Scaling Standard Set Marks to University Grade Boundaries in Health Education

Abstract

Assessments in higher education healthcare programmes can be challenging because they not only need to be fair, valid, and transparent, but it is also necessary to gauge safety, practical skill competence, and professionalism. One way to help maximise validity in practical assessments is to utilise ‘standard setting’ which aims to set a fair ‘cut score’ (pass mark) that reflects the score expected of a ‘minimally competent’ candidate. This purports that if a hypothetically minimally competent candidate could achieve this score, then it should be equivalent to the minimum pass mark for that assessment i.e. everyone who performs better than that should pass. This is effective from an assessment design point of view, but then leads to the new challenge that the cut score is unlikely to be equivalent to the university-level pass mark of 40% (or 50% for postgraduate), which, in cases where the component carries ‘weighting’, can lead to differences in pass mark across different assessments within the same module, or the same programme, which requires a way of scaling the marks for each assessment to make the cut score align with the university-level pass mark.

However, the guidance for this is limited and unclear, so we propose a method of linear interpolation which overcomes any disadvantages of percentage / mark scaling and scaling based on cohort performance. We have also shared an easy-to-use excel file which shows you how to incorporate this method of scaling into your own assessments with ease.