Using AI in Qualitative Analysis: Can Chatgpt and Human Raters Agree?

Poster Session III - Research Proposal Posters

Keyword: Other

Abstract: We propose investigating ChatGPT's potential to code qualitative responses using a human-established codebook. By comparing ChatGPT’s results to trained human raters, we aim to assess inter-rater reliability between ChatGPT and humans, and thus explore the potential of using AI to reduce the cost of qualitative analysis without compromising scientific rigor.

• Evelyn White (Presenting Author)
• Rebecca Lucke (Author)
• Sharn Kang (Author)
• Taiven Haggett (Author)
• Athena Bear (Author)
• Jianjian Qin (Author)