Examining the Individual Differences in Mathematics Achievement: An Application of Quantile Regression

• In Event: Poster Session 5
  In Poster Session: Research and Mathematics Education SIG Posters

Abstract

This study investigated the role of student characteristics in mathematics achievement using 2011 Trends in International Mathematics and Science Study (TIMSS) 8th grade American sample. Both ordinary least squares (OLS) regression and quantile regression were conducted to examine the predictability of exogenous variables (i.e., gender) on mathematics achievement. The results showed that quantile regression yielded a more comprehensive evaluation at various points of mathematics achievement than OLS regression. The quantile regressions results showed that age and SES have a stronger impact on the math scores of lower achieving students, while gender, race, and home language have a stronger impact on the math scores of students who scored higher in the distribution.

Authors

• Jing Yuan

• Letao Sun, University of Kentucky