National study: Charter schools a mixed bag

by David Safier
Stanford University just put out a nationwide study of charter schools. The overall conclusion is that students in charter schools do slightly worse than the same students would do in traditional public schools. But the results are so mixed that the conclusion doesn't mean much. Individual findings in the study are far more important.

A good thing about this study is that the funders include the U.S. Dept. of Ed, which has been strongly for charters in recent years, and the Walton Family Foundation, the WalMart people who are very, very pro charters. So it should be hard for anyone to say the study has an anti-charter bias. I'm a  charter school supporter, with qualifications, and I didn't detect a bias one way or the other.

No studies should be trusted outright, but this one sets up a pretty good model. It creates a "virtual twin" of charter schools in public schools with the same types of students and the same starting test scores. The comparison of the two groups, using longitudinal data (comparing starting scores with later scores) is about as good an approximation as you can come up with to rate the relative success of schools.

Now, for some of the results. As I said, this isn't clear cut, but it does indicate some specific strengths and weaknesses of charters and the way they're regulated.

• Arizona is among six states with lower student growth in charters than in traditional schools (the others are Florida, Minnesota, New Mexico, Ohio and Texas). Five states showed higher growth in charters (Arkansas, Colorado [Denver], Illinois [Chicago], Louisiana and Missouri).
• Allowing multiple entities to authorize charter schools is correlated with low student growth in charters. Arizona is among those states. Charters here can be approved at the state level as well as local school districts and universities. The study hypothesizes that directors of possibly weaker schools can shop around until they find someone willing to give them a charter.
• Elementary school students tend to do better in charters than in traditional schools, and high school students tend to do worse.
• ELL students and students in poverty do better in charters, but blacks and Hispanics do worse. (Now that's a complex finding!)
• Students tend to do worse the first year they transfer to a charter school, then do better in the next few years.

I could get deeper in the weeds, but that's more than enough.

My conclusions from the study conform with what I already believed. (I love it when that happens! Though to be honest, it puts me on guard, since it's very easy to seek out conclusions that support your preconceived notions.) Charters in general are neither good nor bad. If we did a better job of picking and choosing which schools get charters, then tracking them over the years, they would probably equal and maybe surpass traditional schools in student achievement.

I've always maintained that Arizona's charter laws are too lax, and everyone would benefit from greater regulation. In the past, "regulation" has been a forbidden word for conservative charter school advocates in Arizona (they purposely created laws that made careful oversight virtually impossible), but that may be changing. If so, I hope there will be bipartisan groups working to make AZ charters more accountable.