Jonathan Moyer, MST

Cluster randomized trials (CRTs) are ones in which groups of individuals – such as entire hospitals, entire villages, etc. –  are randomly assigned to treatments.  Where they are applicable, CRTs can be more convenient and less invasive than conventional, individually randomized trials. However, because measurements within a cluster are correlated, power calculation for these studies is more complicated than individual randomized trials. Calculating power via simulation is a way to avoid these computational complexities and is the focus of my current research.