Rationality in the teacher evaluation debate

One of the touchiest subjects in all of education right now is the evaluation of teachers. It's easy to understand why. Teachers have been the victims of curious instruments in the past - does anyone have fond remembrance of LaTIP/LaTEP in the late 1980s? That disastrous instrument scored teachers by how many "benchmarks" they were able to hit during observations scored by evaluators with checklists. Teachers aren't trained monkeys, but LaTIP/LaTEP treated them as if they were. It failed miserably.

Yet there are few who could honestly say that the current evaluation method, which depends on subjective judgments by administrators, is the best possible way to judge a teacher's effectiveness.

So how to proceed? Education Week reporter Susan H. Fuhrman has made an effort to cut through the fog and smoke and come up with a rational article about teacher evaluation.

Fuhrman starts with the observation that "value added" is the flavor of the moment, and likely will play a role in future instruments promoted at both the national and state levels. And the idea that teachers can be evaluated in part by measuring the progress of students in their classrooms is attractive.

The idea is not to base the evaluation on the outcome of a single test, but to determine how much student improvement is made. Thus a teacher whose students begin two years behind but end up just half a year behind could be considered a star.

But there are pitfalls, as the EdWeek article points out. That means we should tread carefully. And the political process, which can be very clumsy, has a way of highlighting flaws and codifying poorly thought-out processes along ideological fault lines.

We've been there before, and it didn't work. The right solution is to let research lead the way, with allowances for failure and missteps, and permission to change when necessary.