An examination of the prediction of overall, task, and contextual performance using three selection measures for a service-type occupation

John Michael Avis


This study responds to recent calls to determine the best predictors of job performance in a rapidly growing market segment, namely lower-level cognitive ability (e.g., service-type) jobs (Cascio, 1995). Using data from 203 clerical employees in a Southeastern state government organization, the study examined the extent to which three types of selection measures (e.g., cognitive ability, personality dimensions, and a motor work sample) predicted supervisory ratings of overall, task, and contextual performance (c.f., Borman and Motowidlo, 1993). The results suggest that the best predictor of performance depends on the way in which performance is defined. Specifically, cognitive ability and four of the Big Five personality measures (i.e., extraversion, conscientiousness, agreeableness, and emotional stability) were related to both overall and contextual performance. Moreover, it was found that the joint use of cognitive ability and personality improved the level of prediction achieved when measuring overall and contextual performance in service-type jobs. That is, personality provided incremental validity over cognitive ability in the prediction of both overall and contextual performance, and vice versa. On the other hand, only cognitive ability predicted task performance. Finally, neither customer service orientation nor a work sample was related to any of the criteria. Implications of the findings, limitations and suggestions for future research are addressed.