower bound for the number of common factors. BritishJournal of Statistical Psy- chology,14(1),1. Landy, F.J., Shankster, L.J. & Kohler, S.S. (1994). Personnel se- lection and placement. Annual Reviewof Psychology,45, 261- 96. Larson, G.E. & Wolfe, J.H. (1995).Validity results for g from an expanded test base. Intelligence, 20,15-25. McHenry, J.J., Hough, L.M., Toquam, J.L., Hanson, M.A. & Ashworth, S. (1990). Project Avalidity results: the relation- ship between predictor and criterion domains. Personnel Psychology,43, 335-355. Mount, M.K. & Barrick, M.R. (1998). Five reasons why the ‘‘big ¢ve’’article has been frequently cited. Personnel Psycho- logy, 51(4), 849. Muchinsky, P. M. (1993).Validation of intelligence and mecha- nical aptitude tests in selecting employees for manufacturing jobs. Journal of Business and Psychology, 7(4), 373-382. Murphy, K.R. (1996) Individual di¡erences and behaviour in organisations: much more than g. In K.R. Murphy (Ed.), Individual di⁄rences and behaviour in organisations (pp3 ^ 30). San Francisco: Jossey-Bass. Nathan, B.R. & Alexander, R.A. (1988). A comparison of cri- teria for test validation: a meta-analytic investigation. Per- sonnel Psychology,41,517-535. Nell,T.L. (1994). Diesamestelling van’n persoonlikheidspro¢el virdie keuring van bewakingsdienspersoneel (Compiling a personality LA GRANGE, ROODT42 pro¢le for the selection of prison warder personnel). Uni- versiteit van Port Elizabeth. Olea, M.M.& Ree, M.J. (1994). Predicting pilot and navigator criteria: not much more than g. Journal ofApplied Psychology, 79(6), 845-851. Ones, D.S., Mount, M.K., Barrick, M.R. & Hunter, J.E. (1994). Personality and job performance: a critique of the Tett, Jackson, and Rothstein (1991) meta-analysis. Personnel Psychology,47(1),147-156. Parasuraman, A., Berry, L.L. & Zeithaml,V.A. (1991). Re¢ne- ment and reassessment of the servqual scale. JournalofRetai- ling, 67(4), 420-450. Piedmont, R.L. & Weinstein, H.P. (1994). Predicting supervisor ratings of job performance using the NEO personality in- ventory.TheJournal of Psychology,128(3), 255-266. Ree, M.J. & Earles, J.A. (1992). Intelligence is the best predictor of job performance. Current Directions in Psychological Scien- ce,1, 86-89. Ree, M.J. & Earles, J.A. (1994).The ubiquitous productiveness of g. In M.G. Rumsey, C.B.Walker, & J.H. Harris (Eds), Personnel selection and classi¢cation (pp. 127 ^ 136). Hillsdale, NJ: Erlbaum. Ree, M.J., Earles, J.A. & Teachout, M.S.(1994). Predicting job performance: not much more than g. Journal of Applied Psy- chology, 79(4), 518-524. Robertson, I.T. & Kinder, A. (1993). Personality and job com- petencies: the criterion-related validity of some personali- ty variables. Journal of Occupational and Organizational Psychology, 66, 225-244. Sackett, P.R., Gruys, M.L.& Ellingson, J.E. (1998). Ability- personality interactions when predicting job performance. Journal of Applied Psychology,83(4), 545-556. Salgado, J.F. (1996). Personality and job competences: a com- ment on the Robertson & Kinder (1993) study. Journal of Occupational and Organisational Psychology, 69, 373-375. Salgado, J.F. (1997). The ¢ve factor model of personality and job performance in the European community. Journal of Applied Psychology,82(1), 30-45. Saville & Holdsworth Ltd. (1997). Manual & User’s guide for cus- tomer contact series.Thames Ditton. Schepers, J.M. (1999).The bell curve revisited: a South African perspective. Journal of Industrial Psychology, 25(2), 52-61. Schmidt, F.L., Hunter, J.E. & Outerbridge, A.N. (1986). Impact of job experience and ability on job knowledge, work sample, performance, and supervisory ratings of job per- formance. Journal of Applied Psychology, 71, 432-439. Schmitt, N., Gooding, R., Noe, R. & Kirsch, M. (1984). Meta analysis of validity studies published between 1964 and 1982 and the investigation of study characteristics. Person- ality Psychology, 37, 407-422. Seligman, M.E.P. & Schulman, P. (1986). Explanatory style as a predictor of productivity and quitting among life in- surance sales agents. Journal of Personality and Social Psycholo- gy, 50, 832-838. Seligman, M.E.P., Abramson, L.Y., Semmel, A. & von Baeyer, C. (1979). Depressive attributional style. Journal of Abnormal Psychology,88, 242-247. Tett, R., Jackson, D. & Rothstein, M. (1991). Personality mea- sures as predictors of job performance: a meta analytic re- view. Personnel Psychology,44, 703-734. Tett, R.P., Jackson, D.N., Rothstein, M. & Reddon, J.R. (1994). Meta-analysis of personality-job performance rela- tions: a reply to Ones, Mount, Barrick, and Hunter (1994). Personnel Psychology,47(1),157-172. Thurstone, L.L. (1938). Primary mental abilities. Chicago: Uni- versity of Chicago Press. Verbeke,W. (1994). Personality characteristics that predict ef- fective performance of sales people. ScandinavianJournal of Management,10(1), 49-57. Vinchur, A.J., Schippmann, J.S., Switzer, III, F.S. & Roth, P.L. (1998). A meta-analytic review of predictors of job perfor- mance for salespeople. Journal of Applied Psychology, 68, 587- 593. Wagner, R.K. (1997). Intelligence, training, and employment. American Psychologist, 52(10),1059-1069. Wigdor, A.K. & Garner,W.R. (Eds). (1982). Ability testing: uses, consequences, and controversies (part 1).Washington, DC: Na- tional Academy Press. 43PREDICTORS OF THE JOB PERFORMANCE