An Introduction to Multilevel Modeling for Social and Personality Psychology

Multilevel modeling is a technique that has numerous potential applications for social and personality psychology. To help realize this potential, this article provides an introduction to multilevel modeling with an emphasis on some of its applications in social and personality psychology. This introduction includes a description of multilevel modeling, a rationale for this technique, and a discussion of applications of multilevel modeling in social and personality psychological research. Some of the subtleties of setting up multilevel analyses and interpreting results are presented, and software options are discussed. Once you know that hierarchies exist, you see them everywhere. (Kreft and de Leeuw, 1998, 1) Whether by design or nature, research in personality and social psychology and related disciplines such as organizational behavior increasingly involves what are often referred to as multilevel data. Sometimes, such data sets are referred to as ‘nested’ or ‘hierarchically nested’ because observations (also referred to as units of analysis) at one level of analysis are nested within observations at another level. For example, in a study of classrooms or work groups, individuals are considered to be nested within groups. Similarly, in diary-style studies, observations (e.g., diary entries) are nested within persons. What is particularly important for present purposes is that when you have multilevel data, you need to analyze them using techniques that take into account this nesting. As discussed below (and in numerous places, including Nezlek, 2001), the results of analyses of multilevel data that do not take into account the multilevel nature of the data may (or perhaps will) be inaccurate. This article is intended to acquaint readers with the basics of multilevel modeling. For researchers, it is intended to provide a basis for further study. I think that a lack of understanding of how to think in terms of hierarchies and a lack of understanding of how to analyze such data inhibits researchers from applying the ‘multilevel perspective’ to their work. For those simply interested in understanding what multilevel