Choice Students Score Better

Published January 1, 1997

A new performance analysis of Milwaukee, Wisconsin’s school voucher program has found that low-income minority students do better with choice. After three and four years in the program, those students scored three and five percentage points better in reading than their counterparts in the city’s public schools. In math, the choice students scored five and twelve percentage points better.

The new analysis, published in August 1996 by researchers Jay P. Greene, Paul E. Peterson, Jiangtao Du, Leesa Boeger, and Chris L. Frazier of Harvard University and the University of Houston, corrects flaws in the only previously conducted review of the Milwaukee Parental Choice Program.

That earlier review–conducted on an ongoing basis since the program’s inception by John Witte of the University of Wisconsin-had found high levels of parental satisfaction and teacher morale but no educational benefits for students.

Witte’s data and methodology were made available to outside scholars for evaluation and additional analysis for the first time last summer. After reviewing Witte’s work, the Harvard/Houston team concluded that Witte had selected an inappropriate control group for his study. As a result, “inappropriate comparisons” were made between low-income minority students in the choice program and much less disadvantaged students in the public schools.

The Harvard/Houston researchers noted that the very structure of the Milwaukee choice program allowed for more appropriate “test” and “control” groups than Witte had identified. Because the Wisconsin legislature had limited the number of students who could participate in the choice program, it was over-subscribed. Students were selected for participation at random. That process created two groups of students with similar backgrounds who were then subjected to two different educational programs, one in a public school setting and one in a choice school setting.

During the choice program’s first two years, students in choice schools didn’t perform any better than the comparable public school group. But students who remained in the program for three and four years showed significant improvement in educational achievement when compared to their public school counterparts.

“If similar success could be achieved for all minority students nationwide,” observed the Harvard/ Houston researchers, “it would close the gap between white and minority test scores by at least a third, possibly by more than a half.”

The Harvard/Houston challenge to Witte’s methodology, and the new study’s support for the choice program, may quiet the program’s detractors. The Wisconsin teachers union, for example, relied on Witte’s findings when seeking an injunction to stop the city from expanding the program. His results also were used earlier this year to thwart a choice program that would have allowed students in Washington, D.C. to escape that city’s poorly performing government schools.

The Carnegie Foundation for the Advancement of Teaching has claimed that “Milwaukee’s plan has failed to demonstrate that vouchers . . . can spark school improvement.” Albert Shanker, president of the American Federation of Teachers, declared that Milwaukee’s private schools of choice “are not out-performing public schools.”

Noting the use of Witte’s analysis to discredit and thwart choice proposals in other states, lead Harvard/Houston researchers Greene and Peterson concluded that his study was damaging those who most needed help. “Mr. Witte’s study isn’t just bad science,” they wrote in The Wall Street Journal. “It’s actually harmful to the underprivileged children who most need the opportunities vouchers would provide.”

George A. Clowes is managing editor of School Reform News. His email address is [email protected].