Method for Sample Size Determination for Cluster-Randomized Trials Using the Bayes Factor

Determining sample size is crucial in research study design. The hierarchical structure ofthe data in cluster-randomized trials (CRTs) complicates this process, thereby necessitatingthe determination of the sample size at each level. Most methods for these trials are basedon null hypothesis significance testing (NHST), which has numerous pitfalls. Thesedrawbacks can be avoided, however, by using the Bayes factor. Current methods forsample size determination when using the Bayes factor are limited to trials withoutmultilevel structure. This study presents a method to determine the sample size for aone-period two-treatment parallel cluster-randomized trial using the Bayes factor. Weintroduce the implementation of this method in an R package. Simulation results showthat the required sample size increases with decreasing effect sizes and with increasingintraclass correlation and Bayes factors.