Association for Public Policy Analysis & Management

Universal Preschool programs, such as those established in Georgia and Oklahoma, are designed to increase the accessibility of a quality preschool education statewide. The literature consistently observes that these programs increase both preschool participation and improve the academic achievement of disadvantaged participants. Traditionally, we evaluate the effectiveness of these programs using test scores, which serve as a proxy for human capital. While test scores inform us about a child's cognitive achievements, they tend to provide little insight into a child's non-cognitive skill. In that respect, test scores provide an incomplete measure of human capital. This paper uses the diagnosis of certain intellectual and behavioral disorders as indicators for cognitive and non-cognitive skills, where these outcomes can be thought of as complementary to test scores. Using data from the National Survey of Children's Health and the Current Population Survey October Supplement, I use a two-sample instrumental variables technique to estimate the effect of preschool participation on the probability of ever having been diagnosed with certain intellectual and behavioral disorders in childhood. I find suggestive evidence that, among low-SES children, participating in preschool reduces the probability of ever having been diagnosed with behavioral or conduct problems, and requiring the use of special therapy.