This research evaluated the usefulness of 3 approaches for predicting college grades: (a) traditional regression models, (b) high-school-effects models, and (c) hierarchical linear models. Results of ...

This article proposes a latent variable regression four-level hierarchical model (LVR-HM4) that uses a fully Bayesian approach. Using multisite multiple-cohort longitudinal data, for example, annual ...