Summary

In social research qualitative analysis of verbatim coded interviews is a common yet time consuming task. Artificial Intelligence (AI) like ChatGPT could assist here.
Previous research showed prevalent self-monitoring among researchers, especially when expertise is required. We also found that AI applications in qualitative analysis is are relatively unexplored, perhaps led by doubts on consistent responses.
This paper explores the performance in qualitative analysis of interviews of four general and contemporary AI systems (Bard, Bing-AI, ChatGPT, and WriteSonic). Using one set of quotations to assess the summarizing performance of these four AI systems.
Our research shows that ChatGPT may be a useful tool in summarizing quotations emerging from interview analysis. While variation in response quality of ChatGPT indicates potential pitfalls.