Sangmyung Ha

Indiana University Bloomington · Economics · 2026–27 job market

Fields

Econometrics

Job market paper

Estimating Heterogeneous Teacher Effects with Grouped Three-Way Interactive Fixed Effects

We develop a grouped three-way interactive fixed-effects model for estimating matching effects in multidimensional panel data. We establish aggregate consistency of the global least-squares estimator, derive stochastic expansions and limiting distributions for the slope estimator and aligned factor estimates, and characterize three modespecific incidental-parameter bias terms. Under the maintained identification and strong factor conditions, slope and aggregate factor consistency permit one dimension to remain fixed when the other two diverge. An application to matched student–teacher achievement data reveals that teacher–group matching accounts for approximately 27% of fitted relative teacher value added variation. Using adaptive LASSO, we identify six CLASS, MQI, or teacher survey items associated with the teacher-level effect, and one teacher survey item associated with the matching effect. We also find that achievementgroup heterogeneity and three-way separability structure in teacher effects can improve out-of-sample forecasts of teacher-level effects in the empirical Bayes framework.