Computer-assisted Text Analysis
Description:... Text analysis is a systematic reduction of a flow of text to a standard set of statistically manipulable symbols representing the presence, the intensity, or the frequency of some characteristics - usually called themes - which allows making replicable and valid inferences from text to their context. This entry presents an overview of present developments in which the computer plays an important role. Several types of text analysis based on hand or machine coding are distinguished: (1) thematic approach, which focuses on detecting (the frequency of) the occurrence of themes in texts and which is close to traditional content analysis; (2) semantic approach, considering subject-verb-object relations in several ways; and (3) network approach, an object in one sentence can be the subject in another one. The latter two especially allow researchers to answer many new research questions. These approaches require coding, which can be performed from the perspective of the investigator performing a study or of the author of the text. These perspectives can be different. One may concentrate on the manifest content of a text (i.e., what is really there) or the latent content (i.e., what one reads as it were between the lines). A problem with all texts is that they can be ambiguous. It is the investigator's task to overcome this problem and identify the content that is to be coded. Since about the year 2000, new approaches in which computer programs do "Ball the work" have been arriving. The idea behind such methods is explained here.
Show description