ABSTRACT

Survey research in the social sciences usually involves a multitude of variables. For example, in a questionnaire survey there are many questions related to the survey objective as well as many demographic characteristics that are used to interpret and explain respondents’ answers. The advantage of correspondence analysis (CA) is the ability to visualize many variables simultaneously, but there is a limit to the number of variables that can be interactively coded, as illustrated in the previous chapter, owing to the large number of category combinations. When there are many variables an alternative procedure is to code the data in the form of stacked , or concatenated , tables. The relationship between each demographic variable and each attitudinal variable can then be interpreted in a joint map. In this chapter we give examples of this approach, both when there are several demographic characteristics and when there are responses to several questions.