Multilevel and Longitudinal Modeling with IBM SPSS

Multilevel modeling has become a mainstream data analysis tool over the past decade, now figuring prominently in a range of social, health, and behavioral science disciplines. This text demonstrates how to use the multilevel- and longitudinal-modeling techniques available in IBM SPSS (Version 26). Adopting a workbook format, the text walks readers through setting up, running, and interpreting a variety of different types of multilevel and longitudinal models using the linear mixed-effects model (MIXED and GENLINMIXED) platforms in SPSS.

The text offers numerous examples of cross-sectional, repeated measures, and cross-classified data structures with outcome variables primarily measured as interval/ratio. It also offers several selected models with categorical outcomes. Extended examples in each chapter illustrate the logic of model development to show readers the rationale of the research questions and the steps through each analysis.

Annotated screenshots are provided to help readers navigate the software program and learn the various techniques developed sequentially in each chapter. Readers are also introduced to diagnostic tools and how to identify data management issues. And, annotated syntax is included at the end for those who prefer a programming approach.

Third Edition highlights include:Updated throughout to reflect IBM SPSS Version 26. Introduction to fixed effects regression for examining change over time where random-effects modeling may not be an optimal choice. Additional treatment of key topics specifically aligned with our focus on multilevel modeling (MLM) (e.g., models with categorical outcomes).

Expanded coverage of models with cross-classified (and multiple membership) data structures. Added discussion on model checking for improvement (e.g., examining residuals, locating outliers). Further discussion of alternatives for dealing with missing data and the use of sample weights within multilevel data structures.

Expanded coverage illustrating different model-building sequences and checking output to locate possible improvements. A useful guide for readers to learn more about the basics of multilevel and longitudinal modeling and the expanded range of research problems that can be addressed through their application, this text is an essential resource for graduate students taking courses on multilevel, longitudinal, and latent variable modeling, multivariate statistics, or advanced quantitative techniques. It can be used as a standalone core text or with any multilevel- or longitudinal-modeling text.

It can also work as a supplement to the authors' other companion workbook, Multilevel Modeling of Categorical Outcomes Using IBM SPSS.