A personality-based similar-to-me effect in the employment interview: Conscientousness, affect-versus competence-mediated interpretations, and the role of job relevance

Canadian Journal of Behavioural Science, Jan 2003 by Greg J Sears, Patricia M Rowe

In addition to matching actors on physical appearance and nonverbal behaviour (vocal and visual cues) in the study's design, repeated-measures ANOVAs were computed to test for potential actor effects. Actor mean comparisons on the Big Five and the four hirability dimensions failed to yield significant differences for any of the primary independent or dependent variables (i.e., conscientiousness, perceived similarity, affect toward applicant, perceived competence, job suitability) relevant to the hypothesized similarto-me effect. It should be noted, however, that the .05 significance level was exceeded for two of the ancillary "Big Five" comparisons: extraversion, F(1,36) = 16.88, p

In terms of psychometric characteristics, each of the measures employed in the study evinced sound internal consistency. Mean alpha estimates averaged across the "Big Five" constructs were .84 for both selfratings (prescreening) and observer ratings (across the two candidates observed). Alpha estimates for conscientiousness (self = .87; observer = .86) compared favourably to those obtained in Saucier's normauve sample (self = .84; observer = .86). The mean alpha estimate for the competency scale averaged across dimensions was .82 (for both candidates observed). The mean alpha for the hirability measure across dimensions was .84, eclipsing the coefficient of .82 reported by Howard and Ferris.

MANIPULATION CHECK FOR INTERVIEW VIDEOTAPES

To gauge the construct validity of conscientiousness depicted in the interview tapes, the effect of candidate conscientiousness (high and low) on "Big Five"and competency ratings was examined. As expected, the highly conscientious candidate was rated significantly higher on conscientiousness, F(1,38) = 30.05, p .69; Actor A - high conscientiousness candidate: M = 56.70; Actor B - high conscientiousness candidate: M = 57.55; Actor A - low conscientiousness candidate: M = 47.30; Actor B - low conscientiousness candidate: M = 46.65. Based on these data, it appears both actors displayed desired levels of conscientiousness, and raters correctly differentiated candidates in the high and low conscientiousness conditions.

Consonant with research reporting appreciable intercorrelations among the "Big Five" factors (e.g., Briggs, 1992; Widiger & Trull, 1997), two "Big Five" constructs varied as a function of candidate conscientiousness. Results revealed that relative to the candidate low in conscientiousness, the highly conscientious candidate elicited higher ratings on openness to experience, F(1,39) = 8.02, p

Results pertaining to the competency measure seem to corroborate these interpretations. The highly conscientious candidate received significantly higher ratings than the low conscientiousness candidate on four of seven competencies: cognitive capacity, F(1,39) = 22.70, p

To further verify that conscientiousness constitutes the basis of rater-applicant similarity in this study, self-ratings for high and low conscientiousness raters were compared on the "Big Five" constructs (i.e., openness to experience, emotional stability) that differed between high and low conscientiousness candidates. Other potential rater confound variables (e.g., interview experience, age, faculty/major) were also inspected. No systematic differences were uncovered in any of these analyses. In concert, this overall pattern of results reinforces the validity of the interview videotapes in tapping "Big Five" conscientiousness and portraying designated variations in this construct. Furthermore, these data confirm that in this study conscientiousness forms the foundation of "actual" rater-applicant similarity.