No abstract

This article reports some findings which have emerged from research into assessment that has been carried out over a number of years by the Student Assessment and Classification Working Group in the UK. The findings raise questions at a number of levels about assessment practice. The heterogeneity in the distribution of honours degree classes that is visible in Higher Education Statistics Agency (HESA) statistics has a parallel at the level of the module, which provokes some disquiet about the equitability of assessment across disciplinary areas. Differences between performances in coursework and examinations prompt questions about the purposes and ‘authenticity’ of assessments. Initial work on lecturers’ marking behaviour, coupled with the preceding points, suggests that there is considerable scope for professional development in the area of assessment.

It is commonly believed that the standard of student performance in coursework tends to be higher than that achieved in formal examinations. This view was tested by analysing undergraduat e performances in six subjects at four UK universities. Two measures of relative coursework performance were employed. The rst is the difference between the mean coursework and examination marks for each module. The second considers the proportion of students in each module who achieve a higher mark in the coursework than in the examination. The measures showed that in English and History coursework performances are slightly higher, equivalent to one-third of one honours class (or division) while, in Biology, Business Studies, Computer Studies and Law, coursework performances are higher by as much as two-thirds of one honours class (or division). The differences observed in the latter subjects are very signi cant and have serious implications for parity of treatment in degree programmes where students may choose modules with contrasting modes of assessment.

Intertidal mudflat channels (gullies) in the Solway Firth, Scotland possess width/depth ratios similar to meandering rivers. Most channels deeper than 1 m show cut‐bank slides, but narrow, deep channels also have rotational slides on the point‐bar slopes. The channels display two types of point‐bar. The first, associated with gently curved meanders, is sigmoidal in profile. The second, associated with tight meander bends, possesses a pronounced lower platform. The onset of flow separation in meander bends, a phenomenon which enhances cut‐bank erosion and point‐bar deposition, is a direct function of meander‐bend tightness and Froude number. The effects of flow separation are greatest on tight meander bends at times of high velocity during late spring ebb and also during rainfall run‐off at low tide. These events appear to be responsible for the growth of the point‐bar platforms. A model, predicting the type of point‐bar development to be expected in different channel meanders, is used to reconstruct the sedimentary history of active and fossil point‐bars.