My colleague Elena Silva is pretty bent about this Baltimore idea to pay kids for passing tests. New York City is experimenting with this general idea, too, and a few other places. Like Elena, my first gut reaction is also pretty opposed.
But, how do we know it’s such a bad idea? Where is the evidence? Actually, it’s pretty early and mixed on this sort of idea overall. Meanwhile, Elena cites lots of other things that would plausibly get students engaged in the context of “we already know what works.” But I think she ignores just how great the discount rate is for kids, long time horizons are not great motivators, and also minimizes the thin evidence around student engagement overall. A lot more theories than empiricism there and though I agree with the ideas Elena lays out, we don’t “know” as much as various partisans claim. So, what if this idea could be one more element in the mix? Al Shanker used to point out that kids must have some skin in the game, too, for the system to work.* In other words it can’t all be on teachers.
Doesn’t mean we shouldn’t be leery of this approach, I personally think very leery, just that we should also be humble about what we know and don’t know in this field and consequently be willing to at least give some ideas with some sort of logic model behind them a chance and have our belief systems challenged some. In this case, what’s the absolute worst that can happen?
All that said, the kinds of ideas I’d like to see tried more on this front are bigger. For instance, rather than payments like what Baltimore is doing, what about marrying concepts like “baby bonds” (seeding conditional savings accounts at birth for low-income children and augmenting them over time) with educational goals. I don’t know if this would work either, but seems one plausible way to (a) tie incentives to longer-term goals and (b) make the incentives more meaningful. Also, big potential for public-private partnerships there as well via matching dollars and so forth.
*Paging Rick Kahlenberg! Big debate here, what would Shanker say?
Update: Much more on NYC.
Update II: It’s hard to be ahead of Kahlenberg! He writes (and dissents from the CW here, too):
Since you ask —
During my seance in November, I reported the following:
But Diane Ravitch has a different view on Al Shanker and paying for test scores:
http://www.tcf.org/list.asp?type=EV&pubid=206 (scroll down to Ravitch)
4 Replies to “Suspending Belief…”
“In fact, Bloomberg is on to something. The cell-phone bribe and the pay-for-test-scores scheme, which provides up to $500 a year for seventh-graders who do well on 10 exams, are the brainchildren of black economist Roland G. Fryer. An assistant professor at Harvard who also serves as the New York City Education Department’s chief equality officer, Fryer himself grew up in difficult circumstances (his mother left when he was very young, and his father spent time in prison for sexual assault). But Fryer succeeded, and he became interested in finding out what incentives would motivate more students growing up in like circumstances to do well. His ideas are an intriguing combination of tough and liberal approaches: tough because they take a hard-nosed rather than romantic view of education, and liberal in that the goal is to raise the achievement of low-income kids and foster social mobility.
Fryer’s ideas are closely connected to those of another tough liberal, the late Albert Shanker, who headed the teachers’ unions from 1964 until his death in 1997. Shanker argued that the incentive structure in American public education was terribly biased in favor of well-off kids and against poorer ones. Students who were shooting to attend selective four-year colleges—most of whom were upper-middle class and white—had plenty of incentive to do well academically.” — Rick Kahlenberg, Slate, Nov. 6, 2007
There’s more here:
There is an unseen consequence to student pay for test scores that I think studies of such programs are likely to miss: what happens to motivation when incentives no longer exist down the road. If I understand correctly this is relatively well-studied phenomenon in psychology. I recall a study I learned about in college, where young children were divided into two groups. In the first an adult gave the children crayons and asked them to make drawings. The children proceeded to do so. In the second, the children were told that if they made drawings they would be given candy. The children (much more enthusiastically) proceeded to do so. In the second round of the experiment, the children were again asked to make drawings, but this time, there was no offer of candy for either group. The children who had never been offered candy, proceeded to make another round of drawings. The group who had initially been offered the “reward”, now declined to make drawings. The interpretation the authors of the study made was that the children in the first group had learned that the reason to do the drawings was the external incentive of candy, and when the external incentive was removed, so too was their motivation. Gives me serious pause about pay for test-scores unless we can ensure that such external incentives will continue throughout the rest of children’s academic lives.
from Daniel Willingham’s article on rewards and motivation:
The key factor to keep in mind is that rewards only decrease motivation for tasks that students initially like. If the task is dull, motivation might drop back down to its original level once the rewards stop, but it will not drop below its original level.