More than three-fourths of Indiana charter schools perform worse or at least no better than local public schools, according to a study released last month by Stanford University researchers.
That’s one way of spinning the results of the study from the Center for Research on Education Outcomes, or CREDO – although it’s not the finding that was played up in most news coverage. Nevertheless, there it is:
// In math, 42 percent of charter schools performed worse than traditional public schools, 23 percent performed better, and 35 percent were not statistically different.
// In English, 18 percent of charter schools performed better than traditional public schools, 8 percent performed worse and 74 percent were not statistically different.
Most media reports focused on the study’s finding that charter students, on average, scored better on tests than they would have if they attended their local public schools; or its conclusion that Indiana charter schools are among the best in the nation at improving their students’ test scores.
An Indianapolis Star editorial even argued the study should put an end to the “tiresome debates” over the effectiveness of charter schools. It won’t, and it shouldn’t.
A few issues that deserve continued exploration:
// Are charter-school students representative of the general population? The Star highlighted the surprising finding that Indiana charter schools serve significantly more poor and minority students than traditional public schools. CREDO reaches this conclusion (see Table 1, page 12) by comparing charter students with students in “feeder schools,” which include all public schools where at least one student attends a charter school. CREDO Director Macke Raymond, the author of the study, very patiently tried to explain to me, by phone, why this approach makes sense from an economics standpoint.
But there are charter schools all across Indiana, so the profile of feeder schools whose students attend charters looks similar to the state as a whole. Most Indiana charter schools, however, are in high-poverty urban areas like Indianapolis, Fort Wayne, South Bend and the Calumet region. So a non-economist might expect charter students to look like public-school students in those cities.
In fact, if you compare urban charter schools and public schools, it appears the charters enroll about the same percentage of poor students, more black students but fewer Hispanic and special-needs students. The profiles are similar, but they’re also different in interesting ways.
// CREDO finds that, on average, charter students did “significantly” better on ISTEP-Plus exams and End of Course Assessments than they would have done in non-charter schools. But “significantly better” doesn’t mean “a lot better.” It means enough better that it’s not a statistical error. The latest study follows up on a 2011 CREDO evaluation, which found similar test-score gains for Indiana charter students. Jonathan Plucker, who was then the director of the Center for Evaluation and Education Policy at Indiana University, told School Matters the gap was nothing to crow about. “From my perspective, they found no difference between the two types of schools,” he said.
// For some people, the take-home message from the CREDO study was that charter students are as much as two months ahead of their non-charter peers. Stanford led with that point in its news release. But the study itself says translating test-score differences into months of learning gains “is challenging and can be done only imprecisely,” and it warns that such results “should be interpreted cautiously.”
// The oft-made argument that charter schools may be “creaming” the best students doesn’t require thinking that charters exclude poor, minority or special-needs students, or that they dump or “counsel out” students who don’t make the grade. It’s at least arguable that the rigmarole involved in getting and keeping a child in a charter school – selecting the best fit, entering a lottery, arranging transportation – could deter all but the most motivated and savvy parents. And that those parents are more likely to provide their children with the support they need to do well in school.
// Comparing the charter and public students who qualify for free or reduced-price lunches or who receive special education services doesn’t really tell us too much. But that’s the only measure of poverty that’s available for studies like this one. As Chicago education researchers have shown, it can be much more challenging to improve schools with large numbers of truly disadvantaged students. Also, some studies have found that charter schools tend to enroll special-needs students with less serious disabilities, such as speech and language impairment. It would be interesting to know if that’s the case in Indiana.
CREDO has done these studies for 23 states, and Indiana charters are fifth most effective, according to center director Raymond. CREDO is best known for a 2009 national study of charter schools that’s often cited by both supporters and critics of charters. As Matthew Di Carlo of the Shanker Institute noted, the differences that CREDO finds between charter and non-charter schools tend to be small. You can’t really draw broad conclusions that “charters are good” or “charters are bad.”
Note: Apologies for posting this so late. The CREDO website was down over the holidays, so there was no way to have a live link to the study. I didn’t expect that from Stanford, but maybe even computers need a holiday.