Differential effects of high‐quality child care

In policy research a frequent aim is to estimate treatment effects separately by subgroups. This endeavorbecomes a methodological challenge when the subgroups are defined by post‐treatment, rather thanpre‐treatment, variables because if analyses are performed in the same way as with pre‐treatmentvariables, causal interpretations are no longer valid. The authors illustrate a new approach to this challengewithin the context of the Infant Health and Development Program, a multisite randomized study that providedat‐risk children with intensive, center‐based child care. This strategy is used to examine thedifferential causal effects of access to high‐quality child care for children who would otherwise haveparticipated in one of three child care options: no non‐maternal care, home‐based non‐maternalcare, and center‐based care. Results of this study indicate that children participating in the first twotypes of care would have gained the most from high‐quality center‐based care and, moreover, wouldhave more consistently retained the bulk of these positive benefits over time. These results may have implicationsfor policy, particularly with regard to the debate about the potential implications of providing universal childcare. © 2002 by the Association for Public Policy Analysis and Management.