|
Abstract:
|
Public health interventions are often designed to target identifiable groups of people such as communities, schools, and workplaces. The group randomized trial (GRT) is regarded as the gold standard for evaluating these interventions. However, randomization may cause ethical concerns if some groups are denied a beneficial intervention. In the regression discontinuity (RD) design, individuals are assigned to treatment based on the levels of some pre-test measure, thereby allowing that those most in need of treatment receive it. In this article, we extend previous work on sample size and power for the RD and related cut-off based designs for individual randomized trials to the group randomized setting. We examine the power of these designs as a function of intraclass correlation, number of groups, and number of members per group and compare results to the traditional GRT.
|
