Predictive Relationship Between Math Self-Concept and Math Achievement: An Autoregressive Cross-Lagged Panel Model Approach

• In Event: How Do Students' Self-Concepts of Ability Change Over Time? A Cross-Cultural Analysis

Abstract

The purpose of this study is to investigate the directionality of the predictive relationship between math self-concept and math achievement among school-aged students. Despite the growing body of research in student self-concept and achievement, the temporal order and relative power of prediction between two constructs remain unclear. To close this research gap, we applied an autoregressive cross-lagged panel model to census-based longitudinal achievement and questionnaire data collected in grades 3, 6, and 9 from a student cohort in Ontario, Canada. The results suggest their reciprocal relationship for at least three years with a stronger power of math achievement in predicting math self-concept than that of math self-concept predicting achievement. Theoretical and practical implications of the findings are discussed.

Authors

• Hyunah Kim, Education Quality and Accountability Office

• Kimberly Cho, Education Quality and Accountability Office

• Ashleigh Allen, Education Quality and Accountability Office

• Wei Li, Ontario Ministry of Finance

• Jennifer Hove, Education Quality and Accountability Office

• Laurie McNelles, Education Quality and Accountability Office