Summary
Contents
This volume provides readers with a simple, non-technical introduction to correspondence analysis (CA), a technique for summarily describing the relationships among categorical variables in large tables. It begins with the history and logic of CA. The author shows readers the steps to the analysis: category profiles and masses are computed, the distances between these points calculated and the best-fitting space of n-dimensions located. There are glossaries on appropriate programs from SAS and SPSS for doing CA and the book concludes with a comparison of CA and log-linear models.
Correspondence Analysis and Loglinear Models
Correspondence Analysis and Loglinear Models
In Anglo-American social science, the loglinear model is one of the most widely used methods for analyzing categorical variables. It is a very powerful method for analyses of multivariate contingency tables, especially for uncovering interactions between variables. Correspondence analysis and loglinear models have different advantages, so the choice of method depends on the type of data to be analyzed and on what relations or effects are of most interest. In the following sections the reader is presumed to have some knowledge of loglinear models. An elementary and adequate introduction to ...