Multigrading and child achievement

We study how grouping students of dierent grades into a single class (multigrading)\r\naects children's cognitive achievement. To do so, we build instruments to identify\r\nthe causal eect of multigrading by exploiting an Italian law that controls class size\r\nand grade composition. We focus on seven- and ten-year-old second and fth graders,\r\nrespectively. Results suggest that attendance in multigrade versus single-grade classes\r\nincreases students' performance on standardized tests by 19 percent of a standard\r\ndeviation (24 percent, gross of the class size eect) for second graders, while it has\r\nzero eect for fth graders. The positive impact of multigrading only appears for\r\nchildren sharing their class with peers from higher grades and it is relatively stronger\r\nfor students from disadvantaged backgrounds.