A Mathematical Model of Evaluator-Related Variation in Continuous Internal Assessment
Continuous Internal Assessment (CIA) is widely practised in colleges and universities, but students of similar merit, ability, and diligence may receive different marks because of differences in teachers’ marking tendencies. This study presents a model for a paper with three components, evaluated by three teachers. One teacher is taken as a reference for comparison, and the marking tendencies of the other two teachers are represented by parameters. The paper compares three arrangements: one teacher evaluates all three components, two teachers share the components in the sense that one teacher evaluates two components and another evaluates the remaining one, and three different teachers evaluate one component each. For each of these three arrangements, the variation in the possible total marks of a given student is determined. An analytical calculation shows that this variation is the smallest when three different teachers evaluate the three components, the largest when one teacher evaluates all three components, and intermediate when two teachers share the components. The paper also shows how the parameters representing the marking tendencies of evaluators can be estimated from the marks awarded. The model provides a framework for understanding evaluator-related variation in internal assessment and for studying how shared evaluation may reduce the influence of individual marking styles. Further studies can be carried out using this framework and real assessment data.
Authors
- Sudipto Roy
Publication Details
- Journal
- EdArXiv (OSF Preprints)
- Published
- 2026-11-17
- DOI
- https://doi.org/10.35542/osf.io/hd6e8
- Primary Topic
- Student Assessment and Feedback
- Type
- preprint