Computational Text Analysis for Qualitative IS Research: A Methodological eRfletolcin

Qualitative analysis is an essential component of the dynamic process of sensemaking, where researchers sift through data to extract innovative insights that can contribute to new theoretical perspectives. In most cases, this involves analyzing unstructured text data gathered from naturalistic inquiries and secondary data material. However, due to the predominantly manual nature of qualitative text analysis, there is often a trade-off between feasibility and expanding the scope of a study, giving rise to criticism by quantitative scholars that theoretical generalizations from qualitative research often lack a larger empirical backing, are not reproducible, or are subjectively biased. As computational text analysis (CTA) gradually becomes more accessible, also new research opportunities for qualitative scholars arise, which must be aligned with traditional qualitative thinking and evaluation criteria. In this article, we explore the value and purpose, process, and validation of CTA in qualitative IS research. Drawing from a specific case illustration, we examine potential issues concerning data collection and sampling, analysis, and interpretation of findings. Additionally, we discuss the potential obstacles that qualitative researchers using CTA may encounter when conducting the study but also when submitting their work for consideration for publication in IS journals.