Multilevel Modeling: (Quantitative Applications in the Social Sciences)

Multilevel Modeling is a concise, practical guide to building models for multilevel and longitudinal data. Author Douglas A. Luke begins by providing a rationale for multilevel models; outlines the basic approach to estimating and evaluating a two-level model; discusses the major extensions to mixed-effects models; and provides advice for where to go for instruction in more advanced techniques. Rich with examples, the Second Edition expands coverage of longitudinal methods, diagnostic procedures, models of counts (Poisson), power analysis, cross-classified models, and adds a new section added on presenting modeling results. A website for the book includes the data and the statistical code (both R and Stata) used for all of the presented analyses.

Table of Contents:
Series Editor′s Introduction About the Author Preface 1. The Need for Multilevel Modeling Background and Rationale Theoretical Reasons for Multilevel Models Statistical Reasons for Multilevel Models Scope of Book Online Book Resources 2. Planning a Multilevel Model The Basic Two-Level Multilevel Model The Importance of Random Effects Classifying Multilevel Models 3. Building a Multilevel Model Introduction to Tobacco Voting Data Set Assessing the Need for a Multilevel Model Model-building Strategies Estimation Level-2 Predictors and Cross-Level Interactions Hypothesis Testing 4. Assessing a Multilevel Model Assessing Model Fit and Performance Estimating Posterior Means Centering Power Analysis 5. Extending the Basic Model The Flexibility of the Mixed-Effects Model Generalized Models Three-level Models Cross-classified Models 6. Longitudinal Models Longitudinal Data as Hierarchical: Time Nested Within Person Intra-individual Change Inter-individual Change Alternative Covariance Structures 7. Guidance Recommendations for Presenting Results Useful Resources References

About the Author :
Douglas A. Luke is Professor and Director of the Center for Public Health Systems Science at the Brown School at Washington University in St. Louis. Dr. Luke is a leading researcher in the areas of public health policy, imple- mentation science, and systems science. Most of the work that Dr. Luke di- rects at the Center focuses on the evaluation, dissemination, and implemen- tation of evidence-based public health policies. During the past decade, Dr. Luke has worked on applying systems science methods to important public health problems, especially social network analysis. He has published two systems science review papers in the Annual Review of Public Health, and the first study to employ new statistical network modeling techniques on public health data was published in the American Journal of Public Health in 2010. He was also a member of a National Academy of Sciences panel that produced a recent report, Assessing the use of agent-based models for tobacco regulation, which provided the FDA and other public health scien- tists with guidance on how best to use computational models to inform to- bacco control regulation and policy. Dr. Luke directs the doctoral progam in Public Health Sciences at the Brown School, where he also teaches doctoral courses in multilevel and longitudinal modeling, social network analysis, and philosophy of social science. Dr. Luke received his Ph.D. in clinical and community psychology in 1990 from the University of Illinois at Urbana- Champaign

Review :
With growing statistical software package costs, more researchers are using R than ever before. This book allows researchers to do more when using R. The book offers insights and explanations from which both newcomers and seasoned experts can find benefit. Because of the author’s pedagogically masterful presentation of multi-level modeling, the otherwise challenging journey to this topic now becomes not only smooth but also enjoyable. This is a very well-written and organized book. The author uses practical examples to help the readers understand the reasoning and steps of a complex statistical approach. I have used the first edition of this book in my class, and definitely plan on using the second edition too. This is a book that I would highly recommend to clinical researchers who are interested in learning multilevel modeling. Multilevel Modeling provides a thorough and accessible introduction to multilevel models. Through extensive examples, the author expertly guides the reader through the material addressing interpretation, graphical presentation, and diagnostics along the way. The new second edition is even better than the first. The models presented are closely linked to an extended example that students can readily identify with.