EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Can Generative Artificial Intelligence Reliably Score Open-Ended Question Assessments in Undergraduate Medical Education?

Doreen M Olvet iD, Marieke Kruidering iD, Tracy B Fulton iD, Bao Truong iD, Kumiko Endo iD, Robert Lucito iD, Joanne M Willey iD

Medical Science Educator · 03.03.2026 · 36 (3) · 1539-1552

DOI: 10.1007/s40670-026-02638-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Medical Science Educator, online erschienen am 03.03.2026. Verfasst von Doreen M Olvet und 6 weiteren Autor:innen. Beteiligte Einrichtung: Northwell, New Hyde Park, NY US. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: International Association of Medical Science Educators.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

There are numerous benefits to including open-ended questions (OEQs) in the assessment of medical knowledge, but one of the biggest challenges is the time it takes to grade student responses. With the widescale introduction of generative artificial intelligence (AI), it is plausible that OEQ exams can be automatically scored. The purpose of this study was to establish the accuracy of generative AI when scoring medical student OEQ exams. Students' responses from OEQs administered at two US allopathic medical schools were analyzed. Case vignettes, questions, rubrics, and student responses were fed into the GPT-4 model via the Med2Lab platform. The Med2Lab system was specifically engineered to manage rubric integration and automate prompt workflows. Scores and feedback on students' responses were generated and compared to faculty scores using Cohen's weighted kappa (kw) to evaluate inter-rater reliability (IRR). An error pattern analysis was performed to assess why there were scoring discrepancies between faculty and GPT-4, then this information was used to perform rubric engineering. We ran 3 iterations of GPT-4 scoring after each rubric adjustment. By the third iteration, IRR between faculty and GPT-4 was substantial using the analytic rubric (question 1A: kw=0.94; question 2A: kw=0.88) and the holistic rubric (question 2H: kw=0.89). IRR for question 1H reached moderate reliability (kw=0.54). We identified errors in GPT-4 and faculty scoring, although score discrepancies were typically only 1-point. Our data suggest that generative AI can be used to reliably score OEQ exams using an iterative process of rubric engineering to achieve maximum reliability. Supplementary Information: The online version contains supplementary material available at 10.1007/s40670-026-02638-2.

Schlagwörter

Stichwörter der Autor:innen

AssessmentGPT-4Generative artificial intelligenceOpen-ended questionUndergraduate medical education

Autor:innen und Institutionen

  • Northwell, New Hyde Park, NY US.
  • Department of Science Education, The Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, Uniondale, NY US.
  • Department of Cellular and Molecular Pharmacology, University of California at San Francisco School of Medicine, San Francisco, CA US.
  • Department of Biochemistry and Biophysics, University of California at San Francisco School of Medicine, San Francisco, CA US.
  • Department of Pediatrics, University of California at San Francisco School of Medicine, San Francisco, CA US.
  • Med2Lab, Inc, 44 Montgomery St. Ste 300, San Francisco, 94104 CA US.

ORCID-Profile: Doreen M Olvet, Marieke Kruidering, Tracy B Fulton, Bao Truong, Kumiko Endo, Robert Lucito, Joanne M Willey

Förderung

  • International Association of Medical Science Educators

Zitieren

APA

Olvet, D. M., Kruidering, M., Fulton, T. B., Truong, B., Endo, K., Lucito, R., & Willey, J. M. (2026). Can Generative Artificial Intelligence Reliably Score Open-Ended Question Assessments in Undergraduate Medical Education?. Medical Science Educator, 36(3), 1539-1552. https://doi.org/10.1007/s40670-026-02638-2

Vancouver

Olvet DM, Kruidering M, Fulton TB, Truong B, Endo K, Lucito R, et al. Can Generative Artificial Intelligence Reliably Score Open-Ended Question Assessments in Undergraduate Medical Education?. Medical Science Educator. 2026;36(3):1539-1552. doi:10.1007/s40670-026-02638-2

BibTeX
@article{Olvet2026,
  author = {Olvet, Doreen M and Kruidering, Marieke and Fulton, Tracy B and Truong, Bao and Endo, Kumiko and Lucito, Robert and Willey, Joanne M},
  title = {Can Generative Artificial Intelligence Reliably Score Open-Ended Question Assessments in Undergraduate Medical Education?},
  journal = {Medical Science Educator},
  year = {2026},
  volume = {36},
  number = {3},
  pages = {1539--1552},
  doi = {10.1007/s40670-026-02638-2}
}
Kostenlos starten →