Tuesday, May 08, 2007

Group selection and merit pay at the CSU

Wilson's perspective on group selection illustrates why the continual pressure by the administration of the California State University (CSU) system to institute a "merit-pay" plan is wrong-headed. (Click here for all my posts about Evolution for Everyone, which I liked a lot.) Such a plan may select for "the best professors" — if best professors can even be defined — but it is not likely to produce "the best" University.

Although I'm not recommending this approach, a group selection perspective might suggest the following. For any academic discipline, e.g., Computer Science, compare Computer Science departments across different campuses and "reward" those that did better against some specified metrics. Eventually, the successful departments would take over more of the Computer Science education work in the CSU system as the unsuccessful departments were allowed to wither and die. This would lead to concentrating disciplines at specific campuses rather than spreading them across the state, but it might produce departments that scored higher on given metrics. Although I'm not urging this approach, it would be interesting to see what sort of results it produced.

Of course, one might also apply this approach to the campuses with the system. Reward the successful campuses and penalize the unsuccessful ones. Eventually the successful ones will take over more of the work of the system. The unsuccessful ones will be closed.

There are (at least) two problems with this approach.
  1. We want (by public policy or political need) to have campuses distributed across the state.
  2. There may be an optimum size for organizations. If they are too successful they will grow past their optimal size. By that time, however, the other ones may have been so weakened that they cannot compete. So one will wind up with unsuccessful oversize organizations.

No comments: