A comparison of the predictive validity of assessment center evaluations versus traditional measures in forecasting supervisory job performance: Interpretive implications of criterion distortion for the assessment paradigm.

Collected data on 799 employees at a large manufacturing firm who were promoted to supervisory positions during a 4-yr period, including criterion data gathered from supervisors from 1-mo to several years after the assessment. 71 predictor variables, 29 assessment evaluations, and 21 criterion variables were aggregated for ease of interpretation. Results show that (1) both assessment-center evaluations (personal history survey, short arithmetic test, measure of general ability, interview, and work exercises) and traditional predictors (supervisor's ratings) were unrelated to job performance, but (2) assessment-center evaluations were predictive of promotability. It is suggested that by the nature of the complex issues that assessment centers were designed to address, a psychometrically clean evaluation of them may be precluded. (13 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).