The nation鈥檚 largest tax-credit scholarship program doesn鈥檛 appear to be hurting the academics of students who remain in public schools, a new study shows.
Those students who remained in public schools during the expansion of Florida鈥檚 tax-credit-funded private school vouchers program鈥攖he nation鈥檚 largest, with more than 100,000 students participating鈥� actually saw improvements in their reading and math test scores, and were on average suspended less often and absent less often, concludes the study, which was.
And the most likely explanation for those gains, the study says, appears to confirm one of the arguments made by private school choice boosters: The competitive pressure that comes from students having a lot of school choices led the public schools to improve their offerings.
In all, the study is among the first to examine the long-term effects of a choice program. Most existing studies look only a few years after such a program has been put into place, though they have largely found small benefits.
鈥淚t鈥檚 definitely notable and important to say you continue to see consistent, modest improvement as the program expands in size,鈥� said Cassandra Hart, one of three researchers on the study. She is an associate professor at the school of education at the University of California, Davis.
While the study can鈥檛 resolve some of the deeper ideological questions surrounding these types of programs鈥攖he financial pressure of losing students to private schools is real鈥攊t does suggest that some other concerns aren鈥檛 yet warranted.
It鈥檚 long been a concern, for example, that the most motivated families would benefit from such a program鈥攃oncentrating the neediest students behind them in public schools. But that does not appear to be the case in Florida.
That said, the researchers cautioned, it isn鈥檛 clear when Florida鈥檚 expansion of this program could reach a 鈥渢ipping point,鈥�
鈥淭heoretical models would also say that at some point you should see some declines, because the positive achievement effect is outweighed by things like limited resources,鈥� said Krzysztof Karbownik of Emory University, one of Hart鈥檚 partners on the study. 鈥淪o it鈥檚 very much an empirical question: Where is this tipping point?鈥�
Competitive Pressures
Hart and Karbownik teamed up with David N. Figlio of Northwestern University for the study. It鈥檚 based on a sample of more than 1 million Florida students. The researchers merged Florida birth records with test-score data from grades 3-8 from the 2002-03 through the 2016-17 academic years. Through 2011-12, they also looked at each student鈥檚 records of suspensions or absences.
Building on some prior work of Figlio and Hart鈥檚, the researchers developed five measures of school competition: How many private schools in the same grade levels each school was near, the number of types of private schools that were nearby, how many churches and religious schools were nearby, and the proportion of students served in private schools, among others. Then they analyzed student outcomes for their relationship to these competitive pressures..
The findings showed a statistically significant relationship for nearly all the measures studied, with the students attending schools in areas with more competition seeing greater increases in reading and math test scores. And, in one of the very first pieces of evidence on the effects of these programs beyond academics, the researchers found reductions in both suspensions and absences, suggesting that student behavior also improved.
Importantly, these results weren鈥檛 uniform among all students and schools: Students from low-income families benefited more as schools responded to the tax-credit programs. Reductions in suspensions were most dramatic for Hispanic students.
Here鈥檚 one of the best ways to picture these effects. This diagram from the study shows the effects for all students; for those who at one point received free- or reduced-price lunches, a proxy for poverty; and for those who never qualified for that program. Don鈥檛 worry too much about the finer-grained details and look at the overall patterns: They are largely positive and growing for academic effects, while suspensions and absences fall. And note that those patterns are stronger for economically disadvantaged students than for their more-advantaged peers.
Generally speaking, these are not enormous gains. The researchers characterized them as modest. But it鈥檚 noteworthy that they are there, and that they seem to be increasing as the program expanded.
Importantly, the researchers tried to rule out alternative explanations for the findings. One is that the composition of students left in the school differed drastically from what it looked like before the program started. But the researchers found that those students who stayed were not, on average, higher- or lower-performing students. Second, they theorized, the smaller class sizes could account for the findings, since much research shows a connection between smaller classes and learning. But the researchers found that class-size benefits only contributed a small portion of the overall effects.
The most likely explanation, they said, are instructional changes within the schools themselves.
Putting Findings in Context
One of the reasons the study matters is because so much of the research literature on private school choice programs has been anything but clear-cut. (.)
By and large, the track record on the performance of students who receive vouchers is murky. The most recent studies consistently find negative effects, not positive ones, on the test scores of students who used them to attend other schools. (Longer-term impacts, on high school graduation and college matriculation, are more mixed and seem to depend on program context, Chalkbeat noted.)
And what about the effects on public school students? Most of those studies have focused on the first few years after a private school choice program鈥檚 introduction. Those have mainly found small positive effects. Florida is a different story. The program is more than 15 years old and now serves more than 108,000 students, making it the single largest tax-credit program in the nation. It has also morphed considerably over this time period. Initially restricted to families with incomes just slightly above the federal poverty line, it now serves families with incomes up to 260 percent of the poverty line鈥攐r more than two-and-a-half times.
鈥淚t does a better job than any study I鈥檝e seen to date of asking how the existence of a large voucher program affects students who don鈥檛 use it,鈥� said Martin West, a professor of education at the Harvard Graduate School of Education. (He did not contribute to the study.)
That鈥檚 an important addition in part because of the conventional wisdom on these programs, West added. 鈥淢any people assume that voucher programs have negative consequences for public school systems, and the students who make use of them. Even treating that as an empirical question, rather than a given, is important.鈥�
The study doesn鈥檛 directly address value-laden disputes over voucher programs. Those longstanding arguments include whether the money that funds them should be considered public. (Many argue that it should, since they鈥檙e funded through tax credits, but Florida courts have ruled that its program does not involve public money.)
More recently, Florida鈥檚 program has come under scrutiny because some of the schools families have selected contain explicitly anti-LGBTQ policies, and because there is far less accountability for results than in the state鈥檚 public schools.
But Can It Be Replicated?
The researchers did sound this note of caution: Don鈥檛 rush to the conclusion that any old private school choice program will have these same kinds of effects. The Florida program is unique in a number of ways, particularly its duration and its changing features over time.
And the study wasn鈥檛 designed to peer inside the 鈥渂lack box鈥� of schools to determine precisely what happened; it tells us the 鈥渨hat鈥� but not much about the 鈥渉ow.鈥� We can鈥檛 determine, for example, if these are genuine improvements in learning or, in contrast, whether the public schools doubled down on test-prep or some other strategy to boost scores. The improvement on non-academic outcomes, though, suggests overall climate did improve for students.
It鈥檚 clear, too, that individual families have very different experiences with these programs. (My colleague Arianna Prothero wrote about the struggles of one family, who used two of the state鈥檚 other private school choice programs, although not the one this study is based on.)
The researchers themselves said they鈥檇 welcome some sociological, qualitative studies to help tease out some of those details.
鈥淚 would love to see the results of this being district offices and researchers getting a better sense of what鈥檚 happening in schools that鈥檚 driving these effects,鈥� Hart said. 鈥淚 think that would be extremely valuable for district offices to know鈥攁nd not just in the context of what鈥檚 happening in voucher schools, but in the search of best practices that can help all students.鈥�
And finally, the study doesn鈥檛 say anything about the financial impacts that these public schools may have grappled with even as they apparently served students better. It鈥檚 reasonable to assume that these, too, would differ in Florida from other states. (You probably have more cushioning in your budget as a large countywide district in Florida than as a tiny district in Indiana.)