The Cathie Marsh Memorial Lecture

Overview

The use of Generative-AI tools based on the capabilities of Large Language Models (LLMs) is infiltrating every aspect of the qualitative research cycle, from generating ideas to inform design, through data collection, creation and transcription, all phases of data analysis, and writing about and communicating findings.

What this means for the professions is yet to be fully understood. Some hail the new era with enthusiasm, advocating for the adoption of these technologies to speed-up and improve qualitative research. Others entirely dismiss its use on ethical and/or methodological grounds. Neither extreme offers a perfect response for all situations.

In this lecture we pose a number of important questions that the qualitative community of practice are grappling with, the answers of which will shape the development of guidelines for harnessing Gen-AI for qualitative analysis in different contexts.

These include:

When might the use of Gen-AI for qualitative research be appropriate, and when is it absolutely not?

Is the answer different according to topic focus, data type, analytic method?

How do contextual factors like sector, discipline and geographies influence the debate?

Why do qualitative researchers consider using Gen-AI in the first place?

What problem are they seeking to solve?

What affect does the use of Gen-AI have on the legitimacy and reputation of qualitative research?

Researchers consider such questions for each qualitative project, yet the community of practice also needs to collectively discuss and address them, and this lecture is part of that process.