Writing meta-analytic reviews.

This article describes what should typically be included in the introduction, method, results, and discussion sections of a meta-analytic review. Method sections include information on literature searches, criteria for inclusion of studies, and a listing of the characteristics recorded for each study. Results sections include information describing the distribution of obtained effect sizes, central ten-dencies, variability, tests of significance, confidence intervals, tests for heterogeneity, and contrasts (univariate or multivariate). The interpretation of meta-analytic results is often facilitated by the inclusion of the binomial effect size display procedure, the coefficient of robustness, file drawer analysis, and, where overall results are not significant, the counternull value of the obtained effect size and power analysis. The purpose of this article is to provide some guidelines for the preparation of meta-analytic reviews of literature. Meta-an-alytic reviews are quantitative summaries of research domains that describe the typical strength of the effect or phenomenon, its variability, its statistical significance, and the nature of the moderator variables from which one can predict the relative strength of the effect or phenomenon (Cooper, 1989; Glass,