The National Center on School Choice is funded by the United States Department of Education's Institute of Education Sciences. The Center exercises national leadership in school choice research, including charter and magnet schools, private school vouchers, teacher recruitment, school management, and state policymaking.

Broadening Analysis of Randomized School Voucher
Experiments

Paul Peterson, Harvard University
William Howell, University of Chicago

What are the effects of vouchers on students who switch from public to private schools and how do we best analyze them? These questions frame the focus of our study.  Previously, Harvard researchers sought to answer these questions by conducting randomized field trials of private school voucher experiments in New York City; Dayton, Ohio; and Washington, D.C.  Although randomized experiments have obvious advantages – chiefly, they alleviate the selection effects that regularly plague observational studies of public and private schools – they also present problems of their own.  When these trials are implemented, one unavoidable complication is noncompliant behavior; that is, someone offered a voucher may not use it, or someone not offered a voucher may nevertheless switch to a private school. In many cases there are good reasons to believe that those who do not comply differ systematically from those who do.

During the past year, we revisited the data from the New York City voucher program. We estimated a wide range of alternative models that explored student attrition patterns, considered the influence of additional covariates in models of student achievement, and assessed the influence of non-response rates on estimated impacts. These findings were published in a revised edition of William Howell and Paul E. Peterson, The Education Gap: Vouchers and Urban Schools (Brookings Institution, 2006).