March 20, 2007

Challenging the challenge index

Sara Mead has half of the right argument with Jay Mathews on his challenge index. But because she picks on King High School in Tampa (more on that in a minute), I can tell the other half of the story.

Summary thus far: For a number of years, Jay Mathews (and Newsweek) has published a list of high schools with a high ratio of AP tests taken to graduates. Mathews sees that as a reasonable approximation of the challenge that schools offer students. Previously, he has acknowledged that AP exams are only one type of challenge that schools can offer and that "tests taken" is an imperfect measure of AP participation, let alone success. Today, he argues that Mead (and Andy Rotherham, aka Eduwonk) are falsely impugning the legitimate work done in these schools, even if other measures suggest serious problems:

But in the real world that means C. Leon King High School in Tampa does not belong on the best schools list because of its high dropout rate and low average test scores, even though Newsweek ranked it 73rd in the country in AP and IB test participation last year.

Mathews points out that despite its being a magnet school, the AP and IB tests are not overwhelmingly in the magnet program at King. And one can make an argument that having a magnet (or other specialty) program raises the general capacity of King to offer AP courses and other challenging opportunities. So far, so interesting. Mead responds with a "well, let's look at the other data" argument.

I'm not sure anyone has pointed out the perverse incentive in treating schools like King as a single entity: for every kid in the non-magnet side who drops out, the school looks a little better. Horrific dropout rates can only boost your challenge index so far if the school is a comprehensive high school. On the other hand, a selective-enrollment magnet program has a critical mass of students guaranteed to take plenty of AP classes, and then you have the resources to offer AP classes outside the magnet program as well. So a school that's split into a magnet program and a regular program with lots of poor children who drop out? That should be the gold mine for the challenge index, at least hypothetically.

Anyone want to test my hypothesis?

Note: I don't know what the graduation measures would look like at King if there were decent ones. (I'm skeptical of Florida's official measure.) I know a few teachers at the school and have visited, but I have little global information other than to know from my students that it essentially is two schools on a single campus (the magnet program and the rest of the school). Listen to this article
Posted in Education policy on March 20, 2007 9:57 PM |