There is an ongoing nationwide debate on the issue of merit pay for teachers. Many national policy makers and media pundits have essentially said, "How can we improve the quality of education if we don't reward the best and brightest teachers with more money?" While I am never against giving more money to teachers, this idea will have a detrimental effect on learning. There is no way to determine who should get bonus money without negatively affecting students, and that makes it wrongheaded.
There are two main reasons why for the good of children this idea must be rejected. The first reason is based in my books, including Discipline with Dignity and Meeting Students Where They Live, that demonstrate clearly how rewards stop learning. Not only that, but there is a plethora of research, gathered from Alfred Denning to Alfie Kohn and others, demonstrating that bonus money reduces positive outcomes in the workplace. To use this kind of bonus for teachers is hypocritical, bad role modeling, ignoring what has been documented extensively about rewards and has no place in schools.
The second reason is that, to decide who gets more money, we need a way to measure what good teaching is. This measurement, no matter how it is done, implies that learning is measurable. The theory is that teachers who earn bonuses should be those that increase learning. This belief validates all of the false claims of accountability that currently plague educators and schools.
Let's look at the current criteria that are used to assess teacher value and the problems with each and every one.
The first and most obvious way to measure successful teaching is test scores. What a disaster it would be to increase test pressure by including teacher pay. Because it was test-based, No Child Left Behind only worked for school bus drivers, and Race to the Top faces the same fate for the same reason. Increased use of tests to determine what teacher is rewarded will only produce more teaching to the test, reduce higher level thinking and creativity, and set American education even further back. There are many professions where getting the desired results all the time is impossible, including doctors, lawyers and of course, teachers. Surgeons are not judged by mortality rates; heart surgeons have different rates than plastic surgeons and doctors of aids patients. Lawyers are not judged by verdicts; public defenders have higher guilty verdicts than expensive ones, because, among other things, different clientele is a major factor. What the patient or client brings to the table has much more to do with the outcome than the professional who serves them. The same is true in teaching. Tests include the results of those students who for one reason or another cannot or will not learn regardless of how excellent their teacher is.
The second most obvious way to evaluate teacher performance is through administrator observation. As any teacher or administrator knows, there is little time in the administrator's schedule for more than two or three visits a year per teacher, and even less in a big school. This is hardly enough to make an accurate assessment. In addition, principals and other administrators are influenced by the number of hostile parent calls, other teachers' gossip, trips to the office by students and personal relationships. None of these are effective measures of good teaching. For example, while I generally dislike sending students to the office as a punishment, poor teachers may keep troubled students in the classroom if they knew their pay might be determined by referrals.
The State of Texas, under the influence of Ross Perot, tried measuring teaching performance with a pen and pencil test. As might be predicted, it was a total failure. Most teachers know it's not what you teach but how you teach that makes for excellence. Knowledge of pedagogy or content is a poor predictor of teaching if the teacher does not know how to translate that knowledge into behavior.
Some colleges have tried both formally and informally through student initiatives to determine the best professors through student evaluations. These evaluations have even been published online and in rare instances have had a small influence in tenure decisions. The result: easier grading, lower expectations, teachers who become friends with students, and in general crossing the line between being a professional and personal friend. Being friendly with students is a good thing when it's a sign of caring, but not pandering. There are still lines that professionals can't cross to get good evaluations.
Peer review holds some promise but eventually affects collegiality. Teachers are reluctant to share great lessons for fear that their students will say, "We did that already." Every school I have ever worked in, including college, has been infiltrated by politics, groups who dislike each other, and sets of friends and enemies. There is too much potential for politics to be a factor in determining excellence rather than student learning. And it doesn't resolve the first issue of reward failure.
The only way left that I know to find teachers worthy of bonus pay is bringing in outsiders to do evaluations. The cost alone makes this impractical when the money is better spent on school programs. Outsiders would have limited observation time and suffer from all of the obstacles connected with the other methods described above.
Clearly, money spent on teacher excellence is worth it, if we could only identify who these teachers are without making learning worse in the process. I remember visiting a Japanese school years ago. I asked how they assess teacher performance. A gentle administrator answered me very politely. "We don't have to because we only hire the best teachers in the first place." That's a goal worth pursuing. In the meantime, let's give all teachers more money. No one deserves it more.