The Asymmetric Effects of High Achiever Peers: Experimental Evidence from Ecuador

Abstract

We study the impact on cognitive and non-cognitive skills of exposure high achievers, using a unique experiment from Ecuador. Within each school, students were randomly assigned to classrooms in every grade between Kindergarten and 6th grade. Therefore, the proportion of peers who are high achievers varies across grades and classrooms because of their (random) peer composition. We find that exposure to high achievers reduces test scores in Math, Language, and Executive Function. The results also suggest that male students only affect their male peers, while female students affect both female and male peers. Additionally, we find that the effects of high achievers are concentrated on those in the top quintile of the test score distribution from the previous year. Finally, as with other school inputs, Kindergarten peers have stronger effects but fade-out over time. Contemporary exogenous changes in peer composition do not appear to affect non-cognitive skills.