Power Analysis for Models of Polynomial Change in Four-Level Cluster Randomized Designs

• In Event: 31.038 - Power Analysis in Multilevel Modeling

Abstract

Field experiments in education frequently assign schools to treatment or control conditions. Such experiments sometimes incorporate a longitudinal component where for example students are followed over time and student outcomes are measured repeatedly for several grades. This study provides methods for power analysis in four-level polynomial change models for cluster randomized designs (i.e., treatment assigned to units at the fourth level). Power computations take into account clustering effects at the third (classroom) and fourth (school) levels, the number of measurement occasions, and the impact of sample sizes at different levels (e.g., number of schools, classrooms, and students). An illustrative example shows how power is computed.

Authors

• Spyros Konstantopoulos, Michigan State University

• Wei Li, University of Alabama