Table of Contents

Part I: Preparation for Analysis. What is Multivariate Analysis? Characterizing Data for Analysis. Preparing for Data Analysis. Data Visualization. Data Screening and Transformations. Data Visualization. Selecting Appropriate Analyses. Part II: Regression Analysis. Simple Regression and Correlation. Multiple Regression and Correlation. Variable Selection in Regression. Special Regression Topics. Discriminant analysis. Logistic Regression. Regression Analysis with Survival Data. Principal Components Analysis. Factor Analysis. Cluster Analysis. Log-Linear Analysis. Correlated Outcomes Regression.

About the Author

Abdelmonem Afifi, Ph.D., has been Professor of Biostatistics in the School of Public Health, University of California, Los Angeles (UCLA) since 1965, and served as the Dean of the School from 1985 until 2000. His research includes multivariate and multilevel data analysis, handling missing observations in regression and discriminant analyses, meta-analysis, and model selection. Over the years, he taught well-attended courses in biostatistics for Public Health students and clinical research physicians, and doctoral-level courses in multivariate statistics and multilevel modeling. He has authored many publications in statistics and health related fields, including two widely used books (with multiple editions) on multivariate analysis. He received several prestigious awards for excellence in teaching and research.

Susanne May, Ph.D., is a Professor in the Department of Biostatistics at the University of Washington in Seattle. Her areas of expertise and interest include clinical trials, survival analysis, and longitudinal data analysis. She has more than 20 years of experience as a statistical collaborator and consultant on health related research projects. In addition to a number of methodological and applied publications, she is a coauthor (with Drs. Hosmer and Lemeshow) of Applied Survival Analysis: Regression Modeling of Time-to-Event Data. Dr. May has taught courses on introductory statistics, clinical trials, and survival analysis.

Robin A. Donatello, Dr. P.H., is an Associate Professor in the Department of Mathematics and Statistics and the Developer of the Data Science Initiative at California State University, Chico. Her areas of interest include applied research in the Public Health and Natural Science fields. She has expertise in data visualization, techniques to address missing and erroneous data, implementing reproducible research workflows, computational statistics and Data Science. Dr. Donatello teaches undergraduate and graduate level courses in statistical programming, applied statistics, and data science.

Virginia A. Clark, Ph. D., was professor emerita of Biostatistics and Biomathematics at UCLA. For 27 years, she taught courses in multivariate analysis and survival analysis, among others. In addition to this book, she is coauthor of four books on survival analysis, linear models and analysis of variance, and survey research as well as an introductory book on biostatistics. She published extensively in statistical and health science journals.

Reviews

"This book is an excellent resource for students and researchers of all levels. I have used earlier editions repeatedly in data-analysis courses for advanced undergraduates and graduate students in applied fields. The level of mathematical presentation is well matched to such settings. Not only are there excellent examples from biostatistics and public health, but there are also some very good business financial examples. The new chapter on Data Visualization in the new, sixth edition will be especially useful. Overall, the book is exceptionally well written and readable."
- Stanley Sclove, University of Illinois at Chicago"Editions of Practical Multivariate Analysis have been the mainstay of my graduate-level service course in applied data-analysis since 1985. It remains an extraordinary book -- packed with excellent examples, clear explanation and fine advice -- and has my highest possible recommendation. Among many reasons it remains so extraordinary, are three signaled directly in its title: it is practical rather than theoretical, analytic rather than technical, and it embodies a broader-than-usual conception of utilitarian multivariate methods.
Practical Multivariate Analysis connects readily to its audience’s reality. It uses concrete research questions and real data to motivate its content, illustrated by exemplary analyses using R, SAS, SPSS and STATA. It models how complex findings can be made comprehensible to a broader community.
It reaches beyond the typical spectrum of multivariate methods. It begins sensibly, discussing how multivariate data can be explored and displayed before complex analysis. Then come chapters on useful extensions to multiple regression analysis. While not usually considered “multivariate,” these latter methods connect an incoming audience to earlier acquired skills and extend them. Then follow the core chapters on “standard” multivariate methods, including canonical correlation, discriminant, principal-components, factor and cluster analyses. All are clearly presented, and then extended by excellent chapters on logistic regression, survival and log-linear analyses, and multilevel modeling, techniques that have proven useful and ubiquitous throughout social-science research.
In my view, Practical Multivariate Analysis is an excellent roadmap for conducting such analyses, and a fine model for ensuring that their complex findings can be communicated successfully to others."
- John B. Willett, Charles William Eliot Research Professor, Harvard University Graduate School of Education"The Practical Multivariate Analysis is a fun statistical modeling book to read. I enjoyed the rich insights the book has provided, which can only be accumulated through years of experience with the complexity in real data. It covers a large collection of statistical methods and models with a clear focus on application. Always discussing a model or method along with data examples, the book helps readers focus on important perspectives in applying the model, from choice of appropriate methods to interpretation of the results, while it still manages to maintain the
technique details at a minimal level. Readers with different backgrounds can all benefit from this book. It is valuable for researchers who are interested in analyzing their data with classical statistical models and interpreting the results. It is a good reading for new graduates in statistics who have not had a lot of experience with real data as the book provides many importance guidance in handling real data as well as watch-out advices. It can be used by applied data scientists and serve as a resourceful reference book for experienced consultants."
- Xia Wang, University of Cincinnati"The monograph belongs to the series Texts in Statistical Science and presents the sixth upgraded edition of the popular manual. It was first issued in 1984, and from that time won recognition as one of the best textbooks on the applied statistical modeling and analysis...Most of chapters of the first part of the textbook contain such subsections as “Introduction” or “Definition,” “Discussion” or “Examples,” “Summary” and “Problems”...This structure makes the book very reader-friendly written, helping to students and researchers in various fields to understand what for a statistical tool can serve, how to apply it, and to interpret computer outputs. There is not much of mathematical and statistical derivation, neither modern statistical techniques, but plenty of examples oriented to the easy “know-how” practical implementations of the classical multivariate methods."
- Stan Lipovetsky, Technometrics, Vol 62

"The authors wrote the sixth edition of this book for biomedical scientists, behavioural scientists, and academic researchers, who wish to perform and understand the results of multivariate statistical analyses. The book also describes when to ask for help from a statistical expert on multivariate analysis...The sixth edition has been updated with, in particular, a new chapter on data visualization, a distinction made between exploratory and confirmatory analyses, and a new section on generalized estimating equations. This new edition will enable the book to continue as one of the leading textbooks in the area, particularly for non-statisticians, since it provides a comprehensive, practical, and accessible introduction to multivariate analysis whilst keeping mathematical details to a minimum...The book is an excellent roadmap for multivariate analysis and a fine model for ensuring that complex findings can be successfully communicated in a paper."
- Luca Bertolaccini, ISCB News, July 2020