Another Round in the Public-Private School Debate

Dallasblog.com | Caroline Walker:

“A study released today by the Center on Education Policy concludes that low income students who attend urban public high schools generally do just as well as private school students with similar backgrounds. ... The trouble with relying on conclusions from studies like this is that they have to torture the data to get there. Once you take raw statistics and feed them through the grinder of 'control functions,' 'weighted variables,' and other compensatory 'co-efficients,' you come out with an exercise in hypotheticals, if you ask me.”

Ed Cognoski responds:

Caroline Walker might as well say, "I'm clueless about mathematics, I don't know apples from oranges, but if a study doesn't support school vouchers, I'm ag'in it."

In fact, any comparison between public and private school results needs to control for other differences, such as gender, race/ethnicity, prior achievement, and socio economic status, parental involvement, etc. Otherwise, what might look like an advantage for, say, private schools, might just be a case of parents of children abandoning public schools taking a more active role in their children's education than the parents of the children left behind. This is pretty basic stuff for anyone with an ounce of training in science and mathematics. Caroline Walker considers the data "tortured". In fact, it's her brain that's tortured.

She then concludes with this non sequiter: "Rest assured that you’ll be hearing a lot about this study by opponents of school choice, who will point to it as conclusive evidence that our public schools are serving at-risk students just fine, thank you." In fact, our public schools are not serving at-risk students just fine. It's just that Caroline Walker's solution, private school vouchers, has not been shown to be an effective solution. Let's focus on finding solutions that do work.

P.S. Dallas Blog continues to block me from posting comments on its site. I guess it wants to focus on its pre-conceived notions and not discuss their merits.