Meta-Analysis of Bidirectional Relations in Personality-Job Performance Research

Abstract Standard meta-analysis (Hunter & Schmidt, 1990a) assumes unidirectional (i.e., all positive or all negative) predictor-criterion relations. We challenge this assumption in the context of personality-job performance linkages based on several lines of evi- dence supporting the coexistence of true positive and true negative correlations in- volving the same trait. Subjecting such bidirectional relations to standard meta-analysis will underestimate effect sizes to an unknown degree owing to cancel- lation of positive and negative values. A modification of standard procedures is pro- posed that accounts for the possibility of bidirectionality. It employs successive itera- tion of an initial estimate of the absolute correlation, accounting for sample sizes and the sampling distribution of the correlation.