The previous edition, published in 1993, has been totally revised and extended. Like the first edition, this book is intended to be practically oriented and didactic. The author's wish is to represent in each of the 25 chapter (and five appendices) a fixed amount of material. As a result, each chapter is exactly eight pages in length to offer a fixed amount of material both for reading and teaching. However, this does not always have pleasant consequences in some chapters, where more information would be useful for the reader. A short summary of the main points in the form of a list concludes each chapter.

As in the first edition, the book’s main thrust is toward the practice of correspondence analysis, so that most technical issues and mathematical aspects are gathered in appendix A. The theoretical part here is more extensive than that of the first edition, including additional theory on new topics such as canonical correspondence analysis, transition and regression relationships, stacked tables, subset correspondence analysis and the analysis of square tables. Concerning computational issues, a very strong feature of this edition is the use of the R program for all computations. The lengthy appendix B (45 pages) summarizes (on examples) possibilities of relating macros from R. In addition to that, three different technologies are described enabling the creation of the graphical displays presented in the book. No references at all are given in the 25 chapters; a relatively brief bibliography in appendix C is given to point readers toward further reading containing a more complete literature guide. A glossary of the most important terms and an epilogue with some final thoughts conclude the book.

Summarizing, this is a nice book for all those who wish to acquaint themselves with a versatile methodology of correspondence analysis and the way it can be used for the analysis and visualization of data arriving typically from the fields of social, environmental and health sciences, marketing and economics. Numerous examples provide a real flavour of the possibilities of the method.

Reviewer:

jant