ABSTRACT
Summary: Tertiary educational institutions have various methods of assessing their students’ performance. Ideally, formal summative student assessment should consist of two parts: first, the setting of examination papers which reasonably reflect both the content and difficulty level of the course in question; and secondly, based on the examination results, the assignment of student grades which give a fair and meaningful representation of students’ academic performance. It is imperative that both parts meet so that an unbiased and standardized assessment can be achieved.
This chapter proposes a system that takes the set of raw numerical scores (R-scores) from a properly constructed examination paper and statistically maps them on to a set of standardized adjusted numerical scores (A-scores). The A-scores are then mapped linearly on to a semi-extended letter grade scale which is suggested as the most meaningful way in which to represent the students’ academic achievement. Implications of the proposed system on the numbers of failures and distinctions, staff/student morale and academic standards are discussed.
