Skip Navigation
Technical Methods Report: Do Typical RCTs of Education Interventions Have Sufficient Statistical Power for Linking Impacts on Teacher Practice and Student Achievement Outcomes?

NCEE 2009-4065
October 2009

Chapter 5: Empirical Analysis

This chapter uses the non-centrality parameter formulas in (18), (19), (20) and (26) to conduct a simulated "typical" statistical power analysis for γ1 for school-based RCTs. The goal is to identify the number of study schools that are required to ensure that an RCT has sufficient power for detecting R2 values for mediator effects that are likely to be found in practice. The focus is on RCTs for elementary school students in low performing schools, a common target population for experiments conducted in education research.

The first part of this chapter discusses the key issue of identifying benchmark R2 values that can realistically be found in practice using the approach displayed in Figure 3.1. The second part discusses additional assumptions that are required for the statistical power calculations, and the final part presents the empirical results.