I write a lot about what I oppose, so as a sort of thought experiment, today I'll try to imagine if there are ways to accomplish reformster goals that I could live happily with. The posts in the series include Imagining Charters, Imagining Teach for America, Imagining National Standards, Imagining Vouchers and Choice, Imagining Teacher Evaluation, and Imagining National Assessments.
This one is easy, because I've actually addressed it directly before. In fact, years ago I started working up this model and would have started at least half-heartedly refining and pitching it, but by the time I was ready to give it a shot, here came No Child Left Behind and the idea that we didn't need a complex or sophisticated measure of teacher quality-- just check the test results.
You can read about my thoughts in greater detail right here. Let me hit the highlights.
First, the evaluation technique has to do three things:
1) Provide clear expectations to the teacher.
2) Provide useful feedback and remediation.
3) Provide the district with clear information on whether they need to retain, retrain or refrain from hiring permanently.
I've always been focused on the first point-- teaching can be crazy-making because nobody ever tells you what, exactly, they want you to do. We've learned under the "We want you to raise test scores and that's it" regime of reformsterism that a bad answer is actually worse than no answer, and no answer can be kind of liberating. But I've watched a lot of teachers over the years blindsided by discovering that administration was not happy with the teacher's job performance because reasons.
The thing is, your community probably has teacher standards of a sort. In the grocery store, on the corner, in the car pool, people have a pretty good shared idea of what "being a good teacher" means. But it's a complicated constellation of qualities, so the first big challenge of a teacher evaluation system is teasing out of your community what exactly it is that they want from their teaching staff.
Once you've done that (which I admit is like saying "once you've built your faster-than-light propulsion device"), you come up with a form, and you put it in the hands of every single stakeholder of the school, which means staff, taxpayers, parents, students, teachers, alumni-- and you start crunching numbers.
And hey-- if your community tells you, "Never mind anything else-- all we want to know about is how the kids do on the standardized test" you can go ahead and use that.
That's how I imagine it working-- clear expectations, clear evaluation.
I also like the Peer Assistance and Review model, which has the virtue of not requiring a gazillion dollars and umpy-zillion human-hours (that requirement is admittedly a drawback to my model, but this is an exercise in imagining).
I also offered an alternative plan, which I still think is the best one:
Hire a really good principal and let him do his job.