Hierarchical Linear Modeling

Researchers across various disciplines often deal with data that have multilevel, hierarchical, or nested structure. Methodologists from various disciplines refer to the statistical methods for analyzing hierarchically structured data differently. Researchers in behavioral and social sciences refer to them as hierarchical linear modeling (HLM) or multilevel modeling. In statistics, the process is referred as covariance components modeling. In biometrics, it is referred as random-effects modeling and mixed-effects modeling, and in econometrics, it is referred to as random-coefficient regression modeling.

These hierarchical models explicitly model lower and higher levels in the hierarchy by taking into consideration the interdependence of individuals within the groups. For example, in a two-level HLM analysis, the emphasis is on how to model the effects of explanatory variables (predictors) at one level on the relationships occurring at another level. These multilevel structured data present analytical challenges that cannot be handled by traditional linear regression methods, because there is a regression model for each level.