Publication bias

Meta-analysis is the statistical method for synthesizing studies on the same topic and is often used in clinical psychology to quantify the efficacy of treatments. A major threat to the validity of meta-analysis is publication bias, which implies that some studies are less likely to be published and are therefore less often included in a meta-analysis. A consequence of publication bias is the overestimation of the meta-analytic effect size that may give a false impression with respect to the efficacy of a treatment, which might result in (avoidable) suffering of patients and waste of resources. Guidelines recommend to routinely assess publication bias in meta-analyses, but this is currently not common practice. This chapter describes popular and state-of-the-art methods to assess publication bias in a meta-analysis and summarizes recommendations for applying these methods. We also illustrate how these methods can be applied to two meta-analyses that are typical for clinical psychology such that psychologists can readily apply the methods in their own meta-analyses.