Meta-analysis of correlation coefficients: A Monte Carlo comparison of fixed- and random-effects methods

The efficacy of the Hedges and colleagues, Rosenthal-Rubin, and Hunter-Schm idt methods for combining correlation coefficients was tested for cases in which population effect sizes were both fixed and variable. After a brief t utorial on these meta-analytic methods, the author presents two Monte Carlo simulations that compare these methods for cases in which the number of st udies in the meta-analysis and the average sample size of studies were vari ed. In the fixed case the methods produced comparable estimates of the aver age effect size; however, the Hunter-Schmidt method failed to control the T ype I error rate for the associated significance tests. In the variable cas e, for both the Hedges and colleagues and Hunter-Schmidt methods, Type I er ror rates were not controlled for meta-analyses including 15 or fewer studi es and the probability of detecting small effects was less than.3. Some pra ctical recommendations are made about the use of meta-analysis.