| Popis: |
One effective way to consolidate studies is to use Bayesian meta-analyses. Bayesian methodology allows researchers to integrate their own predictions into the analysis. However, researchers may struggle to decide which prior distribution is appropriate. To help researchers in this pursuit, we outline a method to develop prior distributions based on past data. We constructed three prior distributions based on 100 social psychology meta-analyses of the past 20 years. We fit distributions to the data using maximum likelihood estimation. We then tested the effectiveness of our prior distributions against uninformed alternatives using testing data that had been separated out during the fitting process. Our prior distributions resulted in larger Bayes factors than the alternatives. Our results showcase the methodology’s potential to develop informed prior distributions for various purposes in the future. |
