UC Berkeley

We use quantile treatment effects estimation to examine the consequences of a school voucher experiment across the distribution of student achievement. In 1997, the School Choice Scholarship Foundation granted $1,400 private school vouchers to a randomly-selected group of low-income New York City elementary school students. Prior research indicates that this program had no average effect on student achievement. If vouchers boost achievement at one part of the distribution and hurt achievement at another, zero or small mean effects may obscure theoretically important but offsetting program effects. Drawing upon prior research related to Catholic schools and school choice, we derive three hypotheses regarding the program’s distributional consequences. Our analyses suggest that the program had no significant effect at any point in the skill distribution.