An Analysis of Effects of Small Class Size on Academic Achievement : Focused on the Moderating Effects of Regional Context

This study analyzed the effects of small classes on academic achievement using data from the second (6th grade) and fifth (9th grade) waves of the 2013 Korean Education Longitudinal Study (KELS). To more rigorously estimate the causal effects of small classes, propensity score matching (PSM) was employed, defining small classes as those with 20 students or fewer or 22 students or fewer. In addition, interaction terms with regional size (metropolitan city, large city, medium-sized city, and rural area) were included to examine regional variations in the effects of small classes. The main findings are as follows. First, small classes were most prevalent in rural areas, and students in these classes tended to have lower household income and private education expenditures than the overall average. Second, for 6th grade students, small classes showed no clear overall effect on academic achievement; however, when the threshold was set at 22 students or fewer, interaction analysis revealed positive effects in medium-sized cities and negative effects in large cities. Third, for 9th grade students, the average treatment effect was not statistically significant, though regression models indicated some achievement gains for classes with 22 students or fewer. Interaction analysis also showed positive effects in medium-sized cities and weakly positive effects in rural areas. These findings suggest that the effects of small classes are not uniform but instead vary depending on regional context. Accordingly, policies on class size should move beyond a uniform, student-number-based framework to incorporate regional and educational environmental factors. The results further imply the importance of continued class size reduction and active implementation of small classes to foster tailored learning environments and improve educational quality.