The effect of gender on mathematics attitudes among elementary school students: a multiple indicators, multiple causes (MIMIC) modelling

Abstract

This study examined the effect of gender on students' attitudes towards Mathematics. 1039 elementary students from a Gulf state responded to the Math and Me survey (Adelson and McCoach, 2011) which has two constructs (enjoyment of mathematics and mathematical self-perceptions). Data were analysed using structural equation modelling (SEM), in particular, confirmatory factor analysis (CFA), multigroup CFA and multiple indicators, multiple causes (MIMIC) modelling. Multigroup CFA was used to evaluate the measurement invariance of the Math and Me survey across male and female students. Our results indicated no support for the invariance of both measurement and structural models. Using MIMIC modelling, there was a statistical significant difference for gender in both constructs of the questionnaire, with female students scoring higher than male students in both constructs (enjoyment of mathematics and mathematical self-perceptions). This study has employed a special case of SEM that allows for a dichotomous group of comparison (e.g., gender) to be made using cause indicators as exogenous variables.