Bayesian Evaluation of Replication Studies

Abstract

In this paper a method is proposed to determine whether the result from an original study is corroborated in a replication study. The paper is illustrated using two replication studies and the corresponding original studies from the Reproducibility Project: Psychology by the Open Science Collaboration. This method emphasizes the need to determine what one wants to replicate from the original paper. This can be done by translating the research hypotheses formulated in the introduction into informative hypotheses, or, by translating the results into interval hypotheses. The Bayes factor will be used to determine whether the hypotheses resulting from the original study are corroborated by the replication study. Our method to assess the successfulness of replication will better fit the needs and desires of researchers in fields that use replication studies.