Jumping to conclusions
June 10, 2009
If you're a New Yorker making $500 for nine months' work, you've got a bum gig--if you aren't in elementary or middle school, that is. Indeed, some New York City fourth and seventh graders can earn up to $250 and $500, respectively, for good performance on a collection of 10 assessments. The Sparks program, Harvard economist Roland Fryer's tri-city and 59-school experiment, is a privately funded two-year pilot that tests the relationship between monetary rewards and academic achievement. According to a New York Post analysis, it seems to be "working," at least in the Big Apple; the majority of participating schools in that city saw math and reading scores jump nearly 40 percentage points from last year to this. (Of course, scores were up across the metropolis, but by much smaller margins.) Well, duh! It seems reasonable that the promise of cash rewards for scoring well on a set of tests would be an effective motivator for students to take the same set of tests seriously--and for scores to improve. But whether learning has increased, too, remains to be seen. The problem here, though, is the Post's (and not Fryer's) "analysis." To prove the correlation between monetary rewards and real achievement gains, pay-for-grades programs must be tested against a reliable independent assessment--one whose scores don't carry cash rewards. Otherwise it's possible that Fryer's experiment is demonstrating that, under normal circumstances, lots of kids don't give their all on state tests--an important and distressing finding in its own right.
"Learn-&-Earn Plan Pays Off," by Kelly Magee and Yoav Gonen, The New York Post, June 8, 2009