The link between class size and academic achievement: An econometric study

Abstract

Education is a pivotal sector for economic and social development. Although Moroccan school has made progress in this sector, the education sector still suffers from dysfunctions represented by low internal and external performance. Among these dysfunctions is the average number of students in a class which is larger in comparison with other countries. Comparative studies are not unanimous on the nature of the link between the number of students in a class and academic performance. This study aims to show the limitations of modeling by standard linear regression and proposes to model this relationship by introducing the instrumental variable to resolve the endogeneity of the variable "Class Size". The data are from the provincial education department in EL KALAA province of SRAGHNA for the 2018-2019 school year. According to the main results, the second method shows higher positive impacts than the first method. Indeed, a reduction of one student per class allows for increasing the success rate by 0,4 % instead of a 0,3% increase if we use simple linear regression.