Design and Analysis of Multisite Cluster-Randomized Trials Targeting (Conditional) Mediation Effects

• In Event: New Insights for Designing Multilevel Models: Predicting Random Effects, Using Predictor Aggregates, and Mediation Planning

Abstract

There has been a strong and enduring national interest in improving teacher and student outcomes through teacher professional development programs. A central purpose in studies of teacher professional development is to examine mediational processes to understand how professional development improves instruction in ways that elevate student learning. Within this context, multisite cluster-randomized trials (MSCRTs) have emerged as a principal design. Despite the widespread use of these designs and this theory of action, design and analysis methods for investigating mediation effects in MSCRTs is underdeveloped. In this study, we develop analysis methods and expressions to predict power and guide the design of MSCRTs probing mediation. We illustrate the results with an example and an R ShinyApp.

Authors

• Fangxing Bai, Montana State University

• Benjamin Kelcey, University of Cincinnati

• Amota Ataneka, University of Cincinnati

• Yanli Xie, Florida State University

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