This article, titled “Thematic Analysis and Artificial Intelligence: A Step-by-Step Process for Using ChatGPT in Thematic Analysis,” by Muhammad Naeem, Tracy Smith, and Lorna Thomas, explores the integration of generative Artificial Intelligence (AI), specifically ChatGPT, into the six steps of systematic thematic analysis. Published in the International Journal of Qualitative Methods, this research addresses a critical gap by providing a detailed guide, including specific prompts, for using ChatGPT in each phase of thematic analysis.
The paper’s primary contributions are twofold:
- It offers practical ChatGPT prompts for each of the six steps of systematic thematic analysis, which also serves to aid researcher training in this qualitative method.
- It provides input development for training AI in thematic analysis, detailing how to familiarize an AI system with the research context, as well as the researcher’s methodological and theoretical considerations. This approach aims to reduce human bias and enhance accountability and transparency in thematic analysis.
The authors leverage AI to address common limitations of traditional thematic analysis, such as subjectivity, potential biases due to the human inability to process large datasets, inconsistency, lack of generalizability, and the time-consuming nature of the process. The study highlights that AI can analyze complex patterns in large amounts of data, leading to more comprehensive and efficient analysis.
The paper outlines a modified six-step systematic thematic analysis process (originally introduced by Naeem et al., 2023), which includes:
- Familiarization with data and selection of quotations. This step involves thoroughly familiarizing ChatGPT with the research context, theoretical, methodological, and philosophical underpinnings.
- Selection of keywords based on the “6 Rs” framework (Realness, Richness, Repetition, Rationale, Repartee, and Regal). The article suggests that AI can identify a broader and more diverse range of keywords, enriching the analysis.
- Coding, where keywords and quotations are labeled to produce codes, reflecting the meaning and addressing research questions based on the “6 Rs” coding framework (Robust, Reflective, Resplendent, Relevant, Radical, and Righteous). The sources indicate that AI-generated codes can be more numerous and richer than manually generated ones.
- Theme development, involving organizing codes into categories based on their inter-relationships and theoretical considerations, guided by the “4 Rs” of theming (Reciprocal, Recognizable, Responsive, and Resourceful). AI-developed themes are shown to be broader and deeper.
- Conceptualization, which involves interpreting codes and themes to define and clarify concepts, linking them to theory. AI-generated concepts can offer wider and more comprehensive explanations, potentially expanding existing theoretical frameworks.
- Development of a conceptual model, synthesizing all concepts into a coherent framework that proposes solutions to the research question and contributes theoretically.
The authors used a case study from Naeem et al. (2024a), which explored the constraints and enablers of using scan-and-go apps for shopping, to compare the outcomes of AI-based and manual systematic thematic analysis.
Despite the benefits, the paper also acknowledges challenges associated with AI in research, such as the necessity for verification of results, algorithmic bias, the “black box” problem (lack of visibility into AI inputs and processes), researcher influence, and issues of reliability and accuracy. The authors emphasize that AI should complement and enhance traditional methods, but the researcher must remain fully involved in the process, providing guidance and ensuring consistency.
The toolkit described in the paper not only benefits systematic thematic analysis but can also enrich other qualitative research methods by helping identify patterns, categorize data into themes or codes, and use participants’ language to develop insights.
Reference: Naeem, M., Smith, T., & Thomas, L. (2025). Thematic analysis and artificial intelligence: A step-by-step process for using ChatGPT in thematic analysis. International Journal of Qualitative Methods, 24, 1–18. https://doi.org/10.1177/16094069251333886
