There are certain situations in which a candidate does not have a mark for a component/unit in a GCSE or AS/A level examination. For example, if they were ill on the day of the exam, if their paper was lost, or if their controlled assessment was invalid as a result of individual or centre malpractice. Subject to certain rules, the awarding body can calculate an estimated mark for the component/unit with the missing mark to enable the candidate to certificate, rather than having to wait for the next assessment opportunity.

This article explores the use of statistical methods for handling missing data, specifically regression imputation, to estimate the mark for a missing unit/component in GCSE and AS/A level qualifications. The marks (and grades) obtained in this way are compared with the marks (and grades) obtained applying two different methods currently used by some of the awarding boards in England: the z-score method and the percentile (cum% position) method.