Research into the validity, reliability and difficulty of comparative judgement to assess complex skills (including peer assessment). Research into learning by comparison.
Digging into the task difficulty of comparative judgement: its link with decision quality and low alignability.
AbstractComparative judgement is increasingly used to assess student work in secondary and higher education. Assessors are presented with two pieces of work and asked to indicate which work is better. Research underpins that comparative judgement can become difficult at times. Evidence univocally points to the impeding role of similarity and also indicates that task difficulty differs across assessors. Insight into several other aspects of task difficulty is, however, lacking. This research project focusses on two of these issues: (1) the relation of task difficulty with decision quality and (2) the role of low alignability in task difficulty. To answer these research questions, four samples will be used of which three are already collected. These three samples focus on the assessment of writing. The fourth sample will be collected in an authentic assessment on presentation skills. All samples include the comparative judgements made, assessors' justifications for their decisions or feedback on the pieces of work compared and information on task difficulty. Results of this project will provide more insight into the role of low alignability and initiate a new line of research by linking task difficulty to decision quality.
- Promotor: van Daal Tine