Examining a causal effect of private tutoring in Korea: does private tutoring damage students' self-regulated learning?

Abstract

Previous findings on the relationship between private tutoring and self-regulated learning are inconsistent. In this study, we assumed that the inconsistency of these findings was due to a combination of methodological issues, differing definitions of self-regulated learning and failure to account for the amount of time spent in private tutoring. The purpose of the study is to examine a causal effect of private tutoring on students' self-regulated learning based on a social-cognitive perspective via the marginal mean weight through stratification method (MMW-S). We especially investigated if the effect of private tutoring varies depending on the quantity of private tutoring. Based on the hours of private tutoring per week, 3750 Korean middle school students were classified into six subgroups, and the population average causal effect of multivalued private tutoring was estimated with MMW-S. This finding indicated that private tutoring had a null effect on self-regulated learning across different levels of private tutoring, and no evidence was found that excessive private tutoring for academic subjects had a positive or negative impact on self-regulated learning. The findings imply that private tutoring itself neither progresses nor damages students' self-regulated learning abilities.