Two national retail companies provided rosters of current store detectives with at least 1 year of tenure as a store detective. A total of 277 of these were randomly selected to receive a mailed packet that contained an individual characteristic sheet and a NEO PI-R personality instrument booklet and a self-rating answer sheet. The respondents' supervisors administered and returned the complete Wonderlic Personality Test (WPT) and performance instruments. All the supervisors of the selected participants were mailed a packet that included a job performance review instrument for each participant they managed. A total of 201 detectives and their managers completed and returned the materials. The three hypotheses tested in this study pertained to the relationship between biographical data, personality traits, cognitive ability, and the three measures of job performance. The three measures of job performance were the overall performance assessment score over the previous 12 months, success at apprehending and processing theft offenders or at deterring individual thefts and recovering merchandise, and potential as a future leader as indicated by the supervisor. In the three multiple regression models used, standardized regression coefficients were used to indicate the variables that contributed the most to explaining the variance in job performance scores. Based on the findings, the initially proposed model was revised to indicate the job/role specific nature the WPT (measure of cognitive ability) has in predicting the leadership role, the role of low prior loss-prevention experience, and various personality facets such as neuroticism and conscientiousness. 2 figures and 42 notes