Harnessing the power of AI in qualitative research: Exploring, using and redesigning ChatGPT

Abstract

AI tools, particularly large-scale language model (LLM) based applications such as ChatGPT, have the potential to mitigate qualitative research workload. In this study, we conducted semi-structured interviews with 17 participants and held a co-design session with 13 qualitative researchers to develop a framework for designing prompts specifically crafted to support junior researchers and stakeholders interested in leveraging AI for qualitative research. Our findings indicate that improving transparency, providing guidance on prompts, and strengthening users' understanding of LLMs' capabilities significantly enhance their ability to interact with ChatGPT. By comparing researchers' attitudes toward LLM-supported qualitative analysis before and after the co-design process, we reveal that the shift from an initially negative to a positive perception is driven by increased familiarity with the LLM's capabilities and the implementation of prompt engineering techniques that enhance response transparency and, in turn, foster greater trust. This research not only highlights the importance of well-designed prompts in LLM applications but also offers reflections for qualitative researchers on the perception of AI's role. Finally, we emphasize the potential ethical risks and the impact of constructing AI ethical expectations by researchers, particularly those who are novices, on future research and AI development. • This study identifies the key challenges of applying ChatGPT in qualitative analysis, including prompt design and AI interpretability, and proposes a structured framework to address these issues. • The prompt design framework, developed with researcher feedback and aligned with traditional qualitative methods, enhances ChatGPT's effectiveness in qualitative analysis by improving context definition, methodological guidance, and data structuring. • The adaptability of the framework is discussed, highlighting its potential to support evolving AI models and serve as a reusable resource across future LLM-based tools. • Ethical considerations, such as transparency and accountability in AI- assisted analysis, are examined to promote reliable and responsible use of ChatGPT in qualitative research. • Practical implications for junior researchers are emphasized, as the framework provides a foundational tool to improve prompt design skills and proficiency in AI-supported qualitative analysis.

Ähnliche Arbeiten