School Choice or Best Systems_ What Improves Education_ - Margaret C. Wang [21]
As I learned from Stanford political scientist Terry Moe, vouchers are politically controversial because they tend to split the two major political parties. Two big constituencies of the Democratic Party are African Americans and Latinos, who strongly favor vouchers, and teachers’ unions and other public-sector unions that strongly oppose them. Among Republicans, advocates of the free market support vouchers because they expect competition and choice to enhance efficiency and parental satisfaction, whereas suburban Republicans may oppose vouchers because they do not want low-achieving city children enrolled in their children’s schools, and they may believe they pay high property taxes for what they see as good schools in their neighborhoods.
Because of these political controversies, public voucher programs in the United States exist in only a few cities, and they are generally small in scope. Consequently, their success or failure does not constitute a conclusive test of the education market (or even the voucher idea). Certain programs, however, are sufficiently large and have been studied with sufficient rigor to shed some light on the controversies summarized above and to provide some insight into how schools and parents would react to larger-scale voucher programs. Several voucher programs in other countries, discussed below, have been far more ambitious and also reasonably well researched.
Voucher Effects on Academic Achievement
Although fewer students participate in voucher programs than attend charter schools, the random assignment of students gives researchers access to enough data to compare public and private voucher recipients to students who lost selection lotteries, which results in gold-standard randomized field trials. Eight such studies and three non-random-assignment studies have evaluated the effects of vouchers on academic achievement.
Eight Random-Assignment Studies of Academic Achievement Effects
Led by Paul Peterson, Harvard University’s Program on Education Policy and Governance2 conducted several studies of achievement by students participating for two years in private voucher programs in New York City, Washington, and Dayton. The research team found that African-American students who received scholarships outperformed African-American students who applied for but did not receive scholarships by 4 percentile points in New York, 7 percentile points in Dayton, and 9 percentile points in Washington, DC. No statistically significant effect was found for white students.
Mathematica Policy Research of Washington, DC,3 studied the private voucher program in New York City and found a slightly stronger academic benefit to African-American students receiving scholarships (9.2 percentile points instead of 9.0) than did the Harvard study, though the effect was statistically significant only for one of the three grade levels studied. The Harvard team validly contended that the sample size for each grade level was too small to allow that conclusion to be drawn.
Researchers from Harvard, Mathematica, and the University of Wisconsin4 studied the New York private voucher program again in 2002 and once again found that standardized reading and math test scores for African-American students who used vouchers were 9.2 percentile points higher than those of comparable African-American students who did not use vouchers. The researchers also found that, when asked to assign a grade to their children’s schools, 42 percent of voucher parents gave their school an A while only 10 percent of parents not in the program gave their school an A.
Jay Greene5 studied a private voucher program in Charlotte, North Carolina, and found that scholarship lottery