The Effects of Head Start and Juvenile Delinquency

Nathaniel Balis, Georgetown Public Policy Institute
Jens Otto Ludwig, Georgetown University

Very little is currently known about the effects of the Head Start program on later criminal activity. We re-examine this question using the National Education Longitudinal Survey (NELS), which provides longitudinal data on a nationally representative sample of teens interviewed between 1988 and 1994. The parents of sampled students are asked whether their children participated in Head Start or other pre-school programs prior to enrolling in elementary school. The students themselves are asked to report on their involvement as adolescents in behaviors such as drinking, smoking, drug use, sexual activity, and criminal behavior. We address the problem raised by family self-selection into the Head Start program by controlling for family, student and school characteristics. We also provide instrumental variables (IV) estimates that identify the effects of Head Start using plausibly exogenous variation across states in the criteria through which income-eligible children are chosen to participate in the program, as well as across-state variation in local, state and Federal funding. The paper concludes by comparing the average per-student cost of Head Start with the program's benefits, as measured by the dollar value of changes in non-academic outcomes reported in our study together with the value of academic improvements from previous research.

(Return to Program Resources)

Updated 05/20/2006