Title: The Effect of Educational Vouchers on Academic and Non-academic Outcomes
Economic research examining how educational intervention programs affect primary and secondary schooling focuses largely on test scores although the interventions can affect many other outcomes. This presentation is based on a series of papers that examine how an educational intervention, a voucher program, affected not only educational outcomes, including test scores, but especially how it affected other student outcomes including altruism, patience and confidence. The voucher program used a lottery to allocate scholarships among low-income applicant families with children in K-8th grade. By exploiting the random allocation of the voucher lottery to identify the voucher effects, and using experimental economic methods, we measure the effects of the intervention.
We find that the educational intervention had no significant effect on mathematical test scores, but positively reduced student’s suspensions and increased their parents’ involvement in their education. We note that test scores were not significantly correlated with the non-academic outcome measures we collected, implying that test scores are not a sufficient statistic for all possible effects of the educational intervention policy. Regarding our most important contribution, the effect of the educational policy on non-academic outcomes, we find that the educational intervention positively affects students' altruism towards charitable organizations but not towards their peers. We also find that the voucher directionally increased children’s patience and reduced overconfidence, but neither effect reached the level of statistical significance.
Given the uniqueness of our data to measure children’s behavior, our work also adds to the small literature on children’s economic behavior. We find that older children and girls are more generous, children give more to charities than to their peers, and children and their parents’ generosity are not significantly correlated. Further, children’s choices are consistent with hyperbolic discounting, boys are less patient than girls, older children are more patient, and parent’s and children’s patience are not significantly correlated. We also find that although more than 25 percent of children do not make rational inter-temporal choices within a single two-period time frame, we cannot find variables that explain this behavior other than age and standardized mathematical achievement test scores.