Methodology for Dealing With Duplicate Study Effects in a Meta-Analysis

Abstract

Study effects developed from the same data set and duplicated in more than one publication likely will bias the aggregated effects in a meta-analysis. This threat to the validity of meta-analyses in organizational research is significant. Methods of detecting overt and covert duplication of publication are presented through an analysis of five published meta-analyses. An unpublished meta-analysis of salesperson turnover is used to present the implications of nonindependence. Options for the correction of duplication are discussed. It is recommended that study effects gathered during the literature search use the proposed screening heuristic to detect and address duplicate study effects.