Secondary Analysis of Qualitative Research Data

Utilizing Text Mining to Explore Transcribed Interviews

Abstract

Research data are increasingly being stored in digital repositories and made available to third parties, sometimes restricted to scientific use only. All ethical dimensions considered, the question arises as to how such data can be used for secondary research. This paper gives an exemplary answer by analyzing a large dataset of biographical interviews with 20^th century social scientists. While the original research was about these biographies in general, this secondary analysis focusses on changes brought on work biographies in the 20^th century. The paper describes ways to exploratively approach interview transcripts by using digital methods and reflects on the extent to which conclusions can be drawn from such an analysis. How meaningful are identified interview passages for one’s own research interest? What limitations remain when interpreting these? Exploring this, various standard approaches in text mining are examined in terms of usability and utility, including word frequencies, n-grams, and topic modeling.