Mark distribution is affected by the type of assignment but not by features of the marking scheme in a biomedical sciences department of a UK university

Abstract Marking schemes are a tool to ensure fairness in assessment of student work. Key features of fairness are that different markers would award the same mark to the same work and that the resulting marks effectively discriminate between different levels of student attainment. This study focuses on the ability of assessment to discriminate by analysing the mark distributions resulting from the use of different types of marking scheme in a real-world setting in a research-intensive UK university. This analysis shows that, in qualitative assessment, the mark distribution is unaffected by features of the marking scheme used. Instead, it shows that the type of assignment used has a significant effect on the mark distribution and that these effects are sometimes counterintuitive. Marking schemes are unlikely to be an effective tool in shaping mark distributions. To determine the effectiveness of approaches to assessment, we need to interrogate data rather than make assumptions.