Multilevel modeling, also known as hierarchical regression, generalizes ordinary regression modeling to allow explicit and flexible compromises between simple and complex models. This article provides an elementary introduction to multilevel modeling as a model-averaging technique. Model averaging provides an alternative to model selection, and it emphasizes the role of prior information in finding good models.

Key terms bayesian statistics; empirical-Bayes estimation; hierarchical regression; random-coefficient regression; ridge regression; risk assessment; stein estimation