The use of large language models in generating multiple choice questions for health professions education: A systematic review and network meta-analysis
ChatGPT 4 may create similar quality MCQs to humans, whereas ChatGPT 3.5 and Llama 2 may be of worse quality. Further studies that directly compare these LLMs to human-generated questions and administer MCQs to students are required.
Authors
- Keean Nanji (ORCID: https://orcid.org/0000-0002-7176-6631)
- Wesla Pfeifer (ORCID: https://orcid.org/0000-0002-7909-4139)
- MOIZ LAKHANI
- Lauren Riehm
- Evelina Pankiv
- Dean Hasanee
Institutions
- University of Ottawa (CA)
- University of Toronto (CA)
- Hospital for Sick Children (CA)
- Impact (CA)
- McMaster University (CA)
Publication Details
- Journal
- PLoS ONE
- Published
- 2026-01-02
- DOI
- https://doi.org/10.1371/journal.pone.0340277
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00