Assessing learning gains

Abstract

Over the last 30 years a range of assessment strategies have been developed aiming to effectively capture students’ learning in Higher Education and one such strategy is measuring students’ learning gains. The main goal of this study was to examine whether academic performance within modules is a valid proxy for estimating students’ learning gains. A total of 17,700 Science and Social Science students in 111 modules at the Open University UK were included in our three-level linear growth-curve model. Results indicated that for students studying in Science disciplines modules, module accounted for 33% of variance in students’ initial achievements, and 26% of variance in subsequent learning gains, whereas for students studying in Social Science disciplines modules, module accounted for 6% of variance in initial achievements, and 19% or variance in subsequent learning gains. The importance of the nature of the consistent, high quality assessments in predicting learning gains is discussed.