Do peer effects influence the academic performance of rural students at private migrant schools in China?

This paper examines peer effects on the academic performance of rural migrant students at both the class level and the individual level. The dataset is from a survey of more than 3000 students from all of the 87 migrant schools in Shanghai and Suzhou, China. The two-stage least squares method with an instrumental variable is employed to control for the endogeneity of peer performance variables. We found that peer effects exist among the migrant students at both class and individual levels. A one-point increase in the average of standardized math grade (SMG) of his/her classmates is associated with an increase of 0.5 points in the SMG of an individual (1.01 standard deviation). A one-point increase in a learning companion's SMG can result in an increase of 0.046 points in the student's SMG. The findings reveal that private migrant schools can improve rural migrant students' academic performance by optimizing students' class allocation and building study groups, providing an additional boost to the input-output efficiency of the human capital education of the children of migrant workers in China, but the effects of these measures are limited.