Here:

Via James Joyner comes news that Texas A&M is going to start giving $10,000 bonuses to professors who get good student evaluations. The article, of course, comes along with a lot of caterwauling about how unreliable student evaluations are as a tool for instructor assessment. I’d be sympathetic except for one thing: I grew up among academics. And I have never since met a class of people so contemptuous of teaching. You’d think they were being asked to chew mud. In part, that’s a structure of the rewards system. Teaching takes a lot of time but doesn’t play a big role in tenure or promotion. Which is why Texas’s maneuver makes sense. The question is not whether the evaluations will offer a perfect analysis of teaching skill but whether attaching $10,000 to them creates some real incentive for professors to take teaching seriously.

To solve this problem I’d rather separate teaching from research — have people who are the best at one do one, and the people who are the best at the other do the other. Also, giving profs $10,000 checks for getting students to like them could lead to grade inflation. But it’s an interesting point.