What impact do voucher programs have on the long-term success of students? For decades, studies of the effect of voucher programs relied on near-term data, such as student test scores. As voucher and tax-credit scholarship programs mature in age, so too are the ways researchers are evaluating them.
Two new studies released by the Urban Institute add more nuance to the debate by examining the effects voucher programs have on students’ college going, persistence and graduation in the cities of Milwaukee, WI and Washington, DC.
The Milwaukee Parental Choice Program (MPCP) was the nation’s first private school choice program and now serves nearly 28,000 students in 126 schools. The program provides students with a voucher with a maximum value of around $7,500 for grades K-8 and just over $8,000 for grades 9-12. That’s compared to more than $14,000 that is spent per-student in Milwaukee Public Schools.
In the first Urban Institute paper, coauthored by Patrick J. Wolf (University of Arkansas), John F. Witte (University of Wisconsin) and Brian Kisida (University of Missouri) studied the MPCP and its effect on college enrollment, persistence and graduation by matching MPCP students to similar Milwaukee Public School students for 11 years starting in 2006. While not a “gold standard” random assignment study, the authors were able to match students who lived in the same neighborhood and had similar demographic characteristics and test scores.
The researchers found that voucher students in the MPCP were, on average, more likely to enroll in college than their matched public school peers. This was particularly true for attendance at four-year colleges: ninth graders who were enrolled in the MPCP in 2006 were six percentage points more likely to enroll in a four-year college than their public school peers. Similar results were found for students who were enrolled in elementary or middle school in 2006, where MPCP students were five percentage points more likely to enroll than their Milwaukee public school peers.
There is, however, a wrinkle. As the authors wrote:
We find that students in the MPCP program have greater educational attainment than the comparison group. MPCP students are more likely to enroll, persist and have more total years in a four-year college than their MPS peers. We do not find evidence that MPCP students are significantly more likely to graduate from college.
So, while the MPCP program seems to lead to more students enrolling and staying in college longer, college graduation rates for MPCP students were statistically indistinguishable from public school peers.
The Washington, DC Opportunity Scholarship Program (OSP) has the unique distinction of being the only K-12 voucher program ever enacted and funded by Congress. In operation since 2004, the OSP has served between 1,000 and 2,000 low-income students in the nation’s capital each year since it was authorized. OSP students receive significantly less funding than their counterparts in district and charter schools. Voucher students receive no more than $8,000 in grades K-8 and $13,000 in grades 9-12, compared to $17,500 in DC charter schools and $30,000 in Washington’s district public schools.
The second paper, authored by the Urban Institute’s Matthew Chingos, studies effect of the DC OSP program on college enrollment. Unlike the Milwaukee study, Chingos used the “gold standard” method of random assignment. This is because the OSP program has a funding cap and more students apply to participate than there are eligible spots, with a random lottery utilized to select participating students.
The study finds no significant difference between the college enrollment of OSP participants and students who lost a lottery to participate in the program. However, Chingos notes two important points.
First, unlike the Milwaukee study (and an earlier Florida study) that compared voucher participants mostly to traditional public school students, the DC study compares the voucher program to an environment where nearly half of all public schools are charter schools. The DC charter school sector has been held in high regard as one of the most improving education sectors in the nation.
Second, while the enrollment trends are indistinguishable, the cost to taxpayers is not. Chingos writes that his study “still suggests a possible productivity benefit of the private sector,” since the voucher amount is somewhere between half to one-third of the what DC spends per students in traditional public and charter schools.
The DC and Milwaukee studies come just a few months after a similar study of Florida’s tax-credit scholarship program, which found that program increased a low-income student’s likelihood of enrolling in a two-year college by 15 percent. However, college graduation in that study was also indistinguishable from similar non-participants.
This series of papers from the Urban Institute add tremendous value to the ongoing debate around private school choice. Overall, these studies show that private school choice programs are producing similar or better results for participating students, at a cost far below the per-student amount in traditional district schools.
And, perhaps more importantly, these studies are advancing our thinking around choice, how to measure its success and what is needed to ensure that students are not only more likely to enroll in college – but able to see their way through to a post-secondary credential.
About the author
Adam Peshek @AdamPeshek
Adam Peshek is Managing Director of Opportunity Policy at ExcelinEd, where he provides strategic support to state leaders interested in developing, adopting, and implementing policies that increase educational options for children. He has provided expert testimony in more than a dozen state legislatures and is a frequent commentator on ESAs, school choice, and education policy across the country. He is also the is the co-editor of the first published volume on ESAs, Education Savings Accounts: The New Frontier in School Choice. Adam currently resides in Atlanta, Georgia and is a Senior Fellow with the Beacon Center of Tennessee.