Multilevel modeling for survey data

Bart Meuleman, University of Leuven

Survey researchers are –more often than not- confronted with nested or clustered data. Nesting arises because social reality is layered (individuals are educated within classes, are employed within companies, pass their free-time within organizations, live within countries,…) or is caused by the research design (e.g. two-sample designs, allocation of respondents to interviewers). Multilevel models are becoming an increasingly popular technique to analyze clustered survey data.

This course provides an introduction to the basics of multilevel modeling. Participants are made familiar with the conceptual logic as well as the statistical background of multilevel models. Concretely, we discuss two-level regression models with a random intercept and random slopes. Strong emphasis is put on the substantive interpretation of multilevel parameters. Advantages as well as drawbacks of the multilevel model are discussed.

About the instructor

Bart Meuleman is Assistant Professor at the Centre for Sociological Research, University of Leuven (Belgium), where he teaches research methodology. His main research interests involve cross-cultural comparisons of attitude and value patterns, such as welfare attitudes, ethnocentrism, religiosity, and basic human values. In his work he frequently applies multilevel models. He was invited to teach multilevel courses at various institutes throughout Europe. Recent publications appeared in Social Science Research, Journal of Cross-Cultural Psychology, Journal of European Social Policy, and Social Psychiatry and Psychiatric Epidemiology.