Designing Group- and Multisite Group-Randomized Studies of Teacher Development

• In Event: Roundtable Session 11
  In Roundtable Session: Evaluating Professional Development, Teaching Practices, and Teacher Advancement

Abstract

Though group designs may be theoretically favorable, prior evidence has suggested that they may be challenging in professional development studies because well-powered designs will typically require large sample sizes or large effect sizes. We develop a theoretical and empirical basis for the design of professional development studies by building on previous work in three ways: (a) developing estimates of intraclass correlation coefficients for teacher outcomes using two- and three-level data structures, (b) developing estimates of the variance explained by covariates, and (c) exploring the absolute and relative sensitivity of several types of multilevel designs. The results from these analyses are intended to guide researchers in making more informed decisions about the tradeoffs and considerations involved in selecting different study designs.

Authors

• Benjamin Kelcey, University of Cincinnati

• Jessaca K. Spybrook, Western Michigan University

• Geoffrey C. Phelps, Educational Testing Service

• Nathan Jones, Boston University