Wednesday, November 17, 2004

Is the next X-Prize the "E"-Prize?

Alex Tabarrok over at Marginal Revolution is intrigued by the way that Ronald Fryer, a junior fellow at Harvard, has employed a reward system of paying students for performance in a study he has undertaken to help improve the grades of inner city school youth in New York City. The students can earn $20 if they show improvement on tests given every three weeks.

While this approach seems to be working, I think the methodology is flawed. Because of the nominal amount given to each student, the reward is more akin to giving out shiny star stickers than any attempt at long-term improvement. What most likely will happen over time is a conditioning effect upon the students where ever greater rewards are needed for each marginal improvement. The next time $20 won't be enough, but $40, $75, $100, etc. will be needed to extract the same level of improvement. Not only will schools not have the resources to finance this, but it will set students up to become addicted to these rewards. Meaning if the cash is taken away, you can expect student performance to fall below the initial achievement level.

One question that Tabarrok fails to ask of Fryer is why the mostly Black and Hispanic students do not feel the need to compete for educational attainment, yet they are fiercely competitive when it comes to extra-curricular activities, especially organized sports? Even though most kids know they have no shot at stardom, athletic competition is anything but a passive activity. One could even say that sports go overboard in their attempt get their athletes to win at all costs. Perhaps it is because the reward system in sports is organized to a particular goal with a sizable reward rather than incremental payouts. Therefore, student-athletes channel their energy into achieving that goal whether it is winning the state championship or medaling at an event. In academics, there is no parallel system of long-term rewards for many kids who cannot envision themselves as Ivy-League material.

An alternate method that rewards long-term achievement over short-term gains is modeled after the Ansari X-Prize, recently won by the Paul Allen backed SpaceShipOne team of Scaled Composites. Like the X-Prize, the "E-Prize," would incentivize students to excel beyond normative achievement because of the prospect of a large reward instead of nominal, reoccuring sums. However, unlike the X-Prize, the E-Prize would give each student who improved their grades a chance to win the ultimate prize given either each semester or at the end of the year.

A way for it to work may be similar to this:

*At the beginning of the semester, in each class, students are told that they may be eligible to win a sizable prize of their choice if they achieve certain measurable goals or test well on the final exam.

*Students are allowed to tell the teacher what kind of reward they would like to get if they win. Some may choose a particular gift, others may just want cash, and a few may request a non-monetary prize--like free class time.

*Arranged in lottery fashion, the chance that a student will win is based upon how well a student performed, with high achieving students getting more chances. For instance, if a student improved only marginally, he would get 1 chance, while a student who improved considerably would receive 5 chances. The weighting would dependent on overall achievement.

*Once the chance receipts are allocated, a lottery will determine the winner of the raffle. The winners will be publicly announced and celebrated.

I think the above system would be an improvement over just paying students a few bucks each month to get their grades up. Moreover, it would clearly motivate the entire class to improve, rather than just the kids with the lowest expectations.