Teachers’ Graduate Education and Experience Affecting Elementary Student Achievement: A Crossed Random Effects Growth Model

• In Event: National Databases, Science and Math Student Achievements, and a Methodological Assist

Abstract

Associations between teacher qualifications and student achievement have been studied extensively, yet little consensus has been reached on what qualifications matter. The variety of statistical modeling approaches and data sources that have been employed make consensus difficult, since estimates are sensitive to model specification and the usually observational data sources are prone to selectivity biases. This study has two objectives: addressing whether teacher characteristics matter and determining their sensitivity to model specification. Using large-scale observational data from Florida, we use a crossed random effects growth modeling approach to estimate teacher covariate effects on mathematics achievement growth and use school-mean centered covariates to control for selectivity biases. Preliminary results show significant teacher effects for both graduate degrees and years of experience.

Authors

• Walter L. Leite, University of Florida

• Robert John Sandbach, University of Florida

• Gregory J. Palardy, University of California - Riverside