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 1: Introduction
Chapter 2: Definition of a Mediator
Chapter 3: Theoretical Framework
Chapter 4: Statistical Power Formulas
Chapter 5: Empirical Analysis
Chapter 6: Summary and Conclusions
Appendix A: Proof of Equation (18)
References
List of Tables
List of Figures
PDF & Related Info

Chapter 4: Statistical Power Formulas

This chapter presents formulas for the non-centrality parameter in (6) using OLS and IV methods for estimating the mediator associations in (3) and (4). Asymptotic formulas are presented due to the considerable complexity of calculating finite sample moments for IV estimators. OLS estimators are used rather than GLS estimators (which are more efficient), because the IV approach and corrections for mediator measurement error are much more complex using the GLS approach.