ABSTRACT
In Chapter 10, we presented methods for the analysis of CRTs based on cluster-level summaries. We also discussed how these methods can be extended to adjust for covariates measured at either cluster level or individual level. As we noted, these methods have been shown to perform robustly over a wide range of circumstances. In particular, they provide reliable results when there are a relatively small number of clusters in each treatment arm, as is the case in many CRTs.
