Giving teachers something to lose may work better in raising student performance than awarding bonuses later, according to a working paper by a group of major academic figures.
In the study, some teachers — the “loss” group — were given $4,000 bonuses at the beginning of the school year and told they would have to return some or all of it if their students did not reach performance benchmarks. If their students showed above-average performance at the end of the year, the teacher received an extra $4,000. Other teachers — the “gain” group — were given traditional incentives: a maximum of $8,000 at the end of the year if students did well.
Results showed that those in the first category, called “loss aversion,” pulled student scores up higher than those in the latter. These students made large, statistically significant gains in math test scores — the equivalent of reducing class size by a third, according to the researchers. However, “gain” teachers did not make a significant impact in boosting scores.
The latter, a traditional “merit pay” system that rewards good teachers, has caught on as a popular method of improving test scores in the U.S. Yet the researchers explain that the jury is out on whether or not these incentives work; field experiments in the U.S. have shown “small, if not negative, treatment effects.”
The study is headed by economics heavyweights Steven Levitt, University of Chicago professor and co-author of Freakonomics, and Harvard University professor and MacArthur “Genius Grant” winner Roland Fryer, along with two other professors. The researchers conducted the experiment during the 2010-2011 school year, studying 150 teachers in nine schools in Chicago Heights, Ill.
Framing the incentive as a potential loss works better than as a potential reward, the results indicate. In other words, we’re more afraid of losing $4,000 than motivated to gain the same amount in a bonus.
But such a system could produce detrimental consequences, the researchers acknowledged. Teachers might try to discourage weaker students from taking tests or cheat to induce positive scores. Nonetheless, the study shows that “loss aversion” may have noticeable positive effects, comparable to making the class a third smaller.
Which do you think works better: taking money away from teachers if students don’t improve, or awarding them bonuses after students get good scores?