Bayesian Approaches to Designing Replication Studies

Abstract

Replication studies are essential for assessing the credibility of claims from original studies. A critical aspect of designing replication studies is determining their sample size; a too-small sample size may lead to incon-clusive studies whereas a too-large sample size may waste resources that could be allocated better in other studies. Here, we show how Bayesian approaches can be used for tackling this problem. Bayesian approaches allow researchers to incorporate both the data from the original study and external knowledge into the sample size determination, thereby accounting for uncertainty and potentially leading to efficiency gains. Our approach can be applied to any type of planned replication analysis, catering to various stakehold-ers and enabling conclusive inferences based on multiple approaches. We use data from social-behavioral experiments to illustrate how Bayesian approaches can help design informative and cost-effective replication studies. We also show how the methods can be applied using the R package BayesRepDesign.