ArticleScientific reports2025
Benchmarking ChatGPT-generated multiple-choice questions against faculty-authored items in dental education.
Article in Scientific reports, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 3 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
3 citing papers in PubMed.
- Human-Edited Generative AI-Assisted Multiple-Choice Questions in Postgraduate Family Medicine: Blinded Cross-Sectional Comparative Psychometric Study.JMIR medical education · 2026Article
- Article
- Deep learning-driven discovery of anti-metastatic phytochemicals targeting MMP-1 in breast cancer via advanced contrastive learning and structural attention.Molecular diversity · 2026Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
7 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
Large language models (LLMs) have become an integral part of self-directed student learning. This study aimed to benchmark assessment items generated by ChatGPT against those written by the faculty instructors using the item response theory. A 40-item Oral Medicine assessment was generated for undergraduate dental students using 20 MCQs written by instructors and 20 MCQs were generated by ChatGPT 3.5 and administered to 547 students in four institutions. The results were analyzed using a 3-Parameter Logistic model. The person reliability of ChatGPT items was 0.778 with a Mean Absolute Deviation of Q3 (MADaQ3) of 0.0741 (p < .001). Most items had a guessing parameter of 0.000. The difficulty ranged from -5.57 to 1.64 and discrimination ranged from -0.56 to 3.85. The person reliability of instructor items was 0.845 with a mean absolute deviation of Q3 (MADaQ3) of 0.0698 (p < .001). The guessing parameter was 0.00, while the difficulty ranged from -1.29 to 3.22 and discrimination ranged from 0.50 to 4.64. Overall, the instructor items showed higher average scores and greater variability (instructor mean = 11.6, SD = 5.5 vs ChatGPT mean = 9.01, SD = 4.32). The instructor items exhibited higher discrimination, a wider difficulty range, better alignment with respondent abilities, and consistent fit indices. However, the ChatGPT generated items showed promise and may outperform humans in MCQ generation in the future.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.