An investigation of personality similarity effects (relational and perceived) on peer and supervisor ratings and the role of familiarity and liking.
Most research in the performance rating setting which investigates the effect of similarity relies on measures of demographic similarity (e.g. Judge & Ferris, 1993; Tsui & O'Reilly, 1989; Wayne & Liden, 1995) and/or global perceptions of similarity (e.g. Ensher & Murphy, 1997; Turban & Jones, 1988). As early as 1980, Landy and Farr called for researchers to measure actual similarity effects using variables other than demography (e.g. personality). More recently, Bauer and Green (1996) called for investigation of the personality similarity effect on work outcomes using the Five-Factor Model of personality. The current study extends past research by investigating actual personality similarity (i.e. relational personality) and perceived personality similarity effects on performance ratings. Further, given the increased interest and use of multi-level feedback methods (e.g. 360-degree feedback) in organizations (Fletcher, 1995; Waldman, Atwater, & Antonioni, 1998), this study will assess how personality similarity relates to performance ratings by both peers and supervisors. To better understand how similarity affects evaluations, this study explores the role of interpersonal familiarity (knowledge of the ratee) on the relationship between actual and perceived personality similarity. Also, the study tests the hypothesis that interpersonal attraction (liking) mediates the perceived personality similarity--performance rating relationship (Byrne, 1971; Lefkowitz, 2000). The primary purpose of this paper, then, is to assess the effect of similarity in personality variables (measured using actual and perceived personality similarity) on both peer and supervisor performance appraisals and to investigate how familiarity and liking impact the similarity--performance rating relationship.
The similarity effect
The `similar-to-me' hypothesis (see Byrne, 1971) argues that people will be rated higher the more similar they are to the rater or the more similar the rater believes people are to him/herself. Two theories offer support for the `similar-to-me' hypothesis--`self-categorization' theory (Turner, 1987) and Byrne's (1971) similarity-attraction hypothesis. Self-categorization theory (Jackson et al., 1991; Tsui, Egan, & O'Reilly, 1992; Turner, 1987) argues that our self-concept is based on the social categories we place ourselves in (e.g. age, gender, race) and that we desire to have a positive self-identity. The need for a positive self-identity causes us to have a preference for and, by suggestion, evaluate more positively, those similar to us on the social category on which we base our identity. This theory suggests that the categorization of others may require no interaction (i.e. judgments could be instantaneous).
Byrne (1971) and his colleagues borrow from learning theory--using a reinforcement framework to explain why similarity affects people's evaluations of others. Byrne (1971) presents a model in which evaluative responses are a function of reinforcing stimuli (e.g. proportion of similar attitudes) associated with conditioned stimuli. Similar attitudes, for example, are perceived as being rewarding and are therefore viewed in the model as positive reinforcements, whereas dissimilar attitudes function as negative reinforcements. According to the model, an affective response (e.g. interpersonal attraction) mediates the relationship between the conditioned stimulus (e.g. similarity) and evaluative response (e.g. performance rating). Thus, similarity is the reinforcing stimuli leading to an evaluative response.
As noted earlier, researchers have called for investigations of the similarity effect using personality as the measure of similarity (Bauer & Green, 1996; Landy & Farr, 1980). Bauer and Green (1996, p. 1563) suggest that `... other [personality similarity] measures need to be explored. Studying the "Big Five" measures of personality would be a good place to start'. Thus, a unique contribution of this study is assessing the relationship between personality similarity and performance ratings using dimensions from the five-factor model, a comprehensive, well-researched model of personality. The five-factor model includes extraversion, agreeableness, conscientiousness, emotional stability, and openness to experience.
In general, there has been a resurgence of interest in personality in the workplace. In the United Kingdom, the use of personality measures for management selection increased from 12% to 37% from 1984 to 1989 (Robertson & Kinder, 1993). United States practitioners have also increased the use of personality testing in selection (Behling, 1998), partly due to the increased use of teamwork. Furthermore, research supports the enduring nature of personality traits (Funder, 1991). The personality measure used in this research is the five-factor model, a taxonomy of personality traits found in longitudinal studies, across sources (e.g. self-ratings, ratings by spouses), and in different age, sex, race and language groups (e.g. Digman, 1990; Hogan, 1991). Studies using the five-factor model in a variety of countries have found evidence that personality is a valid predictor of job success (see Barrick & Mount, 1991; Dalton & Wilson, 2000; Hough, Eaton, Dunnette, Kamp, & McCloy, 1990; Salgado, 1997).
Prior to hypothesizing about the relationship between personality similarity and performance ratings, the main effects for personality of the ratee will be considered. For the job assessed in this study (salesperson), extraversion, conscientiousness, and emotional stability (see Barrick & Mount, 1991; Barrick, Mount, & Judge, in press; Hough et al., 1990; Salgado, 1997) have been found to be the most highly correlated with measures of sales performance (i.e. most relevant). Therefore, the following hypothesis will be tested.
Hypothesis 1: Conscientiousness, extraversion, and emotional stability will be positively related to performance ratings.
Relational personality effects
The influence similarity in specific personality traits has on performance appraisal could be determined by the relevance of the trait for successful job performance. As stated above, research has found conscientiousness, extraversion, and emotional stability to be relevant for success in sales positions. Interestingly, prior studies investigating relational personality effects on organizational outcomes using personality measures which tapped components of the five-factor model related to these relevant dimensions found significant results (e.g. Ashkanasy & O'Connor, 1997; Bauer & Green, 1996). For example, Bauer and Green (1996) found that actual similarity in positive affectivity (a trait similar to extraversion) related significantly to performance ratings. Similarly, Ashkanasy and O'Connor (1997) found that actual similarity in one of five similarity dimensions (achievement values, which is similar to conscientiousness) was significantly related to higher quality of leader member exchange.
As noted above, self-categorization theory (e.g. Turner, 1987) suggests that judgments may be instantaneous for social categories on which we base our identity. Given the importance of the three relevant personality dimensions to workplace performance for sales people, by extension, individuals may be more likely to base workplace identity on these dimensions. Thus, based on self-categorization theory and the above cited empirical support, the following is hypothesized:
Hypothesis 2: Similarity in personality between the ratee and rater on specific personality dimensions (extraversion, conscientiousness, and emotional stability) will be positively related to supervisor and peer performance ratings.
Perceived similarity effects
Perceived similarity, as defined by Turban and Jones (1988), is the rater's perception of how similar the subject is to the rater. Perceived similarity effects on work-related outcomes have consistently been found to be significant across laboratory (e.g. Dalessio & Imada, 1984; Turban, Jones, & Rozelle, 1990) and field (e.g. Ensher & Murphy, 1997; Turban & Jones, 1988; Wayne & Liden, 1995) studies. As noted in Ferris and Judge (1991) `... the type of similarity index [e.g., actual versus perceived] employed produces different results'. They suggest that one reason that perceptions of similarity more consistently predict work-related outcomes than actual similarity is that `people react on the bases of perceptions of reality, not reality per se' (p. 464). Information-processing models argue that cognitive processing is required for judgments of similarity to evolve (e.g. Srull & Wyer, 1989). Additionally, Byrne's (1971) `similar-to-me' hypothesis suggests that rates must first perceive the relevant other as similar to them.
Studies investigating perceived similarity effects typically use general or global perceptions of similarity (e.g. Ensher & Murphy, 1997; Turban & Jones, 1988). The current study includes a measure of raters' perceptions of the similarity of their personalities to the ratees' personalities (perceived personality similarity). Based on theory and empirical data, it is expected that perceived personality similarity will have a stronger relationship with performance ratings than will relational personality. Additionally, as with relational personality, perceived personality similarity for job relevant personality dimensions should relate significantly to performance ratings. Thus, the following hypothesis will be tested.
Hypothesis 3: Rater's perceived similarity to the ratee in personality on specific personality dimensions (extraversion, conscientiousness, and emotional stability) will be positively related to supervisor and peer performance ratings.
The role of familiarity
One purpose of this study is to investigate how relational personality is related to perceived personality similarity. Several studies have included both actual and perceived similarity effects. For example, Wayne and Liden (1995) include a measure of relational demography for an `actual similarity' variable. Their measure for perceived similarity is a global perception of similarity. Thus, their `actual similarity' variable differs from the construct underlying their `perceived similarity' variable. However, they found that actual demographic similarity and perceived similarity were significantly related. Alternatively, Turban and Jones (1988), using a perceived similarity measure similar to that used by Wayne and Liden (1995), found a nonsignificant relationship between supervisors' perceived similarity and actual demographic similarity. However, Orpen (1984) found a significant relationship between actual attitude similarity and perceived attitude similarity. Intuitively, one would expect that if two people are actually similar they would perceive themselves to be similar in like characteristics (e.g. actual similarity in personality should lead to perceived personality similarity).
The literature on self-peer agreement in personality judgment clearly underscores the importance of interpersonal familiarity. Research has shown that agreement among judges of personality improves with acquaintanceship (e.g. Funder & Colvin, 1988; Hayes & Dunning, 1997). This could be generalized to perceived personality similarity--as the supervisor or peer knows the subject better over time, he/she would more closely approximate the subject's self-rated personality and thus know whether or not he/she is similar to the subject in personality.
Duck (1977) presents a theory that supports the moderating effect of familiarity on interpersonal relationships. He suggests that inconsistent results found in the past for the similarity effect might be due to a lack of consideration of the depth of intimacy of the relationship. Duck's `Filtering Model' hypothesizes that over time, different subtypes of similarity will predict friendship choice or interpersonal attraction. Relatively more `superficial' levels of similarity should be perceived more readily, and therefore, should influence attraction (and subsequent judgmental evaluations) at earlier stages in the acquaintance process. `Deeper' levels of similarity (e.g. personality) should require longer periods to infer, and therefore, should be associated with attraction only after considerable information about the partner has become available (see Harrison, Price, & Bell, 1998; Neimeyer & Mitchell, 1988, for support of the model).
To date, there are no known studies investigating the role of familiarity (how well you know a person) on the actual-perceived similarity relationship when both measures of similarity are the same construct (e.g. personality). As mentioned previously, however, there are a number of studies that demonstrate that greater interpersonal familiarity results in more self-peer agreement in personality judgment. Thus, according to Duck's filtering model, it is expected that given personality is a `deeper level' of similarity, it should take longer to know another person's personality, and by extension, more closely determine how similar or dissimilar one is from a relevant other. The following hypothesis will be tested.
Hypothesis 4: Familiarity will moderate the relationship between relational personality and perceived personality similarity such that those raters who have known the ratee longer will more closely approximate the degree of actual similarity (or dissimilarity) in personality between the rater and ratee.
The role of liking
The focus in performance appraisal research has shifted to include the role of affect in information processing (e.g. Ferris et al., 1994; Lefkowitz, 2000; Tsui & O'Reilly, 1989; Wayne & Liden, 1995). Research in cognitive processing suggests that an affective response is part of the information processing raters engage in when making evaluations (Srull & Wyer, 1989; Zajonc, 1980). Researchers have clearly demonstrated the effect of similarity on liking (Byrne's similarity-attraction paradigm) (e.g. Byrne, 1971; Judge & Ferris, 1993; Tsui & O'Reilly, 1989; Wayne & Liden, 1995). Additionally, as noted in Lefkowitz's (2000) recent review of the role of affect in supervisory performance ratings, a rater's affective regard for a ratee is frequently associated with higher ratings. Further, Byrne (1971) and Lefkowitz (2000) suggest that an affective response mediates the relationship between similarity and evaluative responses. Given that personality takes longer to infer (see above) and that research on Byrne's hypothesis is primarily on perceptions of similarity vs. actual similarity, the cognitive measure (perceived personality similarity) will be included in this hypothesis.
Hypothesis 5: Liking will mediate the relationship between the perceived personality similarity and performance ratings.
The population in the current study was selected from a national appliance organization with headquarters in the midwest of the United States. A sample of 157 salespeople (subjects) filled out a self-report personality measure and 100 of their supervisors and peers provided complete information across all pertinent variables. The subjects were primarily male (85%), averaging 34 years of age, 3 years with the organization, and 2 1/2 years in the job.
The data were collected in conjunction with a training programme. Participants were told that the data collection was for developmental feedback and research purposes only (e.g. `In order to encourage people to give frank and specific feedback, individual responses will be kept confidential'). Thus, participants were clear that their responses would not be used for administrative purposes. Subjects were sent a packet of measures including a brief demographic form and measures of the Big Five personality dimensions. Subjects' supervisors and peers were asked to complete a rating assessing perceived similarity to their subordinates on specific personality traits, a performance rating, a familiarity rating, a liking rating, and a self-report personality measure.
Personality. Personality was assessed with a shortened version of the personality inventory developed by Goldberg (1992). Goldberg's measure includes a 20-item adjective checklist for each of the dimensions in the five-factor model. Due to the practitioners' concerns about the length of the inventory, Goldberg's original list was reduced by extracting 10 of 20 items for each of the personality dimensions based on the magnitude of the factor loadings reported by Goldberg. To provide evidence of construct validity, these scales were correlated with responses from undergraduates (N ranging from 175 to 198) on three other measures of the five-factor model: the NEO Personality Inventory (NEO-PI; Costa & McCrae, 1995), the Hogan Personality Inventory (HPI; Hogan & Hogan, 1992), and the Personal Characteristics Inventory (PCI; Barrick, Mount, Lafitte, & Callans, 1999). The mean correlation among similar personality constructs on the shortened Goldberg scales and the other three inventories were .62, .61 and .65, for extraversion, conscientousness and emotional stability, respectively, and correlations with dissimilar constructs were considerably lower, with an average correlation of .16.
The accumulated data were also factor analysed by the principal components method followed by varimax rotation. The average factor loading across the 10 items for each FFM construct was .55, providing convergent validity evidence. Furthermore, the average loading on other `nonassigned' FFM constructs was .14, demonstrating divergent validity evidence. As this illustrates, the 50 items used in this study `cleanly' loaded on only one FFM construct.
Examples of traits used were: extraversion (e.g. assertive, inhibited); conscientiousness (e.g. organized, efficient); and emotional stability (e.g. envious, touchy). For the three FFM factors, coefficient [alpha] values were .72, .84 and .74, respectively, for extraversion, conscientiousness and emotional stability. Subjects and supervisors were asked to rate how representative each adjective or phrase was of their behaviour on a 5-point Likert-type scale from 1 = `strongly agree' to 5 = `strongly disagree'. The scores were coded so that higher scores indicate a higher level of conscientiousness, etc.
Perceived personality similarity. Supervisors and peers rated how similar they perceived themselves to be to the subject on each of the personality traits included in the self-report personality measure. In making these judgments, raters were instructed to rate themselves on a scale from `Less', to `Same', to `More'. Each time `Less' or `More' was marked, an absolute difference of one was added to the perceived difference score. The smaller the number, the more similar the peer or supervisor perceived his/her personality to be to the subject. The [alpha] coefficient was .90.
Interpersonal familiarity. Raters' familiarity with the ratee was measured with a 4-item scale. The scale includes items related to familiarity at work and outside of work, such as: `How well do you believe you know this person at work', `How well do you believe you know this person outside of work'. Coefficient [alpha] for this scale was .81. Higher scores represent greater interpersonal familiarity with the ratee.
Interpersonal attraction (liking. Byrne's 2-item measure of interpersonal attraction is the standard measure used in much of the literature on interpersonal attraction (e.g. Graves & Powell, 1988; Griffeth, Vecchio, & Logan, 1989). In the current study, coefficient [alpha] was .67. A 5-point Likert-type scale was used to evaluate how well the supervisor (or peer) liked the person and the desirability of working with the person. Scores on the liking measure were obtained by averaging ratings on the two items. A score of 1 represents a low score on liking; 5 represents a high score.
Performance evaluation. A nine-dimensional measure of job performance assessed the subjects' performance. Raters were aware that it had been developed by the researchers (based on job analysis information provided by the firm) and would be used for research purposes. The dimensions included ratings on job knowledge, quality of work, quantity of work, initiative, customer communications, commitment to job and job attitude, account management, interpersonal skills, and overall job performance. The ratings were completed on a 5-point Likert-type scale. Coefficient [alpha] was .89. Again, a low score on the mean item response represents poor performance; a high score represents good performance.
Means, standard deviations, and intercorrelations for each of the variables are presented in Table 1. Statistics for personality variables (self-ratings from ratee and rater in both supervisor and peer data sets) are presented first, followed by perceived personality similarity scores, interpersonal familiarity, liking (interpersonal attraction), and performance ratings. Lower scores represent greater similarity for the perceived similarity scores.
Tests of hypotheses
Hypothesis 1 predicted the three relevant personality constructs (self-ratings of ratee on extraversion, conscientiousness and emotional stability), would be correlated with sales performance. As shown in Table 1, only conscientiousness was consistently correlated with ratings of sales performance (r = .15 and .17, respectively for supervisor and peer ratings of performance). However, only the correlation reported for the peer ratings was significantly different from zero. The correlations between ratings of performance and extraversion (r = .09 for both supervisor and peer ratings of performance) and emotional stability (r = .01 and .03 for supervisor and peer ratings of performance, respectively) were in the expected direction, but none of the correlations were significantly different from zero. Taken together, these results provide partial support for the predictive validity of conscientiousness, but fail to support the validity of other personality traits, as expected in hypothesis 1.
Hypothesis 2 predicted that relational personality (similarity between actual ratee-rater personality) would significantly relate to performance ratings provided by either supervisors or peers. Difference scores are often used to operationalize self-other agreement. However, Edwards (1994) has criticized the use of D-scores for the following reasons: difference scores are less reliable than their components, difference scores are more difficult to interpret than their components, and they often explain less variance than their components. For these reasons, we have chosen to report results using polynomial regressions to directly test similarity effects of self-other personality traits (i.e. relational personality). The following equation  is used to model the critical response surfaces:
(1) Z = [b.sub.0] + [b.sub.1] (ratee personality) + [b.sub.2] (rater personality) + [b.sub.3] [(ratee personality).sup.2] + [b.sub.4] (ratee personality x rater personality) + [b.sub.5] [(rater personality).sup.2]
As shown in Table 2, contrary to hypothesis 2, relational personality for extraversion and conscientiousness was not related to performance ratings by either the supervisor ([R.sup.2] = .064 and .055, respectively) or the peer ([R.sup.2]= .065 and .094, respectively). However, hypothesis 2 was supported in one of the analyses for emotional stability ([R.sup.2] = .125 for peer ratings of performance), while the other was not significant ([R.sup.2] = .035 for supervisor ratings of performance). Consequently, the effect is inconsistent (e.g. for emotional stability) and the magnitude of these effects across traits is relatively small.
Hypothesis 3 predicted that perceived personality similarity on extraversion, conscientiousness and emotional stability would have a significant, positive relationship with supervisor and peer ratings of performance. As shown in Table 3, perceived personality similarity was significantly related to performance ratings for extraversion ([R.sup2] = .083 and .095 for supervisors and peers, respectively), for conscientiousness ([R.sup2] = .117 and .093 for supervisors and peers, respectively, and emotional stability ([R.sup2] = .044 and .077 for supervisors and peers, respectively). Taken together, these results show that as hypothesized, perceived similarity on three job-relevant personality characteristics, extraversion, conscientiousness and emotional stability, were consistently related to higher performance ratings for both supervisors and peers.
To assess whether the degree of interpersonal familiarity with the ratee moderated the relationships between the relational personality variables and the perceived similarity ratings (hypothesis 4), the formula  reported above is modified by adding a parameter for interpersonal familiarity ([b.sub.6](interpersonal familiarity)) and multiplying the first five parameters in formula  by the familiarity variable to test for the moderating effect. This allows one to account for the main effects of the variables first, then to assess the interactive effects. If the test for the last five parameters (accounting for the interactions) is significant, a moderating effect for interpersonal familiarity is suggested (Edwards, 1994). As shown in Table 4, the increase in [R.sup2] by adding interpersonal familiarity (i.e. the main effect) is only significant in one of the analyses, the analysis for extraversion with peer ratings of performance ([DELTA] [R.sup2] = .067). More importantly for testing hypothesis 4, the increase in [R.sup2] due to the moderating effect of interpersonal familiarity is not significant for either supervisor ratings or peer ratings, whether assessed using extraversion, conscientiousness or emotional stability (i.e. increase in [R.sup2] ranges from .001 to .066). Taken together, these results illustrate that interpersonal familiarity did not have a meaningful moderating effect for any relational personality variables on perceived similarity for either supervisors or peers.
Finally, hypothesis 5 posited that interpersonal attraction (liking) would mediate the relationship between the perceived personality similarity and performance ratings. Baron and Kenny's (1986) method for establishing mediation was used to test hypothesis 5. Briefly, the independent variables (perceived personality similarity) must significantly affect the mediator (liking) and the dependent variable (performance ratings), the mediator must significantly affect the dependent variable when the dependent variable is regressed on both the independent and mediator variables, and the effect of the independent on the dependent variable should be less when the dependent variable is regressed on both the independent and mediator variables.
The findings that all three perceptual scales affect performance (see Table 3) establishes one of the conditions required for testing mediation (Baron & Kenny, 1986). Table 5 addresses the second requirement for mediation, showing how perceived similarity in emotional stability ([R.sup.2] = .047 and .029 for supervisors and peers, respectively), conscientiousness ([R.sup.2] = .048 for supervisors), and extraversion ([R.sup.2] = .076 for peers) relate to liking. However, perceived similarity in extraversion ([R.sup.2] = .023 for supervisors) and in conscientiousness ([R.sup.2] = .017 for peers) were not related to liking. Table 5 completes the analysis for mediation, showing that liking was significantly related to both supervisor and peer performance ratings ([R.sup.2] = .204 and .243, respectively) as required for mediation. Finally, additional evidence that liking mediates the relationship between perceived personality similarity for extraversion, conscientiousness and emotional stability, and the performance ratings is provided in Table 5 by noting that the regression weights were consistently reduced for the perceived similarity variables, after controlling for liking. However, the magnitudes of these decrements were not large, providing mixed support for liking as a mediator. Thus, the relationship between emotional stability and performance was mediated by liking with both supervisor and peer ratings. However, liking was a mediator for both extraversion and conscientiousness in only one of two possible data sets (peer or supervisor, respectively). Taken together, these findings provide moderate support for hypothesis 5, that liking is a mediator of the relationship between perceived similarity and performance ratings.
Interestingly, the results of this study support the suggestion that `people react on the bases of perceptions of reality, not reality per se' (Ferris & Judge, 1991, p. 464). Actual personality similarity (relational personality) did not relate significantly to performance ratings in either the peer or supervisor data set (hypothesis 2) with one exception--similarity in emotional stability was related to peer ratings. However, perceived personality similarity did relate significantly to performance ratings (hypothesis 3). For all three personality dimensions hypothesized to be relevant for sale's positions (i.e. conscientiousness, emotional stability and extraversion), greater perceived similarity was associated with higher performance ratings in both peer and supervisor data sets. Interestingly, in spite of meta-analytic support for the relevance of these three dimensions for sales positions (e.g. Barrick & Mount, 1991; Barrick et al., in press; Hough et al., 1990; Salgado, 1997), only ratee conscientiousness reached levels of significance with peer performance ratings (hypothesis 1).
The finding that perceived personality similarity effects are greater than actual personality similarity effects is supported in a similar body of research. Cable and Judge (1996) found that demographic similarity did not affect person-organization (P-O) fit perceptions, whereas subjective P-O fit perceptions significantly predicted job choice intentions. Similarly, Judge and Cable (1997) found that although both subjective and objective fit were related to organizational attraction, subjective fit mediated the relationship between objective fit and attraction. Given these results, along with theory (Byrne, 1971) and past research on perceived similarity effects (e.g. Turban 8: Jones, 1988), it is not surprising that our study found subjective (perceived) similarity in personality related to performance ratings to a much greater extent than relational personality.
A second purpose of this study was to investigate how personality similarity relates to performance ratings. Given the well-documented work by Funder and his colleagues (e.g. Funder & Colvin, 1988) on the relationship between acquaintanceship and consensus in ratings and Duck's (1977) filtering model (i.e. deeper levels of similarity take longer to infer), we hypothesized that as raters knew their ratees better, they would more `accurately' perceive themselves as similar or dissimilar (hypothesis 4). The `acquaintanceship hypothesis' was not supported in either data set. In other words, the relationship between actual and perceived similarity was not higher for those who knew each other better. Similarly, a recent study by Watson, Hubbard, and Wiese (2000) found no support for the moderating effect of acquaintanceship on self-other agreement in personality for any of three facets of acquaintanceship (e.g. acquaintanceship length, relationship closeness and number of shared activities). Their measure of relationship closeness is similar to the measure of familiarity used in this study. These results seem counter-intuitive. However, given that prior studies have found an acquaintanceship effect for self-other agreement in personality (e.g. Funder & Colvin, 1988), future studies should consider the relationship between actual and perceived similarity with familiarity (or acquaintanceship) as a moderator.
Although we found no support for a moderating effect of interpersonal familiarity on the relational personality-perceived personality similarity relationship, we did find evidence that actual and perceived similarity were related in this study (see Table 4). Actual and perceived similarity in extraversion are significantly related in both the peer and supervisor data sets. This supports past research which has found extraversion to be a `visible trait' and thus easier for others to assess (see Funder & Colvin, 1988). Additionally, there was a significant relationship between actual and perceived similarity in emotional stability in the supervisor data set. Emotional stability (i.e. neuroticism) is also rated as a `visible trait' (see Funder & Colvin, 1988). One explanation for finding a lack of significant relationships between the other measures of actual and perceived personality similarity is Hogan's (1991) argument that personality has two different meanings: (1) it refers to a person's public self or social reputation from the observers' perspective and (2) it refers to the `structures, dynamics, processes, and propensities inside a person that explain why he or she behaves in a characteristic way [the inner nature]' (p. 875). Thus, raters may be judging similarity on `public behaviours', whereas ratees are rating their `inner natures'. The two perspectives on personality judgment may be very different and, as such, may not correlate no matter how well the rater knows the ratee. Given a significant relationship between actual and perceived similarity was found in half of the relationships in the current study, future research should consider what, if not actual similarity, we base our perceptions of similarity on. In either case, these perceptions may be irrelevant and/or incorrect.
Results in the current study support past research findings that perceived personality similarity is related to liking (e.g. Byrne, 1971)--with the exception of perceived similarity in extraversion in the supervisor data set and perceived similarity in conscientiousness in the peer data set. Additionally, results support prior research evidence that liking is related to performance ratings (e.g. Judge & Ferris, 1993; Lefkowitz, 2000; Tsui & O'Reilly, 1989; Wayne & Liden, 1995). Finally, results demonstrate that liking mediated the relationship between perceived personality similarity and performance ratings (hypothesis 5) for emotional stability in both data sets, for extraversion in the peer data set, and for conscientiousness in the supervisor data set. Thus, perceived personality similarity had both direct and indirect effects (through liking) on performance ratings. These findings provide moderate support for Byrne's (1971) and Lefkowitz' (2000) hypotheses. However, future studies should attempt to determine how liking, similarity and performance ratings are related (e.g. it is possible that liking impacts perceptions of similarity).
Returning to hypothesis 2 (relational personality will be significantly related to performance ratings), although there was little support for a relationship between relational personality and supervisor or peer ratings of performance, we examined whether the level of similarity affected the nature of this relationship. Although performance ratings were higher as self-and-other ratings became more similar and higher, none of these analyses were significantly different. Further research should continue to investigate the moderating effect of level of similarity on the actual similarity--performance rating relationship.
It is also possible that some range restriction on the actual similarity scores may have limited the possibility of finding other significant results for hypothesis 2. The sample used in this study was based on one job type--sales. This may lead to a restriction of range in the kinds of people in an organization/department/job type (e.g. Holland, 1985; Schneider, Goldstein, & Smith, 1995). Similarly it may be that subjects and raters evaluate their own personalities in a favourable light (see Hogan, 1991; Paulhus, 1986), reducing the variability in ratings of personality which would lead to less variability on the actual personality similarity scores. Finally, before we assume there is a limited `actual similarity effect', researchers should consider other methods of assessing relational personality. For example, Mount, Barrick, and Strauss (1994) have studied personality assessment by other sources (e.g. customer, supervisor, peer), finding that other sources' ratings of personality dimensions were valid predictors of performance ratings and accounted for significant variance in the criterion measure beyond self-ratings alone. Thus, rather than using self-ratings of personality dimensions to determine actual similarity, multiple sources could assess the ratee and raters' personalities and a similarity score could be computed from these `360-degree' assessments.
Although we found significant results for the relationship between perceived personality similarity and performance ratings, we recognize there are alternative processes which could be operating to predict how perceptions of personality similarity/dissimilarity relate to performance appraisals. Social desirability and/or work-related relevance of the personality dimension could affect ratings such that for those traits found to be more socially desirable and/or relevant, ratees perceived to be high in the dimension would receive higher performance ratings than those low in the dimension regardless of the rater's level of the trait. Funder and Colvin (1988) reported that emotional stability, agreeableness and conscientiousness were rated as significantly higher in desirability than extraversion and openness. Graziano, Jensen-Campbell, and Hair (1996) found that individuals like agreeable partners more than less agreeable partners regardless of their own level of agreeableness. Alternatively, Day and Bedeian (1995) found that subordinates who were similar in agreeableness to their co-workers were rated higher by their supervisors than those dissimilar in agreeableness. However, the unit of analysis in the Day and Bedeian study was an aggregate of the group's similarity/dissimilarity to the individual, whereas the current study considers the dyad as the unit of analysis. In this study, we were not able to assess how absolute differences between the employee's perceived standing on a trait and the rater's actual level on the trait (e.g. high perceived conscientiousness for the ratee and low conscientiousness for the rater) were related to performance ratings. However, we could assess how the rater's perceptions of the direction of difference from the ratee on a personality dimension affected ratings. Thus, for socially desirable traits and/or relevant traits (e.g. conscientiousness), one could assume that if the rater perceived himself/ herself to be relatively less conscientious than the rate, he/she might evaluate the ratee higher than if the rater thought he/she were more conscientious than the ratee. We found this effect for conscientiousness but not for either of the other two personality dimensions--one of which, emotional stability, is also considered to be both socially desirable and relevant (analyses are available upon request from the first author). Consequently, future research should consider both similarity and dissimilarity (direction and level of difference) effects on ratings.
Our results consistently revealed the significant effect perceptions of similarity have on performance ratings. One limitation of these findings, however, is that the research design of this study is cross-sectional and both ratings (perceptions of similarity and performance ratings) are from the same source. Thus, it is possible that common method variance could account for the findings. However, given replication of the findings for perceived similarity effects in earlier research (e.g. Turban & Jones, 1988; Wayne & Liden, 1995), the cumulative evidence appears to be very supportive for the perceived similarity-performance rating relationship.
As noted earlier, another limitation of the current study is determination of causality given the cross-sectional data. Cues (which may or may not be relevant) other than relational personality (e.g. liking, familiarity, ability, actual performance) may be triggering a perception of similarity by the raters (i.e. the `halo' effect). The inability to test causal inferences makes it impossible to rule out alternative explanations for the effects noted in this study. For example, it is possible that a positive performance rating leads a rater to perceive similarity. It is quite difficult to design a study at work that can rule out this causal sequence, however, as the raters (peers and supervisors) will have performance information available at the time they assess perceived similarity. One advantage of the design used in this study is that two rating sources were used (peers and supervisors). While a supervisor's success is related to his/her ratees' performance in this job, a peer's success is not.
Consequently, one would expect the positive performance `bias' on perceived similarity, if it exists, to be stronger for supervisors. However, in this study the magnitude of the relationship between perceived similarity and performance ratings for peer ratings was quite similar to those for supervisor ratings. Nevertheless, the inability to eliminate alternative causal inferences suggests one should be more cautious when interpreting the effect of perceived similarity on performance ratings. Future research should investigate alternative models and longitudinal data collection procedures should be used to determine causality.
Performance appraisal continues to be a topic of great interest to managers, employees, and industrial and organizational psychologists. Although the importance of performance appraisals has been questioned (e.g. Nelson, 2000), the use of performance appraisals for administrative decisions (e.g. Fletcher, 1995; Husek, 1998), motivational and developmental purposes (Fletcher, 1995; Waldman et al., 1998), and legal purposes (protecting against discrimination based on non-job-relevant measures such as race, gender, national origin) is unlikely to end soon. As noted in Kald and Nilsson (2000), the benefits of performance appraisal outweigh the shortcomings. However, if cues, such as personality similarity, are used to judge performance, the potential for bias in the new global environment increases. Thus, organizations may need to emphasize alternative methods (e.g. pay for skills, promotion based on seniority) for promotion, layoff, and compensation decisions. This supports arguments made by Deming and his followers (see Scholtes, 1987) who believe that performance differences are primarily due to the system and objective performance evaluation does not exist. Also, organizations could use multiple raters of performance. Wholers and London (1989) found that data from four sources (supervisors, peers, subordinates and self) were desirable to provide a complete picture of the individual. Using multiple raters may overcome the idiosyncratic biases of any one person's category system. In addition to using multiple raters, the use of `hard' criteria for performance assessment should be encouraged, although it should be recognized that even `hard' criteria (e.g. sales) have uncontrollable influences (e.g. sales region).
Alternatively, a technique that organizations can use to address the accuracy of performance ratings is `Frame of Reference' training. Frame of reference training provides raters with a common reference standard (i.e. `frames') by which to evaluate others' performance (Bernardin & Buckley, 1981). Results from studies examining this type of training have been positive in that training leads to better rating accuracy (Day & Sulsky, 1995; Sulsky & Day, 1992). Training raters to recognize a perceived similarity bias may increase their willingness to evaluate dissimilar others more accurately. This becomes even more critical as human capital flows across national boundaries and organizations are more likely to have a culture mix of employees who are not similar to each other. Without some type of intervention (e.g. frame of reference or diversity training) raters are more likely to base ratings on stereotypes such as similarity, especially when filling in gaps of knowledge on ratee performance.
The results of this study suggest that there is a complicated interaction between interpersonal similarity (perceived or actual) and performance ratings, which requires further exploration along with inclusion of other variables (e.g. familiarity, liking, ability). Additionally, although there was minimal support for the relational personality-performance rating relationship, we believe it is premature to discontinue research on actual similarity effects using relevant measures of the Five Factor Model of personality as the measure of similarity.
Table 1. Descriptive statistics and intercorrelations among variables for supervisors (reported below the diagonal) and peers (reported above the diagonal) Supervisor Peer Variables M SD M SD Ratee self ratings 1. Extraversion 3.88 0.44 3.89 0.46 2. Conscientious 4.32 0.40 4.32 0.38 3. Emotional stability 3.24 0.55 3.18 0.16 Rater self-ratings 4. Extraversion 3.92 0.39 3.84 0.47 5. Conscientious 4.23 0.35 4.31 0.36 6. Emotional stability 3.51 0.59 3.13 0.54 Perceived similarity 7. Extraversion 0.44 0.26 0.41 0.41 8. Conscientious 0.32 0.28 0.24 0.24 9. Emotional stability 0.37 0.24 0.35 0.26 Other variables 10. Familiarity 2.77 0.75 3.03 0.83 11. Liking 8.07 0.83 8.70 1.25 12. Performance 34.65 5.45 36.86 5.52 rating Variables 1 2 3 4 Ratee self ratings 1. Extraversion .16 .10 -.03 2. Conscientious .20 * .26 * -.16 3. Emotional stability .16 .37 * -.02 Rater self-ratings 4. Extraversion -.10 .09 -.18 * 5. Conscientious -.15 -.04 -.02 .25 * 6. Emotional stability -.27 * -.06 -.16 .37 * Perceived similarity 7. Extraversion -.26 * .00 -.02 -.17 * 8. Conscientious .01 .09 -.02 -.22 * 9. Emotional stability .12 .13 -.05 -.15 Other variables 10. Familiarity -.02 .00 .21 * -.05 11. Liking -.06 .04 .09 -.03 12. Performance .09 .15 .01 .16 rating Variables 5 6 7 8 Ratee self ratings 1. Extraversion .07 .00 -.15 -.01 2. Conscientious .09 -.03 .14 -.12 3. Emotional stability -.10 .07 .11 -.07 Rater self-ratings 4. Extraversion .19 * .14 -.11 -.02 5. Conscientious .43 * -.12 -.15 6. Emotional stability .49 * -.09 -.27 * Perceived similarity 7. Extraversion -.06 -.03 .32 * 8. Conscientious -.21 * -.17 * .37 * 9. Emotional stability -.11 -.03 .41 * .44 * Other variables 10. Familiarity -.21 * -.01 -.14 .04 11. Liking -.04 .10 -.15 -.22 * 12. Performance .11 .16 -.29 * -.34 * rating Variables 9 10 11 12 Ratee self ratings 1. Extraversion -.05 -.14 .05 .09 2. Conscientious .04 .10 .10 .17 * 3. Emotional stability -.01 .20 * .03 .03 Rater self-ratings 4. Extraversion -.02 -.02 .16 .20 * 5. Conscientious -.19 * .10 .21 * .25 * 6. Emotional stability -.25 * .12 .17 * .33 * Perceived similarity 7. Extraversion .56 * -.26 * -.28 * -.31 * 8. Conscientious .49 * -.17 * -.13 -.31* 9. Emotional stability -.11 -.17 * -.28 * Other variables 10. Familiarity -.03 .37 * .41 * 11. Liking -.22 * .42 * .49 * 12. Performance -.21 * .31 * .45 * rating * p<.05 (one tailed), 90% CI = .01 [less than or equal to] .17 [less than or equal to] .32. Note. N=100 for both the supervisor and peer analyses. Table 2. Polynomial regression results of relational personality variables on ratings of performance Extraversion Variable B SE B Supervisor ratings of performance Intercept 34.10 81.07 Ratee personality 14.74 20.95 Rater personality -18.88 26.99 Ratee [personality.sup.2] -0.64 1.82 Ratee x Rater personality -2.18 4.01 Rater [personality.sup.2] 3.89 2.81 Model [R.sup.2] .064 Peer ratings of performance Intercept 50.55 50.55 Ratee personality -13.64 18.49 Rater [personality.sup.2] 2.36 15.45 Ratee personality 0.48 1.84 Ratee x Rater personality 2.87 2.46 Rater [personality.sup.2] -1.46 1.88 Model [R.sup.2] .065 Conscientiousness Variable B SE B Supervisor ratings of performance Intercept 74.38 111.92 Ratee personality 15.88 28.08 Rater personality -40.1 38.96 Ratee [personality.sup.2] -1.29 2.74 Ratee x Rater personality -0.62 3.88 Rater [personality.sup.2] 5.38 4.19 Model [R.sup.2] .055 Peer ratings of performance Intercept -28.07 115.93 Ratee personality 30.04 31.40 Rater [personality.sup.2] -5.63 33.16 Ratee personality -2.27 2.87 Ratee x Rater personality -2.06 4.18 Rater [personality.sup.2] 2.16 3.07 Model [R.sup.2] .094 Emotional stability Variable B SE B Supervisor ratings of performance Intercept 49.54 30.70 Ratee personality -5.29 12.94 Rater personality -6.31 10.55 Ratee [personality.sup.2] 0.78 1.49 Ratee x Rater personality 0.20 1.72 Rater [personality.sup.2] 1.09 1.40 Model [R.sup.2] .035 Peer ratings of performance Intercept 60.96 * 27.78 Ratee personality -11.04 9.84 Rater [personality.sup.2] -7.92 12.03 Ratee personality 1.23 1.41 Ratee x Rater personality 1.04 2.11 Rater [personality.sup.2] 1.25 1.68 Model [R.sup.2] .125 * Note. N=100 for both the supervisor and peer analyses. Table 3. Regression results for perceived personality variables on ratings on performance ratings by the supervisor or the peer Extraversion Variable B SE B Supervisor ratings of performance Intercept 37.32 * 1.04 Perceived similarity -6.08 * 2.05 Model [R.sup.2] .083 * Peer ratings of performance Intercept 39.32 * 0.93 Perceived similarity -6.05 * 1.89 Model [R.sup.2] .095 * Conscientiousness Variable B SE B Supervisor ratings of performance Intercept 36.79 * 0.78 Perceived similarity -6.62 * 1.83 Model [R.sup.2] .117 * Peer ratings of performance Intercept 38.82 * 0.81 Perceived similarity -8.11 * 2.56 Model [R.sup.2] .093 * Emotional stability Variable B SE B Supervisor ratings of performance Intercept 36.40 * 0.98 Perceived similarity -4.76 * 2.24 Model [R.sup.2] .044 * Peer ratings of performance Intercept 38.93 * 0.90 Perceived similarity -5.88 * 2.06 Model [R.sup.2] .077 * Note. N=100 for both the supervisor and peer analyses. Table 4. Polynomial regression results testing the moderating effect of interpersonal familiarity on the association between relational personality variables and supervisor ratings of perceived similarity Extraversion Variable B SE B Supervisor ratings of perceived similarity Intercept 2.68 3.89 Ratee personality -1.23 0.96 Rater personality 1.01 1.21 Ratee [personality.sup.2] -0.21 0.23 Ratee x Rater personality 0.73 0.52 Rater [personality.sup.2] -0.59 * 0.28 Model [R.sup.2] .155 * Interpersonal familiarity -0.41 0.30 Model [R.sup.2] .167 * Increase in model [R.sup.2] .012 Familiarity x Ratee personality -28.44 17.93 Familiarity x Rater personality 38.30 20.79 Familiarity x Ratee [personality.sup.2] 0.14 0.09 Familiarity x (Ratee x Rater personality) -0.28 0.18 Familiarity x Rater [personality.sup.2] 0.17 0.09 Model [R.sup.2] .233 * Increase in model [R.sup.2] .066 Peer ratings of perceived similarity Intercept 5.14 2.90 Ratee personality -0.46 0.91 Rater personality -0.90 0.77 Ratee [personality.sup.2] -0.05 0.17 Ratee x Rater personality 0.12 0.30 Rater [personality.sup.2] -1.00 0.16 Model [R.sup.2] 0.109 * Interpersonal familiarity -0.62 * 0.25 Model [R.sup.2] .176 * Increase in model [R.sup.2] .067 * Familiarity x Ratee personality 4.96 14.79 Familiarity x Rater personality 0.48 1.63 Familiarity x Ratee [personality.sup.2] 0.06 0.06 Familiarity x (Ratee x Rater personality) -0.09 0.11 Familiarity x Rater [personality.sup.2] 0.07 0.05 Model [R.sup.2] .220 * Increase in model [R.sup.2] .044 Conscientiousness Variable B SE B Supervisor ratings of perceived similarity Intercept 3.52 7.21 Ratee personality -1.59 1.54 Rater personality 0.30 2.35 Ratee [personality.sup.2] 0.14 0.14 Ratee x Rater personality 0.11 0.21 Rater [personality.sup.2] -0.12 0.23 Model [R.sup.2] .066 Interpersonal familiarity 0.01 0.35 Model [R.sup.2] 0.066 Increase in model [R.sup.2] 0.000 Familiarity x Ratee personality 40.68 * 19.71 Familiarity x Rater personality 35.51 30.14 Familiarity x Ratee [personality.sup.2] 0.00 0.01 Familiarity x (Ratee x Rater personality) -2.98 12.08 Familiarity x Rater [personality.sup.2] 0.00 0.02 Model [R.sup.2] 0.067 Increase in model [R.sup.2] 0.001 Peer ratings of perceived similarity Intercept 0.50 4.55 Ratee personality 0.62 1.26 Rater personality -0.61 1.29 Ratee [personality.sup.2] -0.07 0.13 Ratee x Rater personality -0.05 0.16 Rater [personality.sup.2] 0.11 0.12 Model [R.sup.2] .042 Interpersonal familiarity -0.01 0.25 Model [R.sup.2] .062 Increase in model [R.sup.2] .020 Familiarity x Ratee personality -21.79 28.82 Familiarity x Rater personality 5.62 34.04 Familiarity x Ratee [personality.sup.2] 0.00 0.01 Familiarity x (Ratee x Rater personality) -8.94 12.81 Familiarity x Rater [personality.sup.2] -0.01 0.01 Model [R.sup.2] .068 Increase in model [R.sup.2] .006 Emotional stability Variable B SE B Supervisor ratings of perceived similarity Intercept -1.35 3.23 Ratee personality -0.01 0.57 Rater personality 1.21 1.79 Ratee [personality.sup.2] 0.13 0.15 Ratee x Rater personality -0.27 0.27 Rater [personality.sup.2] -0.01 0.25 Model [R.sup.2] .140 * Interpersonal familiarity 1.18 0.96 Model [R.sup.2] .145 * Increase in model [R.sup.2] .005 Familiarity x Ratee personality 4.32 11.00 Familiarity x Rater personality -0.80 0.60 Familiarity x Ratee [personality.sup.2] -0.03 0.05 Familiarity x (Ratee x Rater personality) 0.07 0.09 Familiarity x Rater [personality.sup.2] 0.09 0.08 Model [R.sup.2] .164 Increase in model [R.sup.2] .019 Peer ratings of perceived similarity Intercept 0.94 3.38 Ratee personality -0.10 2.13 Rater personality -0.20 0.63 Ratee [personality.sup.2] 0.21 0.33 Ratee x Rater personality -0.41 0.33 Rater [personality.sup.2] 0.23 0.19 Model [R.sup.2] .079 Interpersonal familiarity 0.06 1.10 Model [R.sup.2] .083 Increase in model [R.sup.2] .004 Familiarity x Ratee personality -0.03 0.71 Familiarity x Rater personality -4.27 12.03 Familiarity x Ratee [personality.sup.2] -0.05 0.11 Familiarity x (Ratee x Rater personality) 0.12 0.11 Familiarity x Rater [personality.sup.2] -0.07 0.06 Model [R.sup.2] .100 Increase in model [R.sup.2] .017 Note. N=100 for both the supervisor and peer analyses. Table 5. Regression results testing the mediating effect of interpersonal attraction (liking) on the association between supervisor and peer ratings of perceived similarity and ratings of performance Extraversion Variable B SE B Supervisor ratings of performance Regress perceived similarity on liking Intercept 8.28 * 0.16 Perceived personality similarity -0.49 0.32 Model [R.sup.2] .023 Regress perceived similarity [right arrow] liking on performance Intercept 14.68 * 4.91 Perceived personality similarity -4.75 * 1.88 Liking 2.73 * 0.58 Model [R.sup.2] .253 * Peer ratings of performance Regress perceived similarity on liking Intercept 9.20 * 0.21 Perceived personality similarity -1.23 * 0.43 Model [R.sup.2] .076 * Regress perceived similarity [right arrow] liking on performance Intercept 21.39 * 3.75 Perceived personality similarity -3.66 * 1.77 Liking 1.95 * 0.39 Model [R.sup.2] .275 * Conscientiousness Variable B SE B Supervisor ratings of performance Regress perceived similarity on liking Intercept 8.28 * 0.12 Perceived personality similarity -0.65 * 0.29 Model [R.sup.2] .048 * Regress perceived similarity [right arrow] liking on performance Intercept 15.37 * 4.89 Perceived personality similarity -4.94 * 1.72 Liking 2.59 * 0.58 Model [R.sup.2] .266 * Peer ratings of performance Regress perceived similarity on liking Intercept 8.89 * 0.19 Perceived personality similarity -0.79 0.60 Model [R.sup.2] .017 Regress perceived similarity [right arrow] liking on performance Intercept 20.74 * 3.43 Perceived personality similarity -6.50 * 2.27 Liking 2.03 * 0.38 Model [R.sup.2] .302 * Emotional stability Variable B SE B Supervisor ratings of performance Regress perceived similarity on liking Intercept 8.34 * 0.15 Perceived personality similarity -0.75 * 0.34 Model [R.sup.2] .047 * Regress perceived similarity [right arrow] liking on performance Intercept 13.13 * 5.11 Perceived personality similarity -2.67 2.09 Liking 2.79 * 0.60 Model [R.sup.2] .217 * Peer ratings of performance Regress perceived similarity on liking Intercept 8.94 * 0.21 Perceived personality similarity -0.67 0.40 Model [R.sup.2] .029 * Regress perceived similarity [right arrow] liking on performance Intercept 20.66 * 3.51 Perceived personality similarity -4.50 * 1.84 Liking 2.04 * 0.38 Model [R.sup.2] .287 * Liking Variable B SE B Supervisor ratings of performance Regress perceived similarity on liking Intercept 10.80 * 4.79 Perceived personality similarity 2.96 * 0.59 Model [R.sup.2] .204 * Regress perceived similarity [right arrow] liking on performance Intercept Perceived personality similarity Liking Model [R.sup.2] Peer ratings of performance Regress perceived similarity on liking Intercept 17.93 * 3.41 Perceived personality similarity 2.18 * 0.39 Model [R.sup.2] .243 * Regress perceived similarity [right arrow] liking on performance Intercept Perceived personality similarity Liking Model [R.sup.2] Note. N=100 for both the supervisor and peer analyses.
The authors thank Daniel Turban for thoughtful comments on an earlier draft of this manuscript.
Ashkanasy, N. M., & O'Connor, C. (1997). Value congruence in leader-member exchange. Journal of Social Psychology, 137, 647-662.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.
Barrick, M. R., & Mount, M. K. (1991). The big five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1-26.
Barrick, M. R., Mount, M. K., & Judge, T. A. (in press). The FFM personality dimensions and job performance: Meta-analysis of meta-analyses. International Journal of Selection and Assessment.
Barrick, M. R., Mount, M. K., Lafitte, L., & Callans, M. (1999). The Personal Characteristics Inventory. Libertyville, IL: The Wonderlic Corporation.
Bauer, T. N., & Green, S. G. (1996). Development of leader-member exchange: A longitudinal test. Academy of Management Journal, 39, 1538-1567.
Behling, O. (1998). Employee selection: Will intelligence and conscientiousness do the job? Academy of Management Executive, 12, 77-86.
Bernardin, H. J., & Buckley, M. R. (1981). A consideration of strategies in rater training. Academy of Management Review, 6, 205-212.
Byrne, D. (1971). The attraction paradigm. New York: Academic Press.
Cable, D. M., & Judge, T. A. (1996). Person-organization fit, job choice decisions, and organizational entry. Organizational Behavior and Human Decision Processes, 67, 294-311.
Costa, P. T., & McCrae, R. R. (1995). Domains and facets: Hierarchical personality assessment using the revised NEO personality inventory. Journal of Personality Assessment, 64, 21-50.
Dalessio, A., & Imada, A. S. (1984). Relationships between interview selection decisions and perceptions of applicant similarity to an ideal employee and self: A field study. Human Relations, 37, 67-80.
Dalton, M., & Wilson, M. (2000). The relationship of the five-factor model of personality to job performance for a group of Middle Eastern expatriate managers. Journal of Cross-Cultural Psychology, 31, 250-258.
Day, D. V., & Bedeian, A. G. (1995). Personality similarity and work-related outcomes among African-American nursing personnel: A test of the supplementary model of person-environment congruence. Journal of Vocational Behavior, 46, 55-70.
Day, D. V., & Sulsky, L. M. (1995). Effects of frame-of-reference training and information configuration on memory organization and rating accuracy. Journal of Applied Psychology, 80, 158-168.
Digman, J. M. (1990). Personality structure: Emergence of the five-factor model. Annual Review of Psychology, 41, 417-440.
Dipboye, R. L. (1985). Some neglected variables in research on discrimination in appraisals. Academy of Management Review, 10, 116-127.
Duck, S. W. (1977). The study of acquaintance. Farnborough, UK: Saxon House.
Edwards, J. R. (1994). The study of congruence and organizational behavior research: Critique and a proposed alternative. Organizational Behavior and Human Decision Processes, 58, 51-100.
Ensher, E. A., & Murphy, S. E. (1997). Effects of race, gender, perceived similarity and contact on mentor relationships. Journal of Vocational Behavior, 50, 460-481.
Ferris, G. R., & Judge, T. A. (1991). Personnel/human resources management: A political influence perspective. Journal of Management, 17, 447-488.
Ferris, G. R., Judge, T. A., Rowland, K. M., & Fitzgibbons, D. E. (1994). Subordinate influence and the performance evaluation process: Test of a model. Organizational Behavior and Human Decision Processes, 55, 101-135.
Fletcher, C. (1995). New directions for performance appraisal: Some findings and observations. International Journal of Selection and Assessment, 3, 191-196.
Funder, D. C. (1991). Global traits: A neo-allportian approach to personality. Psychological Science, 2, 31-39.
Funder, D. C., & Colvin, C. R. (1988). Friends and strangers: Aquaintanceship, agreement, and the accuracy of personality judgment. Journal of Personality and Social Psychology, 55, 149-158.
Goldberg, L. R. (1992). The development of markers for the big-five factor structure. Psychological Assessment, 4, 26-42.
Graves, L. M., & Powell, G. N. (1988). An investigation of sex discrimination in recruiters' evaluations of actual applicants. Journal of Applied Psychology, 73, 20-29.
Graziano, W. G., Jensen-Campbell, L. A. & Hair, E. C. (1996). Perceiving interpersonal conflict and reacting to it: The case for agreeableness. Journal of Personality and Social Psychology, 70, 820-835.
Griffeth, R. W., Vecchio, R. P., & Logan, J. W. Jr. (1989). Equity theory and interpersonal attraction. Journal of Applied Psychology, 74, 394-401.
Harrison, D. A., Price, K. H., & Bell, M. P. (1998). Beyond relational demography: Time and the effects of surface- and deep-level diversity on work group cohesion. Academy of Management Journal, 41, 96-107.
Hayes, A. F., & Dunning, D. (1997). Construal processes and trait ambiguity: Implications for self-peer agreement in personality judgment. Journal of Personality and Social Psychology, 7, 664-677.
Hogan, R. T. (1991). Personality and personality measurement. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (vol. 2, pp. 271-326). Palo Alto, CA: Consulting Psychology Press.
Hogan, R. T., & Hogan, J. (1992). Hogan Personality Inventory manual. Tulsa, OK: Hogan Assessment Systems.
Holland, J. L. (1985). The SDS Professional Manual--1985 Revision. Odessa, FL: Psychological Assessment Resources.
Hough, L. M., Eaton, N. K., Dunnette, M. D., Kamp, J. D., & McCloy, R. A. (1990). Criterion-related validities of personality constructs and the effect of response distortion on those validities. Journal of Applied Psychology, 75, 581-595.
Husek, G. (1998, May). Sharing the risk. Industry Week, 247(9), 50-53.
Jackson, S. E., Brett, J. F., Sessa, V. I., Cooper, D. M., Julin, J. A., & Peyronnin, K. (1991). Some differences make a difference: Individual dissimilarity and group heterogeneity as correlates of recruitment, promotion, and turnover. Journal of Applied Psychology, 76, 675-689.
Judge, T. A., & Cable, D. M. (1997). Applicant personality, organizational culture, and organization attraction. Personnel Psychology, 50, 359-394.
Judge, T. A., & Ferris, G. R. (1993). Social context of performance evaluation decisions. Academy of Management Journal, 25, 349-358.
Kald, M., & Nilsson, F. (2000). Performance management at Nordic companies. European Management Journal, 18, 113-127.
Landy, F. J., & Farr, J. L. (1980). Performance rating. Psychological Bulletin, 87, 72-107.
Lefkowitz, J. (2000). The role of interpersonal affective regard in supervisory performance ratings: A literature review and proposed causal model. Journal of Occupational and Organizational Psychology, 73, 67-85.
Mount, M. K., Barrick, M. R., & Strauss, J. P. (1994). Validity of observer ratings of the Big Five personality factors. Journal of Applied Psychology, 79, 272-280.
Neimeyer, R. A., & Mitchell, K. A. (1988). Similarity and attraction: A longitudinal study. Journal of Social and Personal Relationships, 5, 131-148.
Nelson, R. (2000). Are performance appraisals obsolete? Compensation and Benefits Review, 32(3), 39-42.
Orpen, C. (1984). Attitude similarity, attraction, and decision-making in the employment interview. The Journal of Psychology, 117, 111-120.
Paulhus, D. L. (1986). Self-deception and impression management in test responses. In A. Angleitner & J. S. Wiggins (Eds.), Personality assessment via questionnaires (pp. 144-165). Berlin: Springer.
Robertson, I. T., & Kinder, A. (1993). Personality and job competencies: The criterion-related validity of some personality variables. Journal of Occupational and Organization Psychology, 66, 225-244.
Salgado, J. F. (1997). The five factor model of personality and job performance in the European Community. Journal of Applied Psychology, 82, 30-43.
Schneider, B., Goldstein, H. W., & Smith, D. B. (1995). The ASA framework: An update. Personnel Psychology, 48, 747-774.
Scholtes, P. R. (1987). An elaboration on Deming's teachings on performance appraisal. Madison, WI: Joiner and Associates.
Srull, T. K., & Wyer, R. S. (1989). Person memory and judgment. Psychological Review, 96, 58-83.
Sulsky, L. M., & Day, D. V. (1992). Frame-of-reference training and cognitive categorization: An empirical investigation of rater memory issues. Journal of Applied Psychology, 77, 501-510.
Tsui, A. S., Egan, T. D., & O'Reilly, C. A. III (1992). Being different: Relational demography and organizational attachment. Administrative Science Quarterly, 37, 549-579.
Tsui, A. S., & O'Reilly, C. A. III (1989). Beyond simple demographic effects: The importance of relational demography in superior-subordinate dyads. Academy of Management Journal, 32, 402-423.
Turban, D. B., & Jones, A. P. (1988). Supervisor-subordinate similarity: Types, effects, and mechanisms. Journal of Applied Psychology, 73, 228-234.
Turban, D. B., Jones, A. P., & Rozelle, R. M. (1990). Influences of supervisor liking of a subordinate and the reward context on the treatment and evaluation of that subordinate. Motivation and Emotion, 14, 215-233.
Turner, J. c. (1987). Rediscovering the social group: A self-categorization theory. Oxford: Blackwell.
Waldman, D. A., Atwater, L. E., & Antonioni, D. (1998). Has 360 degree feedback gone amok? Academy of Management Executive, 12(2), 86-94.
Watson, D., Hubbard, B., & Wiese, D. (2000). Self-other agreement in personality and affectivity: The role of acquaintanceship, trait visibility, and assumed similarity. Journal of Personality and Social Psychology, 78, 546-558.
Wayne, S. J., & Liden, R. C. (1995). Effects of impression management on performance ratings: A longitudinal study. Academy of Management Journal, 38, 232-260.
Wexley, K. N., & Klimoski, R. (1984). Performance appraisal: An update. In K. M. Rowland & G. R. Ferris (Eds.), Research in personnel and human resources management, 2, 35-79. Greenwich, CT: JAI Press.
Wholers, A. J., & London, M. (1989). Ratings of managerial characteristics: Evaluation difficulty, co-worker agreement, and self-awareness. Personnel Psychology, 42, 235-261.
Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35, 151-175.
Received 20 April 1999; revised version received 3 January 2001
Judy P. Strauss * Department of Management/HRM, California State University, Long Beach, California, USA Murray R. Barrick Department of Management & Organizations, University of Iowa, Iowa, USA Mary L. Connerley Department of Management, Virginia Polytechnic Institute and State University, Blacksburg, USA
* Requests for reprints should be addressed to Professor Judy Strauss, Department of Management/HRM, California State University, Long Beach, California 90840-0604, USA (e-mail: firstname.lastname@example.org).
|Printer friendly Cite/link Email Feedback|
|Author:||Strauss, Judy P.; Barrick, Murray R.; Connerley, Mary L.|
|Publication:||Journal of Occupational and Organizational Psychology|
|Date:||Dec 1, 2001|
|Previous Article:||Assessors' use of personality traits in descriptions of assessment centre candidates: A five-factor model perspective.|
|Next Article:||Performance disparities between whites and ethnic minorities: real differences or assessment bias?|