A new study published on arXiv explores the potential of large language models (LLMs) as AI-interviewers for qualitative data collection. In a randomized experiment involving university students, AI-interviewers, using a small 7B parameter model, conducted interviews via synchronous text chat. The AI systems asked more questions and elicited more respondent text compared to human interviewers, while maintaining similar interview durations and response relevance. The findings suggest that AI-interviewing is a viable method for large-scale qualitative data gathering, addressing concerns about data security and reproducibility. AI
IMPACT AI-interviewing shows promise for scalable qualitative data collection, potentially bridging the gap between surveys and in-depth interviews.
RANK_REASON The cluster contains a research paper detailing an empirical comparison of AI-led vs. human-led interviews. [lever_c_demoted from research: ic=1 ai=1.0]
- 7B parameters
- AI-interviewers
- AI-interviewing
- Ai System
- arXiv
- General Data Protection Regulation
- large language model
- qualitative interviews
- randomized experiment
- synchronous text chat
- Tobias Priesholm Gårdhus
- university student
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
