Scaling (from Section 3.3)

1. The scaling procedure is intended to ensure that unexpected outcomes in the examination are taken into account and that there is consistency with the QMDs of the examination scripts. Scaling should not be used to compensate for systematic under- or over-achievement in associated CA components.
2. If an Examiner determines that the outcomes of an assessment do not align with the QMDs, the Examiner should propose a scaling by setting a new value for each of the borderline marks: 40, 50, 60, 70. Intermediate marks will be scaled (by the mark sheet) using linear interpolation.
3. The Examiner should give a short explanation of the proposed scaling, or the proposed decision not to scale, as detailed on the Examination Paper Report Form.
4. The Examiner may wish to use the contextual indicators detailed below to justify their scaling decisions.
5. Every overall module mark should be in line with the QMDs regardless of the size of the cohort. This means that Examiners should always consider classification boundaries regardless of the module’s average.
6. Since every overall module mark should be in line with the QMDs, the default scaling for resits will be set to the accepted scaling for the main sitting in the same academic year. This assumes that an exam and resit set at the same time are of a similar level. The Examiner may still propose a different scaling.

The contextual indicators referred to above (from Section 1.1.3)

When setting an examination, the Examiner may wish to use the following contextual indicators as guidance:

1. The mean mark of the examination paper should normally lie between 55% and 68%;
2. A cohort’s performance, as measured by their mean score in the module, should not be substantially different from the same cohort’s performance in earlier year(s). [Generally a difference of more than 5% is considered substantial.]