The (Il)logic of Teacher Evaluation

Cross-posted from Education Week

Yes, illogic is a word, which my dictionary defines as the “quality of lacking logic.”  Let me count the ways.

Not so long ago, the critics derided school administrators for failing to evaluate their teachers.  The research, they said, made it abundantly plain that no school-related variable is as powerful an explainer of student performance as the quality of a student’s teacher.  Yet, they said, teachers almost invariably win high performance ratings.  In their minds, nothing so clearly revealed the overwhelming perfidy of school administrators as their willingness, year after year, to provide so many fine ratings to so many obviously incompetent teachers.  It was that sense of outrage that apparently drove so many of the critics to advocate measures that would force school principals to do what they so obviously should have done, but would not do, to improve the performance of their students.  The logic of that response seemed so clear and compelling to the critics that they did not feel they needed to defend it.

But that is not how I see it.  To the contrary, I see the principals as rational actors responding to the incentives they faced in a rational way.  Consider the situation from the principal’s point of view.  Advancement for principals is not, except in extreme cases, a function of student performance.  Principals are part of system in which advancement is based mainly on loyalty to one’s superiors.  There are good reasons for that.

In public schools generally, parents want schools that are safe, close to home, staffed by people who care about their kids and, at the high school level, home to winning sports teams.  For a great many parents, all these things trump their children’s performance on standardized tests.

For top managers in public school districts, the path to the top is most often barred, not by poor student performance, but by opposition from vocal groups in the community.  That opposition can arise from many quarters for an infinitude of reasons.  If you are an ambitious administrator, you want to develop a reputation for keeping a wide range of constituents happy.  The best way to do that is make sure that the people below you do not do something that will make your constituents unhappy.  The last thing you want is someone who, possessed by their own vision of educational excellence, will do something that angers some group whose support you need.  And that makes loyalty a prime virtue as you look for people in whom to place your trust.

Now add to that the fact that it is, in most school districts, extremely expensive and very time consuming to remove a teacher for incompetence through formal proceedings.  That is why, if you are a conscientious principal, the last thing in the world you want to do is start a formal procedure to get rid of an incompetent teacher.  No, no, no.  That will cause trouble and everything you know about the system is that it is allergic to trouble.  But you have another route.  You can prevail upon the central office to reassign your bad apple to another school or to the central office, which, of course, sets off the famous ‘dance of the lemons,’ but is a perfectly rational response to the situation you face.

But, in that circumstance, it makes no sense to give your bad apple an honest rating.  All that will do is make that teacher furious and, at the same time, make him unmarketable to another school, maybe even the central office.  Other teachers will be worried that they might be next, morale will be shot, and you, the principal, will be isolated.  Most important, you will be less likely to achieve your objective, which is to get rid of the offending teacher.  Bide your time, play the game, continue the good ratings and things will work out.

As long as this incentive structure prevails, it is a form of insanity to expect principals to behave differently, precisely because their behavior is highly rational.  The rage for teacher evaluation will give way to the usual compliance behavior.  There will be change on the surface, but the old game will continue as before.

What would have to change to make the system work better?  First, the incentives operating on the superintendents and their direct reports would have to change dramatically.  They would have to believe that their own advancement and tenure in office depended, first and foremost, on substantial improvements in student performance.  That is not the case now, not even close, because improved student performance would have to trump schools’ athletic records, ideologically-based attacks from parents on value grounds, who wins jobs in the district and who gets the contracts for school bus services and so on.  The politics of local control are at the root of much of the incentive structure that school principals and superintendents face.

Let’s get real, for a change.  Teacher evaluation is not going to get us a supply of great teachers.  And an oversupply of great teachers is the only thing that is going to fill our schools with great teachers.  The logic of test-based teacher evaluation is deeply faulty, a strategic dead end.