Filler: To get better teachers, evaluate the principals

Two weeks ago Gov. Andrew M. Cuomo proposed yet another way to test teacher competence that, while it can't hurt, also won't help much. The idea is a rigorous professional test for new teachers, something like the bar exam for lawyers.

Understand, I'm not against it. Math teachers need to know what X equals (it's 11). We don't want geography teachers who think that of the 57 states, Texas is the largest, when the correct answer is Canada. And a sex education teacher who doesn't know where babies come from (also Canada) does more harm than good.

But the idea comes on the heels of another well-intended but problematic effort to improve schools, and when we talk about making sure teachers are competent and dedicated, there are a few simple truths we need to consider.

Evaluating teachers is easy, as is coming up with a system that weeds out the bad ones and rewards the good ones.

The teachers unions don't want a system that weeds out the bad ones and rewards the good ones.

Lots of smart people do not want to be teachers.

The teacher evaluation system being implemented here in New York has a huge flaw. The 40 percent of evaluations to be based on student progress on standardized tests will have a margin of error of about 25 percent, plus or minus, because classes are not large enough to provide statistical reliability. So when a teacher rates out at 70 percent on improving student performance on the tests, it will mean that educator deserves something between a 45 and a 95. We'll then conclude, decisively, that this teacher's performance falls in the range between horrible and heavenly.

For evaluating principals, though, schools generally are large enough to provide a meaningful measure of student improvement on these tests, experts say. Such measures, then, can tell us quite clearly what kind of ships principals run.

So if we punish or reward principals on these results, they would keep and promote the best teachers and ditch the bad ones at the speed of light (1 jillion mph, right? In my day, teachers took their bar exams at an actual bar, and we turned out fine).

They wouldn't play favorites and keep lovable, ineffective teachers, because they'd care more about their own careers than the jobs of even the most delightful suck-ups. They wouldn't fire the talented pains in the neck unless these teachers were so disruptive it actually hurt the school's results, because the principals would be more psyched by big bonuses than they would be aggravated by the pesky but effective employees.

If principals were allowed to fire and reward teachers, that is.

But enacting such a system, and streamlining the process of firing teachers, known as Education Law §3020a because it involves 3,020 steps, is not in the best interests of teachers unions. Union leaders don't earn their pay protecting the best teachers: The best teachers don't need protecting. This system of evaluating principals would make it impossible to protect the cruddy, loopy teachers, and thus remove a main union task.

But I also think such a system would help address the other problem, which is that so many smart people don't want to be teachers. One reason they don't is that the profession comes under so much fire, and it comes under fire because its protectionism makes educators seem untouchable and entitled.

We can keep dreaming up bar exams and performance rubrics, but they're all intended to show something parents, students and principals already know: who the good and bad teachers are. What we need to do is use the performance standards to evaluate the principals, and let the principals deal with the teachers. Do that, and before you know it, New York will have the best schools in the country, and the other 56 states, even Canada, will be playing catch-up.