Feds: Public Schools Outperform Private Schools

Here's one for all you supporters of the Milwaukee Parental Choice Program, who agitate against the Milwaukee Public Schools and for expansion of the voucher program. As it turns out, public schools outperform--or at least perform as well as--private schools on most fourth- and eight-grade measures.

In a Friday newsdump, the Department of Education released a a report (.pdf) that compared the results of fourth- and eight-grade reading an math tests between public and private schools. The report was competed last year, but, because its findings didn't really fit the ideological leanings of the administration, it went through six months of extra review and scrutiny, with all kinds of qualifying labels attached inside the report to minimize the significance of the findings.

Here's the New York Times from this morning:

The study, carrying the imprimatur of the National Center for Education Statistics, part of the Education Department, was contracted to the Educational Testing Service and delivered to the department last year.

It went through a lengthy peer review and includes an extended section of caveats about its limitations and calling such a comparison of public and private schools “of modest utility.”

Its release, on a summer Friday, was made with without a news conference or comment from Education Secretary Margaret Spellings.

Reg Weaver, president of the National Education Association, the union for millions of teachers, said the findings showed that public schools were “doing an outstanding job” and that if the results had been favorable to private schools, “there would have been press conferences and glowing statements about private schools.”

This study is similar, both in results and in how it was treated by the feds, to one I mentioned in passing last January, that found that students at charter schools, as well, typically do worse, or at least not much better than, their counterparts in the public schools.

What's important about this study for those of us in Milwaukee, though, is the way it was done. There is absolutely no question that if you look only at raw scores, private schools do much better than public schools. I won't deny that, and no one on my side should. The reasons for this are widely-known: Private schools tend to attract better students with wealthier parents, and private schools tend to have the ability to say no when someone undesirable (say, a special education student) asks for admission. Public schools have none of those options; schools like mine have to take any student who walks in the door, and we are held accountable based on that student's performance regardless of where that student comes from.

Hence, the report's methodology. From the report itself:

In grades 4 and 8 for both reading and mathematics, students in private schools achieved at higher levels than students in public schools. The average difference in school means ranged from almost 8 points for grade 4 mathematics, to about 18 points for grade 8 reading. The average differences were all statistically significant. Adjusting the comparisons for student characteristics resulted in reductions in all four average differences of approximately 11 to 14 points. Based on adjusted school means, the average for public schools was significantly higher than the average for private schools for grade 4 mathematics, while the average for private schools was significantly higher than the average for public schools for grade 8 reading. The average differences in adjusted school means for both grade 4 reading and grade 8 mathematics were not significantly different from zero.

The researchers did not simply make two bar graphs and see which one was taller. They looked at what happens when similar students go to public and priate schools by accounting for demographic factors. This allows researchers to say that students with a particular demographic profile would do very well in either a private or public school. A private school that, by its admission policies or because of the popluation it serves has a greater percentage of those students, will do better than the public school down the way without such a high saturation of them.

For Milwaukee, the question of whether the MPCP is truly effective in increasing student achievement rests not on whether private schools in general outperform public schools, but whether individual students would achieve better in the voucher school than if they had stayed in MPS. This study, though not about Milwaukee specifically, suggests that generally, there is no benefit to moving to a private school and, in fourth grade, at least, students in private schools might be at a real disadvantage.

I know that I have complained previously about the kind of sampling and demographic matching that this study did; my complaints, then, however, were in a different context. That context was one of "holding voucher schools" accountable, which is not at all what a statistically-sampled look at voucher schools would do. It would only tell us--as this new Deptartment of Education study does--in general whether the voucher schools were doing any better or worse than the public schools. The whole point of accountability--at least, in my point of view, the point of view of the Milwaukee Public Schools, the point of view of the state Department of Public Instruction, and the point of view of the DoE and No Child Left Behind--is that parents and the community can see how well or poorly individual schools do with their students. No sampling study can ever tell a parent whether the voucher school or the public school in her neighborhood is the better place for her child.

What a study like the one reported today can tell us is whether it is a good idea to keep soaking the Milwaukee taxpayers and siphoning money away from the public schools to keep funding an unaccountable shadow system of schools of indeterminate quality.

I have for a long time said no, it's not worth it. This study confirms that.

(Via Maha.)