Washington, D.C., students who won vouchers to attend private schools under a federally funded program were more likely overall to graduate from high school than students who applied for a voucher but did not win one in a lottery, according to a new report on research into the DC Opportunity Scholarship Program (OSP).
The study also looked at student achievement, but the findings were less conclusive. Overall, standardized test scores in both reading and mathematics were not statistically significantly different for students who won vouchers and those who did not. The program did appear to boost achievement for some students, however. Voucher winners in certain subgroups – girls, students with higher academic performance before the program, and those who did not come from schools deemed in need of improvement under No Child Left Behind – had significantly higher reading scores. No statistically significant differences were found in math.
The report is the sixth and final annual evaluation of the program conducted by a team of researchers led by scholars at the University of Arkansas. The OSP was created by Congress in 2004 to give low-income students a chance “to attend higher performing schools” by providing vouchers worth up to $7,500 to cover tuition, fees, and transportation costs. If too many students applied, voucher recipients were to be chosen by lottery, with higher priority given to students from “needs improvement” schools.
The lotteries provided the research teams with randomly assigned treatment and control groups, allowing them to use rigorous experimental methods to measure the effects of vouchers. The evaluators looked at effects of the program over at least four years on both students who were offered vouchers and those who accepted the offer and actually attended a private school. The final report uses data from the 2008-09 school year. Congress voted last year to end the program but allowed students already enrolled with vouchers to continue until they graduate.
Graduation rates were higher overall than control groups both for students who were offered vouchers and those who actually used the awards to attend private schools, with the impact being greater for the latter group. The rates were also higher for some subgroups – girls, students from schools in need of improvement, and those with higher achievement levels before the program. The report notes that graduation data came from parental reports, not school records.
The evaluators also measured parents’ and students’ assessments of their school’s safety and their overall satisfaction with the school. Parents of children who were offered vouchers rated schools more positively than parents whose children lost the lotteries, while students in the two groups reported gave their schools generally comparable ratings.
The research team also tracked which students who were offered vouchers used them to attend private schools and investigated why other students did not use them or attended private schools only sporadically. Some families reported that they declined the vouchers because their chosen school did not have space or did not participate in the program. Others said they did not find special services they needed or that their student was accepted at a desired charter school. Some families left the program because they left Washington or gained enough income to exceed the program’s participation criteria.
Read the full report here and the executive summary here.
FOR MORE INFORMATION
Learn about more voucher research associated with the NCSC.
Read materials on the 2009 evaluation, which was discussed at the NCSC conference in October 2009.
These NCSC projects study several aspects of voucher programs.
Competitive Effects of Vouchers (Florida)
Advanced Analyses of Randomized School Voucher Experiments
Impact of Choice and Competition in Milwaukee