Understanding contexts and mechanisms through which video based benchmarking promotes alignment of examiners' scoring...

Aligning examiners judgements to a shared standard is desirable within Objective Structured Clinical Exams (OSCEs) because it accords with OSCEs' epistemic assumptions and purpose. Video-based benchmarking (VBB) involves examiners scoring station specific videos and comparing their judgements against an agreed score. Despite promising potential to enhance examiner alignment, VBB has been little researched. As the intervention's efficacy is likely influenced by context, we used Realist Evaluation to study how, why and under what circumstances VBB may aid examiner alignment. Within stimulated interviews, examiners viewed videos of year-3 student OSCE performances and received expert-panel derived benchmark information. Different participants experienced adaptations of VBB, including varying benchmark performance standards, and different timings of receiving the benchmark information. Our resulting middle range theory suggests that despite conscientious preparation with written material, examiners' expectations remain uncertain until encountering concrete performance examples. Judging a video serves this role, whilst comparison to a standard allows examiners to locate their judgement. Examiners experience dissonance when their scores are discrepant, and several contextual factors mediate whether they dismiss the suggested standard or adjust their perspectives. Examiners' degree of engagement with video observation and score reflection is critical to VBB's efficacy, and is influenced by examiner factors and other procedural contexts. These procedural contexts underpin tentative recommendations for practice which may enhance the likelihood of VBB's effectiveness. When conscientious examiners engage effectively with well-constructed benchmarks, they perceive themselves as more prepared and aligned in their judgements because comparative reflection produces adjustment to their frame of reference.