Degree Classification: Does the Calculation Model Affect the Award?

Background Universities have the freedom to define their own calculation model to define the degree classification awarded. The output profile features as a key metric in ranking tables, yet this conceptually could be affected by the calculation method and provide a source of inequality. Method The scores from Level 5 & 6 modules from a group of final year students (n=50) was selected. Four different (A,B,C,D) models were applied to the same data to calculate the final degree score and subsequent award classification and analysed based on raw scores and rounded values. Results All four models appear to deliver similar calculated scores (Mean: A=62.9%: B=65.7%: C=64.8%: D=62.7%) however there is a distinct impact on the degree classification profiles. The proportion of students achieving First or Upper Second class awards for models A to D are 72%, 80%, 74%, 70% respectively. If rounding is applied this changes to 72%, 82%, 78%, 70%. Additional application of discretion at classification boundaries may further positively impact the results. Calculation models have minimal impact on lower class awards. Conclusion The results demonstrate that the calculation model has an effect on the degree classification awarded. In particular, models B and C produce more favourable outcomes. Universities using these models may benefit from an improved contribution to ranking performance. Burgess, R. (2007) Beyond the honours degree classification: Burgess Group Final Report, Universities UK