The Effects of Vouchers and Private Schools in Improving Academic Achievement: A Critique of Advocacy Research

Article excerpt

I. INTRODUCTION: VOUCHERS AND ACHIEVEMENT IN PUBLIC AND PRIVATE SCHOOLS

Proposals and programs to use publicly-funded vouchers to move children from public to private schools are perhaps the most controversial educational reform of the last two decades. While there are a number of compelling arguments in support of vouchers, among the most prominent is the idea that they will result in improved student achievement. From the earliest days of voucher proposals, advocates have argued that competitive pressures will drive schools to improve, thereby improving results for children.

In the current educational policy environment, characterized by choice programs such as vouchers, charter schools, and openenrollment plans, it is increasingly important for reform measures to demonstrate tangible gains in student outcomes. Student achievement scores, as measured by standardized tests, are by far the most prominent of these outcomes and serve as the primary measure for school accountability. Reforms at all scales, from classroom instructional practices to federal initiatives, are evaluated on these grounds, and programs that do not produce results often lose support. As policymakers have increasingly emphasized achievement scores, the argument that choice plans will improve student performance has become increasingly central to school choice advocacy. Choice proponents regularly turn to achievement data to demonstrate the effectiveness of choice plans, and some have asserted that there is a consensus that school choice "works" in this regard.

As choice plans have developed over the last generation, we now have a relatively extensive empirical basis from which to evaluate outcomes of programs designed to send children to private schools. However, despite the rhetoric coming from many policy advocates, the overall results reflected in the research do not provide the compelling support for voucher programs that many expected to find. Indeed, a comprehensive review of the research indicates that the initial optimistic expectations from theoreticians and policy advocates for improved academic outcomes are not supported by the growing body of research on this question.

This Article focuses specifically on the use of achievement data in the assessment of vouchers for private schools. Vouchers seem to be the most controversial form of school choice because they distribute public funds directly to schools beyond the purview of public accountability mechanisms. After a brief review of the history of voucher programs and the role of achievement outcomes in voucher advocacy, we examine the initial, influential research on student achievement in public versus private schools and assess the claim that private schools are more effective than public schools at raising student achievement. The superiority of private schools is often presented as common sense, but the research results are far less clear. Turning to more recent research, we analyze the claims of a "consensus" about the effectiveness of voucher programs for improving student outcomes. Researchers supported by voucher advocacy organizations typically use flawed methodology in their attempts to find a positive academic impact for vouchers, misrepresent the findings of other research studies, and selectively ignore studies that contradict their claims. In the final section, we examine recent large-scale studies regarding student achievement in public and private schools. The picture that emerges suggests that public schools do remarkably well in comparison to private schools when student background is considered. This comprehensive evidence indicates that public schools are on average at least as effective, and in some cases more effective, as private schools when measured by student achievement outcomes. We conclude that while improved student achievement remains the most prominent argument in favor of voucher programs, this claim is not supported by the weight of the best available evidence. …