Learning and Peer Effects

Abstract

Peer effects are possibly very important for educational performance but hard to identify. This paper confirms the existence of peer effects in a learning process with data from an experiment. The experimental approach circumvents key econometric problems which greatly restrict the analysis of educational peer effects with administrative or survey data. The experimental setting offers some insight into the mechanisms of peer interaction. The results show that prospective cooperation has a motivational effect. There is no evidence with respect to an optimal group composition. The benefit from the pair treatment is largely independent of the characteristics of the partner.