Towards an Integration of Content Analysis and Discourse Analysis: The Automatic Linkage of Key Relations in Text

Content analysis is a firmly established technique for textual data analysis. In particular, the notion of fully automatic, or at least computer assisted, content analysis has remained a desideratum among researchers working with large bodies of text, with the development of systems such as General Inquirer (Stone et al., 1966), Words (Iker and Harway, 1969), Quester (Cleveland, McTavish and Pirro, 1973), and Textpack (cf. Tesch, 1990). At the same time, however, the technique of content analysis has come under criticism in some quarters for the decontextualization of words from the discourse being examined. Billig’s (1989:206) criticism is typical: ‘This sort of methodology can count words, but it cannot interpret them. Under some circumstances mere counting can lead to misleading conclusions.’