Experimental Design and Analysis of Multiple-Group Individually Randomized Group Trials

• In Event: Experimental Design and Hierarchal Linear Models

Abstract

We consider a class of individually-randomized group trials that introduces a (partially)cross-classified structure in the treatment condition(only). The novel feature of this design is that the nature of the treatment induces a clustering structure that involves two or more non-nested groups among individuals in the treatment condition but not in the control condition. Although a literature review suggests that multiple-group individually-randomized group-trials are present in theory and practice, their appropriate analysis has largely gone unexamined. We develop statistical theory that motivates this design, map out its use in practice, and develop estimation methods to track treatment effects, their sampling variability and the power to detect them. The results provide the core tools to design and analyze these types of studies.

Authors

• Benjamin Kelcey, University of Cincinnati

• Fangxing Bai, Montana State University

• Amota Ataneka, University of Cincinnati

• Yanli Xie, Florida State University

• Kyle T. Cox, University of North Carolina - Charlotte