Researchers from the University of Arkansas recently studied achievement effects on students participating in the Louisiana Scholarship Program (LSP) between 2012 and 2016. Using a rigorous experimental design, the authors find negative effects for participating students on state assessments, particularly in math.
Some evidence suggests students enrolled in larger schools with higher tuition experienced more favorable outcomes from receiving scholarships. Other papers from the research team found scholarship students were equally likely to attend college as non-scholarship peers, and that children applying for scholarships were more likely to come from less affluent households.
So, what should we make of this new study and how does it fit into the broader research on private education choice?
For education choice programs to succeed, policies should be designed to encourage participation from quality private schools.
The LSP happens to be one of the nation’s most heavily-regulated programs. It is the only program in which private schools cannot maintain their own admissions criteria and are required to administer state tests to LSP students. ExcelinEd has written previously about the LSP’s regulatory burden and the effects this has on private school participation.
Remember, too, that private choice programs typically provide significantly less funding per student than is spent in traditional public schools. That is the case in Louisiana.
When it comes to testing, ExcelinEd advocates that private schools participating in publicly-funded choice programs be permitted to choose whether to administer state tests or nationally-normed tests. This approach makes achievement information available to parents while allowing schools to maintain autonomy. And it acknowledges that perhaps what makes private schools unique from public schools is what makes them appealing to parents—and could be a contributing factor as to why many private schools educate children successfully.
Most evaluators of schools that participate in choice programs try to understand a program’s impact on math and reading scores. However, the link between a program’s effect on short-term achievement with its effect on long-term attainment is unclear. For example, an American Enterprise Institute meta-analysis presented a weak relationship between a choice program’s effect on test scores with its effect on later-life outcomes.
At ExcelinEd, we are supportive of test-based accountability as a short-term indicator of student progress. It’s unreasonable to wait for a cohort of students to graduate from college before assessing an educational intervention. But there are indicators that shed light on whether schools are working. These include achievement in other subjects, post-secondary attainment, civic engagement, and parental satisfaction, among others.
Consider, then, some other data from Louisiana. A recent survey found more than 92 percent of parents participating in the LSP approve of the school their child attends, and more than 96 percent are pleased with their child’s academic performance. A 2017 study found that 82 percent of student transfers resulting from the LSP reduced racial stratification in the students’ former public schools. The University of Arkansas researchers also made this observation:
“In spite of scoring lower than their control group peers, especially in math, students who participated in the LSP were accepted to and enrolled in college at a rate that was statistically similar to the students who lost their LSP placement lottery. One possible explanation for those divergent results is that private schools in the LSP are teaching students content and skills that help them get accepted to college but are not measured by the state test.”
Other studies have shown private school choice has significant positive impacts on non-cognitive outcomes. If this is true in Louisiana, it could explain why LSP participants are enrolling in college at similar rates. However, further study is needed to understand if these students persist in college, despite indications that they are not as proficient as their peers in core subjects.
Research such as this from the Urban Institute on Florida’s Tax Credit Program could serve as a useful guide for measuring the effects of choice on college enrollment and achievement.
ExcelinEd recommends that student achievement in choice programs is evaluated annually by independent third parties. Our philosophy with respect to research is the more, the better.
Fortuitously, a few weeks after publication of the Louisiana research, a new gold-standard evaluation was released by the research arm of the U.S. Department of Education on Washington DC’s Opportunity Scholarship Program. The authors found that, after three years, students using these vouchers to attend private schools had similar norm-referenced test scores as their public school peers, despite earlier studies that showed lower test scores for voucher students. One explanation is that transitioning to a voucher program could temporarily depress test scores.
Both studies on Louisiana and Washington, DC should be considered within the broader existing literature on private education choice.
To this end, a recent publication by EdChoice aggregates more than 140 empirical studies on participating student test scores, attainment, parental satisfaction, public school students’ test scores, civic values, racial integration, and fiscal effects. It seems clear that private choice has a strong record of producing positive results on a variety of indicators, in both the short- and long-run.
