California Center for Population Research

Critics of Head Start contend that many programs spend too much money on programs extraneous to education. On the other hand, Head Start advocates argue that severely disadvantaged children need a broad range of services. Given the available evidence, it has been impossible to assess the validity of these claims. In this study, we match detailed administrative data with data on child outcomes from the National Longitudinal Survey of Youth 1979, including test scores, behavior problems, and grade repetition. We find that former Head Start children have higher reading scores and are less likely to have been retained in grade where Head Start spending was higher. Holding per capita expenditures constant, children in programs that devoted higher shares of their budgets to education and health have fewer behavior problems and are less likely to have been retained in grade. However, when we examine specific educational inputs holding per capita expenditures constant, only pupil/teacher ratios matter.