Educational Policy Symposium: A Unified Conceptual and Empirical Approach to Understanding Treatment Heterogeneity
Tran Dang Keys, Ph.D. Student
George Farkas, Professor
Margaret R. Burchinal, Professor
Weilin Li, Ph.D. Student
Erik A. Ruzek, Ph.D. Student
Abstract
While policymakers justifiably attach considerable weight to experimental evaluations of child care programs, there is much to be learned from rigorous analyses of longitudinal data that are more representative of the population. Our paper applies meta-analytic techniques to summarize results from original analyses of four longitudinal data sets to estimate variation in preschool. In summary, the consistency of these generally null results and the precision with which they are estimated across the different databases, multiple outcomes, and multiple child care quality measures suggests the following: