ChatGPT in Thematic Analysis: Can AI become a research assistant in qualitative research?
Description of Resource
The release of ChatGPT in November 2022 heralded a new era in various professional fields, yet its application in qualitative data analysis (QDA) remains underdeveloped. This article presents an experiment involving applying ChatGPT (Model GPT-4) to thematic analysis. By employing an adapted version of King et al.’s (2018) Template Analysis framework, this article aims to assess how ChatGPT can help with QDA in a full analytical process of a sample dataset provided by Lumivero. My experiment includes applying ChatGPT in four stages: data familiarization; preliminary coding and initial template formation; clustering and template modification and finalization; and theme development. Findings reveal GPT-4’s capacity in efficiency and speed in grasping the data and generating codes, subcodes, clusters, and themes, alongside its learning and adapting capabilities. However, the current version of the model has limitations in terms of effectively handling detailed analysis of large databases and producing consistent results, as well as the need to move across workspaces and the lack of relevant training data for QDA purposes.