[en] This study examines academic gender differences in variance and at the extreme tails of score distributions in reading, mathematics and science. Ten databases from IEA and PISA were used to analyse such gender differences in an international perspective from 1995 to 2009. Differences in standard deviation, effect size and variance ratios were computed. The main results may be summarised as follows. The gender differences at the extreme tails of the distribution are often more substantial than the gender differences at the mean, which may suggest the need to rethink education policies for low-achieving boys in reading and for high-achieving girls in mathematics and science. The “greater male variability hypothesis” is confirmed, although our results suggest that this may depend on the test content, the study design and the computed statistics.
Disciplines :
Education & instruction
Author, co-author :
Baye, Ariane ; Université de Liège - ULiège > Département d'éducation et formation > Analyse des systèmes et pratiques d'enseignement
Lafontaine, Dominique ; Université de Liège - ULiège > Département d'éducation et formation > Analyse des systèmes et pratiques d'enseignement
Monseur, Christian ; Université de Liège - ULiège > Département d'éducation et formation > Psychométrie et édumétrie
Language :
English
Title :
Gender differences in Variability and Extreme Scores
Alternative titles :
[en] Différences de genre : variance et scores aux extrémités des distributions
Publication date :
15 September 2011
Event name :
ECER - The European Conference on Educational Research 2011