Monday, April 12, 2010

Quasi-meta-sorta Control Study for Incentives

I like Claus's analysis of Fryer's initial "pay-for-grades" study. I think you sort of have to set aside your normal squeamishness about the policy itself, and recognize that it is a necessary, and possibly benign, exploration in the larger development of "business model" reform.

First off, look at it this way, if instead of all the crap the Gates, Broad, Walton, etc. have directly and indirectly (and via ED) foisted on the schools in this city and my neighborhood, they'd simply been passing out money to kids for high test scores, good attendance, whatever, I think we'd all be better off.

More realistically and concretely, the entire premise of many of their strategies (ostensibly) is that they've got a better handle on what motivates teachers, students, etc., and by adjusting rewards, punishments, etc., we'll get the results we want. I don't really agree with the premise, but I also don't agree that their models of actor motivation are correct, and that seems easiest to test if you just look at "what if we give people X dollars to do Y" instead of or in addition to more traditionally structured and indirect reforms and incentive structures.

If you aren't highly attuned to this discourse, you won't realize how significant it is that, as Claus highlights, Fryer suggests that rewards for "process" may be more effective than ones that only look at "outcomes," or test scores. This is heresy in many circles today.

2 comments:

gnuosphere said...

If it's true that rewards improve results related to relatively thoughtless tasks (e.g. showing up to class) and that they impede results requiring higher-order thinking, it would seem we have a good point to start from.

Leroy's Mom said...

Well, this "solution" to the staffing problem at my school hits on both this post, AND the donut hole piece you did, as existing teachers in target schools are not eligible (along with any teacher (like kinder and first grade) that doesn't have a test score to point to.
Alice Mercer
http://mizmercer.edublogs.org