« More on TFA: It's Elementary | Main | Are Teachers Jumping the Charter School Ship? »

Dueling Studies: Privately Managed Schools in Philly


The Philadelphia school system began a long-running — and much-studied — experiment with privately managed schools in 2002. That's when frustrated state officials took over the struggling district, parceling out some of the worst-performing schools to for-profit and nonprofit providers.

Studies continue to differ, though, on how well that little venture has worked out. The newest such study, posted online today at the American Journal of Education, makes the case that, at least in the middle grades, the privately managed schools have not kept up, academically, with the rest of the system's schools.

Researcher Vaughan Byrnes of Johns Hopkins University analyzed 10 years of reading and math test scores for 88 district schools enrolling students in the middle grades.

While students at most of the schools improved their test scores after the takeover, the schools run by outside managers did so at at a slower pace, according to the study.

But Harvard University researcher Paul Peterson, in a 2007 study, claimed just the opposite. He studied two successive waves of 5th graders and found that the students in the privately run schools outperformed their peers in district-run schools.

In a subsequent study, Peterson also gave an achievement edge to for-profit schools over those managed by either the district or by nonprofit organizations. Read this article in Ed Week's archives for more details on that.

So why the difference between the studies? It may be a matter of timing, according to Vaughan. His study, spanning from five years before the takeover to five years after, covers a longer period of time.

Plus, he says, 2001, the starting point for Peterson's study, was an unusual year for the schools that were "on the list" for potential outside management.

"There was a big panic in the district and teachers were starting to transfer out," he says. "It turns out it was a unique year."

You may have to read both articles and judge for yourself.


The very fact that researchers have to break out their statistical magnifying glasses to discern any difference between the effects of the two models suggests that we haven't stumbled across any miracle cures by changing governance. We still have much more fundamental questions to answer about teaching and learning.

Theres a famous study done in a factory years ago that I wonder about when reading this kind of results. It was on something like factory lighting or some such. They turned up the lights, and productivity went up. Then they turned down the lights--and productivity went up. There's a name for this kind of effect--when the simple fact of a change, any change, perhaps couples with people knowing they are being watched/studied, that gives positive results. Maybe someone out there knows it?

Anyway--makes me wonder if there may not be some such effect confounding the results.

Comments are now closed for this post.

Follow This Blog


Most Viewed on Education Week



Recent Comments

  • Jean: Theres a famous study done in a factory years ago read more
  • Claus: The very fact that researchers have to break out their read more