Hierarchical modeling, also known as multilevel modeling or random coefficients modeling, is a useful technique for dealing with nested data, or data that are grouped into a hierarchical structure (Draper, 1995; De Leeuw & Kreft, 1995). This technique is used, for example, when analyzing individual-level data that are collected within groups. Hierarchical modeling is a method that has appeared in numerous articles in this Journal, and the Editor believed it would be worthwhile to summarize some pertinent aspects of this approach to inform the general readership about its uses and limitations. In hierarchical data, general units of observation are nested within successively higher levels. For example, students may be nested within classes, then schools, then districts, and finally states; clinical trial participants may be nested within hospitals or treatment centers; or time points may be nested within an individual subject (De Leeuw & Kreft, 1995).