Quantile analysis of the mathematics achievement - attitude relationship by gender

Quantile analysis of the mathematics achievement - attitude relationship by gender

In this paper, we report some findings of a study on attitude and mathematics achievement at the senior secondary school level in Botswana. We adopted a new methodology in the analysis of the relationship between attitude, gender bias and achievement in mathematics with gender as a dummy. The methodology, quantile regression (QReg), was used because it unveils the relationship at the quantiles (percentiles). In addition, this method does not require the parametric assumptions used in the Ordinary Least Square Regression (OLS) conditioned on the mean. The findings showed that attitude variable that significantly influenced the mathematics achievement of students in the OLS model was not significant at the 95th percentile in the QReg. These findings implied that at the uppermost tail of mathematics achievements, there were some other variables that better explained mathematics achievements. Furthermore, using this method, gender bias and being female did not influence achievements. Finally, we hope that this study would open up a new research frontier by encouraging researchers to study the mathematics achievement-attitude relationship at the quantiles rather than generalizing the relationship across all levels of students' ability.

File reproduced in accordance with the copyright policy of the publisher/author.