Assessment profiles in undergraduate dental anatomy education: a five-year retrospective analysis with implications for future AI-supported question Bank Development

A five-year retrospective analysis of undergraduate dental anatomy final examinations found significant year-to-year variation in assessment scores, according to the study’s abstract. The analysis included 1,216 eligible records from first- and second-year courses across the 2021–2022 through 2025–2026 academic years. Overall mean final examination performance was 63.91 ± 20.00, with a median score of 66.0. Scores differed significantly across academic years (p < 0.001), while the numerical advantage for second-year students was not statistically significant (p = 0.072).

The researchers analysed archived individual-level final examination records from first- and second-year undergraduate dental anatomy courses. They treated scores both as continuous variables and as categories assigned to predefined achievement bands. Class-level differences, variation between academic years and achievement-band distributions were evaluated through non-parametric and categorical statistical analyses. The eligible dataset comprised 666 first-year records and 550 second-year records. Artificial-intelligence-generated questions were not administered to students and did not contribute to grades, so the study did not test the performance or educational effects of an AI-assisted question bank.

The abstract reports that 2021–2022 performance was elevated relative to the other academic years in the dataset. This pattern coincided with pandemic-era online assessment conditions, but the available analysis cannot establish that the online examination format caused the higher scores. The findings therefore document longitudinal variation rather than a demonstrated causal effect of assessment format. Likewise, the absence of a statistically significant class-level difference does not establish that the two years achieved equivalent performance; it indicates that the observed numerical difference was not significant in the analysis.

Researchers described the results as a local retrospective assessment baseline that could inform future work on AI-assisted question bank development. The abstract also identifies prospective item-level psychometric validation and expert review as necessary before AI-assisted items are used in student assessment. Because the supplied evidence is limited to the study abstract, detailed information about the institution, examination content, achievement-band thresholds, statistical procedures and other methodological parameters cannot be verified from the available material. The study is a retrospective analysis of existing examination records, not a clinical trial, review, simulation or animal or cell study, and it provides no evidence that AI-generated questions improve dental students’ learning or performance.

Subject of Research: Five-year retrospective analysis of undergraduate dental anatomy examination performance and implications for AI-assisted question bank development.

Article Title: Assessment profiles in undergraduate dental anatomy education: a five-year retrospective analysis with implications for future AI-supported question Bank Development

Article References: Kiliç Şafak, N., & Karaca Bozdağ, Z. (2026). Assessment profiles in undergraduate dental anatomy education: a five-year retrospective analysis with implications for future AI-supported question Bank Development. BMC Medical Education. https://doi.org/10.1186/s12909-026-10422-4

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10422-4

Keywords: dental anatomy, undergraduate education, retrospective analysis, assessment profiles, examination scores, longitudinal variation, artificial intelligence, question banks, psychometric validation, Assessment, profiles, undergraduate