Good or bad? Short- versus long-term effects of multigrading on child achievement

This paper studies the effect of multigrading-mixing children of different ages in the same classroom-on students' short- versus long-term academic achievement in Italy. We cope with the endogeneity of multigrading (and class size) through an instrumental variable identification strategy based on a law that disciplines class composition. By relying on longitudinal data that follow a cohort of Italian students over their compulsory school career, we show that multigrading has a positive short-term effect on achievements. This effect fades away over time to become negative in the long run if students spend several years in a multigrade class. The analysis of mechanisms points to the fundamental role of teachers and suggests that no negative long-term effect arises when multigrade classes are taught by more experienced and motivated teachers. These results reconcile contrasting findings in the literature based on cross-sectional data and a short-term focus.