Statistical Power and Optimal Design of Multisite Experiments Detecting Moderation and Main Effects

• In Event: Experimental Design and Hierarchal Linear Models

Abstract

Studies probing moderation effects in conjunction with main effects provide a richer and more comprehensive portrait of the effectiveness and equity of interventions. Prior research has separately developed strategies to design studies that probe main and moderation effects, including sample size determination through power analyses and optimal sampling plans. However, in most instances, requisite sample sizes and optimal sampling will differ (often remarkably) by effects. In this study, we develop a joint optimal design framework for multisite randomized trials investigating both moderation and main effects, which can identify the jointly optimal sample allocation that uses the minimum budget to detect both main and moderation effects. We illustrate the methods through examples and implement the methods in the R package XXX.

Authors

• Zuchao Shen, University of Georgia

• Benjamin Kelcey, University of Cincinnati

• Zhenqiu Laura Lu, University of Georgia

• Jingyang Li, University of Georgia