What Does New Research Say About Louisiana's Voucher Program?
Note: Joshua Cowen, an Associate Professor of Education Policy in the College of Education at Michigan State University, joins us this week as a guest blogger. You can follow him on Twitter at @joshcowenMSU.
Last month, a pretty big earthquake shook the world of school choice. A team of economists studying the Louisiana Scholarship Program released the first evidence of substantially negative school voucher impacts on student achievement. The rigor of the study was beyond dispute, so instead of the usual back-and-forth debating the quality of the work or the motives of the researchers themselves, both sides of the choice debate have jumped ahead to wondering how to explain what amounts to a shocking bit of news. We'll learn more in a couple of weeks when some new work from a separate team will come out with an extra year of data and outcomes other than test scores. But it's clear that, at least initially, voucher users did really poorly in Louisiana.
The simplest explanation, and the one I'm going to push here today, is that on their own, these results shouldn't be so surprising. We've seen positive and negative effects for charter schools in different places for years so it's probably time we saw something like this for vouchers. What's stunning, though, is just how large the negative Louisiana results appear to be. And it's their size that makes the results susceptible to over-interpretation and over-explanation.
The most interesting debate about all of this has taken place among school choice supporters. There are several theories being tossed around, but by far the one that's drawn most of the attention is the idea that overregulation of private schools is to blame. The idea is that we should expect bad things to happen when states or school districts tell private schools what to teach and how to measure results, especially if those mandates discourage high quality, innovative private schools from joining voucher programs. This is all an intellectual possibility for sure, although as Doug Harris, my colleague down in New Orleans, has pointed out, it rests on "mostly theory" with little in the way of evidence to support it. In fact, the only study out there to look directly at the effects of regulation on voucher programs showed positive effects on test scores. I happen to have co-authored that paper.
So I want to put on my "Captain Obvious" hat for a moment and simply use these ridiculously bad Louisiana voucher results to remind everyone that whatever part an explanation like regulation does or doesn't play in the story, it remains true that some private schools are just not very good at educating kids. There are also some awful charter schools. And, as we know, some pretty miserably run traditional public schools. There are also some amazing examples of truly effective schools in each of those sectors. And so, collectively, we're going to keep seeing some studies show positive results of school choice, some negative, and some no differences between sectors at all. That kind of variation is built into the argument for school choice in the first place. If the whole point of choice is to allow parents to find schools across sectors that meet their kids' needs, it follows that schools are expected to be very different.
And we're not just talking about differences between good and bad schools here, either. There are also differences between what good schools are good at doing in the first place. This is why we're starting to see a number of studies that show no positive impacts of charter or voucher schools on test scores, for example, but big positive effects on other measures like high school graduation or college attendance. On a more basic level, what it means to be a charter school, or a participating school in a voucher program, varies substantially by state. So it's not even clear that when we see certain results—out of Louisiana or any other single context—we can infer a whole lot about similar programs nationally. Just remember that next time you read that charters or vouchers are all bad, or all good. The truth is right there in between.
That may sound like I'm trying to have it both ways. What I'm really trying to do is argue that as much as we need to keep studying these programs, the evidence we obtain from that effort really can't be used to make overly general arguments about which schools—charters, vouchers, traditional public—are best for kids. There are big differences between individual schools, and the more school choice we have, the more individual school differences are going to matter a lot more than the sector in which each school is located. And that's why the only hard and fast lesson we can reliably draw from any new study, regardless of what new information it offers, is the importance of obtaining that information in the first place. Private schools may not release test scores on their own, so studies like the one out of Louisiana—and the testing requirements that made it possible—are especially valuable. If you're a choice opponent, this is the only way to keep these individual schools honest. And if you're a choice supporter, well, there's a flip side to that coin: information is vital to any competitive market. So, now we know: voucher schools in Louisiana just aren't that great. But that knowledge tells parents and policymakers in that state a lot more than it tells the rest of us about the pros and cons of expanding school choice.