Do peers affect private tutoring engagement in Korea? Evidence from a quasi-experimental approach

Abstract

This study examines the influence of peers who receive private tutoring on an individual student's engagement in private tutoring. Using data from the Gyeonggi Education Panel Study of 7th and 10th graders in Korea, we leverage quasi-experimental variation generated from the random allocation of students to classes in schools. To address the bidirectionality of peer influence, an instrumental variable strategy is combined with a school fixed effects approach. The findings indicate that peer's private tutoring leads to an increase in student's engagement in private tutoring. The peer effect on private tutoring is more pronounced among girls than boys. More precisely, increasing the proportion of classmates who receive private tutoring by 10% points increases the likelihood that a girl engages in private tutoring by 7.1% points. Policies that leverage peer influence may be a cost-effective way to reduce the personal and social ramifications of excessive consumption of private tutoring in Korea.