Politics, Testing, and CT

In an interesting post about symbolic politics Sherman Dorn chides me for having the right attitude about the NAACP’s intervention in the CT No Child case but then ignoring that the state wants to keep test quality high something that he notes I’ve written about previously and that my colleague Tom Toch wrote about recently in this ES report.

In theory Sherman’s right that there might be a contradiction there but I don’t think that CT’s case is going to hold up on the specifics. First, they’ve been quite slippery about what exactly they’ve been spending money on and the case is not as strong on the money as some are claiming. Ultimately, though, that’s for the courts to decide. Second, there is a rhetorical trap here where multiple-choice testing becomes axiomatically equated with low-quality testing. Leave aside basic issues like curricular alignment, there are high-and low-quality multiple choice tests. Conversely, just because a test has open-ended or constructed response questions doesn’t guarantee that it is a high quality test for the purposes that it is intended for or overall. In fact, there are some complicated issues here in terms of test reliability that have legal implications for high-stakes testing with non-multiple choice tests. As a matter of policy and practice, for my money annual testing can and in fact should be done with a combination of different kinds of tests to minimize the time students spend on testing but maximize the data educators and policymakers need.

This all matters in the case of CT because the argument that CT cannot do high quality tests without more money from the feds is far from a slam dunk. It’s entirely possible that the state can do just fine on the quality front but that they want to do different kinds of testing. That’s a legitimate desire and though I’ve argued for a long time that we should be spending more on assessment from the federal level than we do now, desire is likely not controlling in this case rather a minimum floor, laid out in the statute itself, is. In other words, Secretary Spellings may have been overly dismissive of CT’s concerns but she’s probably right on the substance and will probably win in the end (though perhaps not as quickly as in the Michigan case because the issues are more complex). The disconnect here speaks to the need for a centrist coalition that supports testing but wants high quality testing. The still not-too-hidden testing good v. testing bad debate hinders progress on this front and of course the weird politics about the federal role which transcend left and right only further complicate matters. In the end though the CT case was/is mostly about politics not substance so again go read Sherman’s apt post! Also, while you’re there check out this interesting post about Sherman’s time in the pokey.

Leave a Reply

Your email address will not be published.