Considering the recent weakening of the teacher’s unions (a Very Good Thing) in many states, the issue of how to evaluate teachers is bound to come up. We can count on the whole concept of seniority finally getting punted, and the possibility that teachers will be promoted based on their ability rather than how long they’ve managed to stick to a gig.

Which brings up the question: HOW do you evaluate a teacher. I’ve been told that they should be evaluated on how may of their students pass their classes (really. if not enough students pass, it must mean the teacher is a bad one, and if they all pass, the teacher must be great, right? especially because the teacher assigns the grades. This person must be a public school graduate.) I’ve also heard that it is unfair to evaluate a teacher based on students performance, because there is so much variety in students. Then, there’s the idea of rating teachers based on how the kids do on standardized tests (which will ensure that teachers teach to the test – another way of making sure that what gets taught is completely useless). Lets take a look at these one by one, and see what we can find.

The idea of rating a teacher based on students passing their courses is simply insane. All this does is make sure that the good teachers (those that are willing to fail non-performing students) will end up at the bottom of the pay scale. The teacher that will pass anyone will end up as the best teacher (and that is more or less where we are now, so why bother).

How about basing it on the students performance (or mastery of the course material). this has one critical flaw (and a number of less critical flaws). the critical flaw is that in order to evaluate a teacher on how well the students have mastered the new curriculum, you have to assume that the students are coming INTO the class with the basic skills (prerequisites) for the course. If the teacher has to spend half the year on remedial work, there isn’t any way that (s)he will end up being evaluated as a good teacher. So instead of evaluating on actual course material, why don’t we just measure how much progress a student makes in a given year? The problem here is that if a student doesn’t keep up (due to a bad teacher, crappy student, or any of a slew of possibilities), the student still moves into the next grade, and by the time you’re in third grade, you’ll have kids that cover the spectrum of knowledge and abilities, and by the time you get to high school it will be even worse. How can you teach a high school class where only some of the students can read? Or only a few of the students can write a cohesive paragraph? (of course, that is what we’re doing now….) So the problem with this idea is that you can’t hold the teachers accountable with this model until you can be sure that they are starting out with students that have the prerequisite skills for their class – in other words, until the schools themselves are accountable, and have a mechanism for dealing with students that can’t or won’t perform.

OK, so how about the old standardized tests. If we can create tests that actually measure mastery of useful and relevant material, this might work. Of course, we all know that standardized tests are inherently flawed, and that there are people that are really good at taking standardized exams, and other people that are really bad at it. So at best, this would be a weak measure of students mastery of mostly irrelevant material. Not so good.

So what is the best way to evaluate teachers? How about we take a look at business, and see how they evaluate their people. In some instances, it is simply a measure of counting up points. A successful salesperson will have a lot of sales, a bad one selling the same product won’t have many. Easy. Also not a good model for teachers – we’ve already decided that there aren’t any solid, easy metrics to measure teachers by. So how about a part of industry that is much squishier – management. In general, management is evaluated based on a whole slew of poorly defined skills and tasks that sum up to indicate if a manger is successful or not. How are they evaluated? Their bosses evaluate them. Yep, purely subjective, based on the day to day interaction of the boss, the manager, and the employees. Are there cases where a good manager gets screwed because the boss doesn’t like him? Of course. Just like there are incompetent managers that keep their jobs by blowing the boss (or the equivalent). Not completely fair, but overall it works. So could this model work for teachers? Let the administrator of the school/department evaluate the teachers, determine who gets raises, who doesn’t, and who gets punted. In other words, let the people who understand the situation, and the strengths and weaknesses of the teachers, students, and the school itself decide who is good. Of course, you have to do the same thing with the administration, which means this would probably never actually happen. School boards and administrative units tend to be magnets for small people that want to be In Charge, just like any bureaucracy. So maybe privatization of the schools would be a good path….. But that’s a different discussion.

Bottom line: Until the schools are actually able to hold non-performing students back, expel problem students, and make sure that students only move on to the next grade when they are qualified to do so, there isn’t any real way to evaluate teachers. The best model is to count on the administration, and hope for the best. Either that or scrap the current fiasco that we call public education, and start from scratch.