incentives and education

If you’ve been following Roland Fryer, you know that he’s been trying a pay for grades experiment with children in New York City. The question is simple: does paying kids improve performance? As I mentioned yesterday, the preliminary evidence is that children are more likely to participate in the test, but they are not more likely to get better grades. I noted that this does not disprove the link between incentives and performance. People in all kinds of job learn well when their pay depends on it (e.g., med school admissions, law school, armed services).

So what’s the deal with school kids? Before I get to my main point, I need to explain how learning happens:

• Motivation: At first people, people have to want to learn the topic. This may be due to social pressure, love for the subject, or pay. Or even threats.
• Ability: People need some minimal level of ability so they can actually do the task.
• Training: People need to invest a certain amount of time in learning the skill.
• Social support: You need people who can properly coach, correct and motivate you.

Now, about Fryer type experiments, where you pay kids to do well on tests. Does the pay actually address any of the issues? First, let’s assume that the kids have at least the minimal ability needed for the test. Then, the pay has to offset low motivation, the lack of training (bad teaching) and meager social support (or all of the above). With regard to motivation, it’s likely to be a moderate or weak effect for these kids. A few hundred dollars may get you one or two pairs of new shoes. Definitely something, but not a huge life changing pay off, not enough to make people want to invest the time to master challenging stuff like algebra. Not enough to divest from socializing, nor does it create a space at home in which to work. It also doesn’t magically conjure up the teaching you might need to get the most from your work.

But some of you may ask, what about those parents who “incentivize” their kids with money or threats? My argument is that (a) parents are spending a lot time persuading their kids that the pay off for study is huge (“Do well in 8th grade algebra and you get into medical school!”), or (b) parents provide additional training and social support that the student needs, or (c) in a few cases, the parents actually change the kid’s preferences for how much they want to work. Parents who pay kids for grades are just using the pay to change the kid’s entire social context. They pay itself is beside the point.