Stephen Dubner's profile of Harvard economist Roland Fryer is really worth reading, though I particularly liked the part about Fryer's idea to pay poor students for doing well in school:

Fryer recently ran a pilot experiment with third graders at P.S. 70 in the Bronx. If a child achieved a certain score on her reading test or improved by a certain percentage, she got a small prize. In some classrooms, every student competed for herself; in others, each kid was assigned to a group of five. Fryer is trying to find out whether the individual or group incentives work better. He suspects the latter – "because no stigma of being the smartest kid applies.'' But the P.S. 70 data was inconclusive.

At a dinner party held by Larry Summers, Fryer met Joel Klein, the chancellor of New York's public schools, and explained his project to him. Klein asked Fryer if he might be interested in expanding his incentive experiment into 15 or so low-achieving schools. At P.S. 70, the rewards had been pizza parties or field trips. This time around, Fryer planned to give cash -- $10 per good test for third graders and $20 for seventh graders. Now it was time to sell the idea to the principals of those 15 schools.

Very interesting! There's nothing morally wrong with rewarding students for high test scores, of course—plenty of middle- and upper-class families do the exact same thing, only in the privacy of their own homes. And "learning for its own sake" is a nice goal and all, but at least in my experience it's mostly an ideal held by those who were pushed to succeed as children and gradually came to enjoy being dorks later on—and, at that point, they pretend they felt that way all along. Besides, test scores are so dire (two, three, four grades behind) in some schools that the moral problems with paying off students become more or less moot.

At the same time, knowing what I know about the fun task of classroom management (admittedly not a whole lot), doling out cash to a few lucky duckies can certainly breed a good deal of resentment. Back in the day, as a "smart kid" at a public middle school, I would have rather done poorly on an exam than earned a high score plus 20 bucks plus the wholly unwanted attention from all the, um, bigger kids that would come with it. So like Fryer, I suspect group incentives would work better too, but I guess we'll have to see how the pilot program does. I'm a bit skeptical—it's not like teachers have never tried to reward high-achieving students before—but it sounds promising enough.