Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

From Assistance to Autonomy -- A Researcher Study on the Potential of AI Support for Qualitative Data Analysis

Elisabeth Kirsten, Annalina Buckmann, Leona Lassak, Nele Borgert, Abraham Mhaidli, Steffen Becker

TL;DR

The paper investigates how AI, especially large language models, can be integrated into qualitative data analysis (QDA) without compromising interpretive depth or researcher autonomy. Through semi-structured interviews with 15 HCI researchers, the authors map real-world QDA workflows, pain points, and conditions for AI involvement, and they propose a three-level framework (minimal, moderate, high) with concrete scenarios across QDA stages. The findings reveal broad openness to AI assistance for repetitive tasks and data exploration, coupled with strong demands for offline operation, explainability, and ongoing human oversight to ensure rigor and ethics. The work contributes a practical, adaptable blueprint for responsible human–AI collaboration in QDA and informs the design of AI-enabled QDA tools that balance efficiency with scholarly integrity.

Abstract

The advent of Artificial Intelligence (AI) tools, such as Large Language Models, has introduced new possibilities for Qualitative Data Analysis (QDA), offering both opportunities and challenges. To help navigate the responsible integration of AI into QDA, we conducted semi-structured interviews with 15 Human-Computer Interaction (HCI) researchers experienced in QDA. While our participants were open to AI support in their QDA workflows, they expressed concerns about data privacy, autonomy, and the quality of AI outputs. In response, we developed a framework that spans from minimal to high AI involvement, providing tangible scenarios for integrating AI into QDA practices while addressing researchers' needs and concerns. Aligned with real-life QDA workflows, we identify potential for AI tools in areas such as data pre-processing, researcher onboarding, or conflict mediation. Our framework aims to provoke further discussion on the development of AI-supported QDA and to help establish community standards for responsible Human-AI collaboration.

From Assistance to Autonomy -- A Researcher Study on the Potential of AI Support for Qualitative Data Analysis

TL;DR

The paper investigates how AI, especially large language models, can be integrated into qualitative data analysis (QDA) without compromising interpretive depth or researcher autonomy. Through semi-structured interviews with 15 HCI researchers, the authors map real-world QDA workflows, pain points, and conditions for AI involvement, and they propose a three-level framework (minimal, moderate, high) with concrete scenarios across QDA stages. The findings reveal broad openness to AI assistance for repetitive tasks and data exploration, coupled with strong demands for offline operation, explainability, and ongoing human oversight to ensure rigor and ethics. The work contributes a practical, adaptable blueprint for responsible human–AI collaboration in QDA and informs the design of AI-enabled QDA tools that balance efficiency with scholarly integrity.

Abstract

The advent of Artificial Intelligence (AI) tools, such as Large Language Models, has introduced new possibilities for Qualitative Data Analysis (QDA), offering both opportunities and challenges. To help navigate the responsible integration of AI into QDA, we conducted semi-structured interviews with 15 Human-Computer Interaction (HCI) researchers experienced in QDA. While our participants were open to AI support in their QDA workflows, they expressed concerns about data privacy, autonomy, and the quality of AI outputs. In response, we developed a framework that spans from minimal to high AI involvement, providing tangible scenarios for integrating AI into QDA practices while addressing researchers' needs and concerns. Aligned with real-life QDA workflows, we identify potential for AI tools in areas such as data pre-processing, researcher onboarding, or conflict mediation. Our framework aims to provoke further discussion on the development of AI-supported QDA and to help establish community standards for responsible Human-AI collaboration.

Paper Structure

This paper contains 66 sections, 15 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. Qualitative Data Analysis in CSCW and HCI
  4. From Computational Support to AI-Assisted Analysis
  5. Potentials and Risks of AI in the QDA Workflow
  6. Designing for Human-AI Collaboration
  7. Research Method
  8. Study Materials
  9. Pre-Questionnaire
  10. Interview Guide
  11. Recruitment & Participants
  12. Sample Description
  13. Ethical Considerations & Data Protection
  14. Data Analysis
  15. Results
  16. ...and 51 more sections

Figures (15)

  • Figure 1: Detailed overview of participants' real-world coding workflow, consisting of three stages for codebook creation, codebook refinement, codebook application, and different steps taken in inductive and deductive approaches.
  • Figure 2: Our framework for integrating AI into QDA, depicting 3 stages from less to more AI involvement, visualizing possible roles for AI we deduced from the data for each stage.
  • Figure 3: AI provides step-by-step guidance within QDA software, suggesting functionalities based on user behavior and answering questions to support with problems in tool usage.
  • Figure 4: AI generates summary reports of QDA projects, such as identifying the most or least frequent codes, or calculating IRR, allowing the researcher to review these insights and make informed decisions more efficiently.
  • Figure 5: AI supports new researchers in learning and practicing QDA through personalized, interactive instructions, real-time feedback, and simulated collaboration exercises to develop critical thinking and improve coding consistency.
  • ...and 10 more figures