Little Green Book

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 the method is to be found in Knoke and Burke (1980). Considering the current increasing interest in correspondence analysis, it is natural to consider the relationship between the two methods. This issue is discussed in detail in, for example, Goodman (1987) and van der Heijden and de Leeuw (1985), but in different contexts. In this chapter the two methods of ...

Users without a subscription are not able to see the full content on this page. Please, subscribe or login to access all Methods content.

Click here to see full text

Articles in Google Scholar by