OBJECTIVE To investigate the impact of a higher publishing probability for statistically significant positive outcomes on the false-positive rate in meta-analysis. DESIGN Meta-analyses of different sizes (N=10, N=20, N=50 and N=100), levels of heterogeneity and levels of publication bias were simulated. PRIMARY AND SECONDARY OUTCOME MEASURES The type I error rate for the test of the mean ef...

Publication bias is one threat to validity that researchers conducting meta-analysis studies confront. Although statistical methods for detecting publication bias have surfaced in the literature (e.g., Begg Rank Correlation, Egger Regression, Funnel Plot Regression, and Trim & Fill), many researchers rely on visual inspection of funnel plots. This program was created to provide meta-analysts wi...

Previous research has introduced the threat of publication bias to meta-analytic reviews in management and industrial/organizational (I/O) psychology research. However, a comprehensive review of top-tier journals demonstrates that more than two thirds of meta-analytic studies in management and I/O psychology ignore the issue. Of the studies that do empirically evaluate publication bias, almost ...