Abstract / Summary
Background/Objectives: Oral examinations can assess clinical reasoning and professional judgement, but examiner-dependent variation and unclear expectations may compromise reliability, procedural fairness, and accountability. This study investigated German dental students’ perceptions of oral examinations and identified learner-informed priorities for fair, transparent, and ethically responsible assessment. Methods: This national cross-sectional mixed-methods study used a participatory approach to questionnaire development. The main online survey was distributed to undergraduate dental students through national student networks. Three supplementary questions were administered separately to participants at the 2026 National Conference of German Dental Student Representatives. Quantitative data were analysed descriptively. Free-text responses were examined using qualitative content analysis conducted collaboratively by two researchers. Results: The main survey included 368 students from 26 dental schools, representing 84% of German dental schools; the separate conference-based dataset comprised 168 participants. Students strongly endorsed clearly defined competencies required to pass (mean 4.74 ± 0.81), transparent assessment criteria (4.64 ± 0.87), and structured questions to promote consistent and fair assessment (4.42 ± 0.99). Conference participants perceived a substantial influence of examiners on grades (4.25 ± 0.81). The qualitative findings contextualised the quantitative results across four areas: fairness and examiner influence; transparency of expectations and criteria; structure and consistency; and psychological safety, feedback, and preparation. Conclusions: Students viewed fairness, transparency, and accountability as integral to assessment quality. Structured oral examinations may support more equitable assessment while preserving opportunities to evaluate clinical reasoning and professional judgement. Because qualitative coding was conducted collaboratively rather than independently, no formal inter-rater reliability coefficient was calculated, limiting the reproducibility of the qualitative findings.