Peut-on améliorer l’apprentissage des mathématiques au primaire avec des activités basées sur l'estimation numérique et le sens du nombre?

Abstract

This communication presents research which evaluates the effects of mathematical progression in CP based on the contributions of developmental psychology, neuroscience and mathematics teaching. More particularly, this progression emphasizes the importance of numerical estimation in mathematical learning (Vilette, 2016) and the systematic familiarization of students with the system of analog representation and approximate processing of numbers and calculation (Dehaene, 1997). Six experimental classes from the ordinary sector (N=133) implemented this progression in its entirety. Two control groups were formed. The first control group (T1) is made up of 4 ordinary classes (N=90) who partially experienced the same progression without specific training in number sense and numerical estimation. The second control group (T2) is made up of 10 ordinary classes (N=193) which implemented a classic progression (i.e., the one that teachers usually use). An assessment of mathematical performance is carried out at the beginning and end of the school year in the three groups. The results of the analyzes of variance (ANOVA) highlight: 1) a significant improvement in arithmetic performance in the experimental classes compared to the control classes T1 and T2; 2) also a more marked improvement in mathematical performance in the T1 control classes compared to the T2 control classes. Overall, the results of this research highlight the interest in mobilizing the analogical and approximate representation system of numbers during mathematical learning at school. Repeated mapping of this analog system to symbolic numbers and arithmetic signs develops students' "number sense" and allows for "sensible" calculations (for example, a subtraction in N cannot increase the largest operand) . The results will be discussed regarding the transfer of the progression to a larger scale and their educational implications.