Does the small class size really matter?

An empirical replication for Switzerland on the relationship between class size and school performance in lower secondary school using ÜGK/COFO/VeCoF data

Abstract

This study examines whether the size of the school class has a causal influence on mathematics performance in Switzerland. For this purpose, data from ÜGK/COFO 2016 (Verification of the Attainment of Basic Competencies), a national school performance study on the basic skills of students in the last grade of compulsory schooling in mathematics, are used. Firstly, bivariate analyses suggest that larger rather than smaller school classes are associated with higher test scores in mathematics. Using multivariate estimation methods that are suitable for uncovering causal effects of school class size, this finding is validated while controlling competing influences. In Switzerland, at least for lower secondary schools, there is no need to reduce class sizes with the aim of improving academic performance among students.