Abstract
It is argued that the most likely explanation of the upward drift in degree results over the past two decades is that the assessment instrument has been changed, through the inclusion of coursework (so that it now does not measure what it measured before), although the degree boundaries have been left unchanged. If this is so, then the effect reflects bad practice on the part of examiners rather than changes in student achievement. To discover whether there have been changes in student achievement requires a comparison of student performance with fixed standards. This article discusses the extent to which this can be achieved in practice and proposes appropriate means for achieving it.

This publication has 4 references indexed in Scilit: