Learning with Differing-Ability Peers: Evidence from a Natural Experiment in South Korea

Whether to group students based on their prior academic achievements has been at the center of policy and research debate. This article explores a quasi-experimental setting in South Korea where the “Equalization Policy” replaced ability-tracking in high school students’ allocation. The policy abolished high school entrance exams and began assigning students to high schools without considering students’ prior academic performance, which exposed students to an ability-mixing learning environment. Using a difference-in-difference approach, we show that ability-mixing considerably reduces the number of low-performers in a national college entrance test. At the same time, high-performers are hardly affected by the policy changes. We document that the behavioral changes of low-performing students may drive the main results. In contrast, we find no evidence that grouping mechanisms affect teacher-pupil interaction and teacher quality.