The Religious Replication Project: Using Pre-registered Replications and Bayesian Statistics to Improve the Experimental Study of Religion

The scientific study of religion can provide fascinating insight in the potential causes and consequences of religiosity, but the reliability of many studies in the field remains to be determined – especially in the light of the recent ‘replication crisis’ that has haunted scientific research in general. Although recently many new methodological initiatives have been taken to improve the credibility of scientific research (e.g., the use of pre-registration; adversarial collaboration; independent replication studies), the scientific study of religion has largely remained beyond the scope of these approaches. Therefore, we propose to initiate the Religious Replication Project (RRP). The project activities will consist of (1) conducting a pre-registered replication report, (2) setting up multi-lab pre-registered replication studies and (3) seeking adversarial collaboration with the original authors of theoretically disputed effects. Multiple labs in different countries will collaborate on conducting pre-registered adversarial replication attempts of at least three studies. Bayesian statistics will be used as a methodological tool that allows to directly quantify and continually monitor the evidential value of the studies with respect to the null and the alternative hypothesis. Through this approach the ultimate aim of this project is to establish the robustness of at least three findings that have shaped theory development regarding the causes and consequences of religion (e.g., the relation between religion and agency, prosociality, and self-control), thereby giving a strong methodological as well as a theoretical impetus to the field. In addition, by directly fostering replication studies in the field of the psychology of religion and the cognitive science of religion, this project will serve as a hallmark for the scientific study of religion in general, by setting a new standard for good scientific practice.