Uniting Differences

In the latest entry in an extended back-and-forth between Edwize and The Quick & The Ed on teacher policy, Leo Casey holds forth on the limitations of standardized testing. Some of it is quite thoughtful, particularly in distinguishing the union position--pro-standards, pro-standardized testing in limited contexts--from the absolutist anti-standards & testing people here.

But it seems like a strange response to my original question, which Leo helpfully includes right at the top of his post. Namely:

How good does student assessment information--from standardized tests or any other source--have to be before it would be appropriate for use in determining teacher salaries?

Note the words "or any other source." I think their meaning is clear--I don't think that standardized tests are the only measure of student learning. In fact, I think it would be wrong to base teacher salaries solely on growth in student standardized test scores, or even solely based on any measure or combination of measures of growth in student achievement. (I'd note that the Aspen Commission seems to feel the same way; they recommended that "Student achievement can count for no less than 50 percent of the determination of [Highly Qualified Effective Teacher] status.") Teachers do lots of important things, like help other teachers. All of them should be considered in any high-stakes definition of "merit."

So if it's true, as Leo says, that:

standardized achievement tests can provide useful data to be considered as one piece of evidence, weighed together with performance assessments, classroom performance and teacher observations, in making important decisions on a student’s promotion or graduation

Why don't the same principles hold for making important decisions on a teacher's promotion, or how much they get paid? Why can't we combine standardized test data with information from the other, more complex qualitative assessments that Leo describes, as well as principal evaluations, peer evaluations, etc., and come up with a responsible way to assess teacher effectiveness--and thus determine an element of teacher pay?

The answer, I think, is that such assessments, if done fairly, would reveal something that everyone already knows: there are huge differences in effectiveness among classroom teachers, even teachers with very similar credentials, experience, etc. And that seems "divisive," to use a word often employed by teachers unions opposing merit pay.

The distinction is important--there are many very legitimate technical concerns about merit pay plans, including the limitations of standardized tests. But this isn't one of them. It's a principled objection, one that seems to rest on the conviction that labor unity requires a certain level of uniformity, that making differences in effectiveness clear, obvious, and meaningful would undermine a larger cause. If that's the real argument here--and I think it is--then Leo should acknowledge it and defend it.

Also, I'd note that in conceding that "There is no reason, as least in principle, why a standardized test could not provide reasonably accurate measures of basic reading comprehension, basic computational skills or the ability to recall and use discrete pieces of information," Leo seems to be suggesting that we really could use those test scores appropriately as part of determining the performance of elementary school teachers, since they're more focused on teaching basic, testable skills.

Finally, Leo asserts that "Carey, Finn and Hess know, without doubt, a large number of ‘hard nosed,’ no nonsense politicians and elected officials of the center to right." I won't speak for Finn and Hess, but all the politicians I know and have worked for have been of the left. This isn't just a Republican thing. Ted Kennedy had kind words for the Aspen Commission report that makes Leo so unhappy--I wonder why that is?