2026 ASEE Annual Conference & Exposition

This paper is a conceptual paper exploring knowledge stratification. Designing assessment tasks to adequately test what has been taught and to meet the learning objectives of a course is a non-trivial task. In the light of Generative AI it has become even more challenging. Whilst much ink has been spilled on the issue of assessment tasks being outsourced to LLMs the issue that is more pertinent to engineering education is establishing what needs to be taught to ensure best engineering practice. To this end developing a course specific knowledge stratification framework or epistemic framework may be helpful.
In this paper, the use of Blackie’s epistemic assessment framework will be demonstrated using mass-energy balances as an example. The implementation of the framework to facilitate the adaptation of existing assessment tasks and the necessitation of the development of new assessment tasks will be discussed. This framework which distinguishes between ‘discipline specific vocabulary’, ‘procedural knowledge’, ‘conceptual knowledge’ and ‘powerful knowledge’. The framework may need augmentation or adaptation for each specific course but provides a useful starting point for knowledge stratification. The balance in the kinds of knowledge required as the use of AI tools and AI agents becomes more common place in engineering practice is likely to change. This framework gives a foundation from which strategic decisions can be made to make such changes.