Associations between Approaches to Learning and Academic Achievement Using Correlated Random Effects

• In Event: Longitudinal Studies Poster Session

Abstract

The present study examines the associations between approaches to learning (ATL) and academic achievement (reading, mathematics, and science) from kindergarten to fifth grade using ECLS-K 2011 data. We utilize a linear unobserved effects panel data model called correlated random effects (CRE) to facilitate causality. The CRE model provides unbiased estimated coefficients of ATL when controlling properly for time-varying covariates. Meanwhile, the model controls for all unobserved student-level time-constant effects and shows the coefficients of important, observed, time-constant variables. Results indicate that ATL is significantly associated with academic achievement for all three subjects across the six grades. We also find that time-constant variables, such as socioeconomic status and gender, are important indicators of academic achievement.

Authors

• Bixi Zhang, The City University of New York

• Spyros Konstantopoulos, Michigan State University