Designing Multilevel Studies in Sub-Saharan Africa

• In Event: Poster Session 3
  In Poster Session: Research Methods and Approaches to Examining Academic and Behavioral Outcomes

Abstract

In planning school-based education studies that assign schools or districts and samples of their respective students to treatment conditions, effective and efficient design requires plausible values of the intraclass correlation coefficient and the variance explained by covariates during the design stage. The purpose of this study was to improve the design of school improvement studies in 15 countries in sub-Saharan Africa by providing empirical estimates of these design parameters. We built on previous work by developing estimates of intraclass correlation coefficients for schools, districts, and countries for student achievement outcomes and explore how these coefficients might be reduced through covariance adjustment on common covariates. In turn, we illustrate the use of these estimates in multisite (blocked) and cluster-randomized designs.

Authors

• Zuchao Shen, University of Cincinnati

• Benjamin Kelcey, University of Cincinnati