In contradiction of some other recent research, a large new study has found that offering students rewards just before standardized testing can improve test performance dramatically. One important factor in this finding might be the immediate pay-off — students received their rewards right after the test. Another might be in the participants, who were attending low-performing schools.
The study involved 7,000 students in Chicago public schools and school districts in south-suburban Chicago Heights. Older students were given financial rewards, while younger students were offered non-financial rewards such as trophies.
Students took relatively short, standardized diagnostic tests three times a year to determine their grasp of mathematics and English skills. Unusually for this type of research, the students were not told ahead of time of the rewards — the idea was not to see how reward improved study habits, but to assess its direct impact on test performance.
Consistent with other behavioral economics research, the prospect of losing a reward was more motivating than the possibility of receiving a reward — those given money or a trophy to look at while they were tested performed better.
The most important finding was that the rewards only ‘worked’ if they were going to be given immediately after the test. If students were told instead that they would be given the reward sometime later, test performance did not improve.
Follow-up tests showed no negative impact of removing the rewards in successive tests.
Age and type of reward mattered. Elementary school students (who were given nonfinancial rewards) responded more to incentives than high-school students. Younger students have been found to be more responsive to non-monetary rewards than older students. Among high school students, the amount of money involved mattered.
It’s important to note that the students tested had low initial motivation to do well. I would speculate that the timing issue is so critical for these students because distant rewards are meaningless to them. Successful students tend to be more motivated by the prospect of distant rewards (e.g., a good college, a good job).
The finding does demonstrate that a significant factor in a student’s poor performance on tests may simply come from not caring to try.
 . The Behavioralist Goes to School: Leveraging Behavioral Economics to Improve Educational Performance. National Bureau of Economic Research; 2012. Available from: http://www.nber.org/papers/w18165