Distributional inequality in mathematics achievement: Quantile regression evidence from low-performing secondary schools in Ghana

This study examines how demographic, family, motivational and contextual factors are associated with mathematics achievement across the achievement distribution among students in low-performing public Senior High Schools in Ghana. Using a cross-sectional survey of 725 final-year students, of whom 418 listwise-complete cases were retained for the main ordinary least squares and quantile models, the study estimates quantile regression at the 25th, 50th, and 75th conditional quantiles and interprets the findings through family capital theory and expectancy-value theory. Results show clear heterogeneity across achievement levels. Mathematics self-confidence is a strong positive correlate at the median and upper quantiles, classroom engagement is negatively associated with achievement at the lower and median quantiles, and regional disadvantage is most evident among lowerachieving students in the Middle Belt. The findings show that meanbased estimates can mask important differences across achievement levels and support a distribution-sensitive understanding of mathematics achievement inequality in under-resourced school settings.