Multilevel Analysis for Grouped and Longitudinal Data

Joop Hox
Professor of Social Science Methodology, Utrecht University

Abstract
In multilevel modeling, the data have a hierarchical structure, with units nested within groups. Classical examples are organizational studies, with individuals nested within organizational groups. Examples relevant to the survey field are data from multistage cluster samples, interviewer research with respondents nested within interviewers (who may be nested within organizations), and longitudinal research with measurement occasions nested within respondents. This course is a basic and non-technical introduction to multilevel analysis. It starts with a description of some examples, and shows why multilevel models are necessary. It then covers the basic theory of two- and three-level models, drawing on an example of respondents nested within interviewers. Next it explains how multilevel models can be applied to analyzing longitudinal data, and why and when this may be an attractive analysis approach. It will end with a brief introduction to software that has been written specifically for fitting multilevel models: HLM and MLwiN. The course assumes reasonable familiarity with analysis of variance and multiple regression analysis, but prior knowledge of multilevel modeling is not assumed.

Instructor
Joop Hox is Professor of Social Science Methodology at the Faculty of Social Sciences of Utrecht University. His research interests are data quality in surveys and analysis models for complex data. His recent research focuses on non-response problems in surveys and interviewer effects. The complex data he works with are often multilevel or clustered data. He is the author of an introductory book on multilevel analysis, Applied Multilevel Analysis, and the more recently published Multilevel Analysis, Techniques and Applications.