Multivariate Analysis of Categorical Data: Applications

A bonus is the insightful introduction by series editor Jan de Leeuw, putting applied statistics into perspective, blurring the somewhat artificial distinction between exploratory and confirmatory analysis. . . . Throughout the volume, strong attention is paid to the practical issues and to limitations of the analytical approaches (e.g., the effects of the number of categories within a variable on the resulting model). This seems especially useful, given that the intended audience includes ′nitty-gritty′ users and less theoretically inclined readers. John P. Van de Geer is careful to head off the reader′s temptation to base interpretation on the old ′the computer says it, so it must be right′ routine, by pointing out easy but faulty conclusions, such as those that might result from interactions and confounding factors. --Elizabeth L. Rose in Structural Equation Modeling Quote for both books "The set would be appropriate for use in a graduate course, with guidance from an instructor who has expertise in this approach to multivariate analysis. The interested researcher will find the set to be very helpful, particularly in terms of developing a coherent and accurate interpretation of the results." --Elizabeth L. Rose in Structural Equation Modeling Nonlinear analysis of categorical variables, that is, a variable that can sort objects into a limited number of distinct groups called "categories" is a useful technique for social scientists, particularly for those who do survey research. This book introduces readers to the application of a particular approach to categorical analysis, the GIFI system or multiple correspondence analysis. Using illustrative examples from a variety of disciplines, Van de Geer shows how to perform these techniques using standard computer programs, such as SPSS. He explains when to use such programs as PRIMALS, HOMALS, ANACORS, PRINCALS, CANALS, and OVERALS, what conditions need to be met for effective use of each program, and how to interpret the results based on the use of each of these programs. Detailed examples are used throughout the book to enable readers to comprehend what is "going on behind" the computer program.

"This set would be appropriate for use in a graduate course, with guidance from an instructor who has expertise in this approach to multivariate analysis. The interested researcher will find the set to be very helpful, particularly in terms of developing a coherent and accurate interpretation of the results."