Abstract:
|
Cluster randomized trials (CRTs), where clusters are randomized to comparison arms but measurements are taken on individuals, are commonly used to evaluate interventions in public health, education, and the social sciences. Analysis is often conducted on individual-level outcomes, and such analysis methods must take into account that outcomes for members of the same cluster tend to be more similar than those for members of other clusters. A popular analysis technique is generalized estimating equations (GEE). However, many studies randomize a small number of clusters (e.g., < 30), and in this case, the GEE standard errors obtained from the sandwich variance estimator will be biased, leading to inflated type I errors. A number of bias-corrected standard errors have been proposed and studied to account for this bias, yet most have not yet been implemented in Stata, which is a statistical software commonly used for the analysis of CRTs. We illustrate the implementation of bias-corrected standard errors using our newly-created Stata program xtgeebcv, discuss suggestions about which finite-sample corrections to use in which situations, and consider areas of future research.
|