A DIDACTIC AND PSYCHOSOCIAL PERSPECTIVE ON NATIONAL MATHEMATICS ASSESSMENTS AT THE BEGINNING OF ELEMENTARY SCHOOL IN FRANCE

Prepublication

Abstract

In France, assessments carried out on all pupils at the beginning of each year of primary school reveal performance gaps between girls and boys in mathematics, starting as early as the first year of primary school. To better understand this result, we studied the validity of the inferences generated from the statistical model used, focusing on the two aspects of validity considered in mathematics education: epistemological and psycho-didactic validity. Our research is based on two types of study. An a priori analysis of the test tasks, based in a reference praxeology, provides evidence of validity from the knowledge perspective. This is completed, in relation to the test exercise set for Grade 2 pupils (CE1), which in 2023 showed the largest performance gap between boys and girls, by an analysis of the responses of 260 pupils and interviews conducted with around thirty pupils. We first demonstrate that the content of the test, as presented, does not allow for a precise fulfilment of the assessment objectives; we show that the format of the exercise analysed and the instructions given to pupils strongly influence their responses and lead to interpretations that may be incorrect, particularly when referring to the diagnostic report produced for each pupil after this assessment. Our analyses also provide explanations for the differences in scores between boys and girls and help to put into perspective the recurring findings of boys’ superiority in mathematics at this school level.