Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We examine the effect of relative evaluation on test performance by implementing a classroom-level experiment in which students are financially incentivized either individually or in a tournament. Linking the experimental data with student-level administrative data allows us to study two aspects of competitive environments: tournament structure and one’s perceived position in the ability distribution. At least in our setting, we find only limited evidence that effort responses to competition are sensitive to tournament size or prize structure. However, in contrast to previous studies that examine effort responses to exogenously assigned competition, we find a large negative competition effect for students who believe they are relatively low in the ability distribution, and no competition effect for those who believe they are relatively high ability. Using additional treatments, we further show that the divergence between our results and past results is likely driven by task type and not by differences in selection into participation between lab and classroom environments.