There has been a worldwide increase in the number of postgraduate students over the last few years and therefore some examiners struggle to maintain high standards of consistency, accuracy and fairness. This is especially true in developing countries where the increase is supervision capacity is not on a par with the growth in student numbers. The aim of this research is to deploy freely available technology in order to find a way to help examiners to cope with this extra pressure, while maintaining the rigour of the assessment process. In terms of methodology, we commenced by mining the literature to ascertain exactly what criteria dissertation examiners were assessing, and how they went about doing this. We discovered that examiners tend first to gain an initial impression of a dissertation by reading the summary sections of the report: the abstract, introduction and conclusion. This delivers a helpful overview that eases the subsequent thorough examination of the dissertation, where they work their way through each chapter. This “overview then zoom” practice is reminiscent of the primary information visualisation mantra. This led us to consider whether knowledge visualisation could be the ameliorative mechanism we were looking for. We then carried out a systematic literature overview in order to determine whether knowledge visualisation had been used in this context. This revealed a surprising lack of research on the use of knowledge visualisation for assessment. We thus commenced to study extant use of visualisations. A case study approach was employed to study extant use of visualisations, in terms of how adequately they provided evidence of students having satisfied the previously identified assessment criteria. A number of experienced supervisors were then surveyed to gather their opinions about the role of knowledge visualisations in dissertations. Our findings indicate that knowledge visualisations can indeed provide evidence that particular criteria have been satisfied within a dissertation, and they do this more efficiently than text. Given the advances in technology, all postgraduate students are now able easily to produce computer-generated visualisations, so requiring their inclusion would be no great impediment. We conclude that knowledge visualisations demonstrate promise in terms of supporting assessment of postgraduate dissertations. Our recommendations are that the deliberate deployment of knowledge visualisations in this context be investigated further to determine whether this initial promise can be realised in actual practice.A video abstract of this article can be viewed at https://www.youtube.com/embed/Y7mcF2ZBNT8
展开▼
