Author Information
Ying Wu, University of Dundee, United KingdomAndrew Ross, University of Dundee, United Kingdom
Malcolm Stewart, University of Stirling, United Kingdom
Fahima Alam, Queen Mary University of London, United Kingdom
Abstract
Universities must prepare marketing graduates for practice in which generative artificial intelligence (AI) is routine, while preserving assessment validity and independent disciplinary judgement. Detection-led prohibition is difficult to enforce reliably, and permission without evidential requirements can weaken assessability, yet the literature does not specify the design conditions under which authentic assessment converts permitted AI participation into demonstrable evaluative judgement. This study examines two iterations of an undergraduate marketing module, with a postgraduate module as an embedded contrast, across 191 enrolled and 182 marked students. It combines a documentary audit of review forms, briefs and rubrics with a documentary census of 182 submissions across four assessment sets, analysed abductively against stated criteria. Three findings are advanced. Client authenticity supplied constraints against which generated output could be judged, but the assessment architecture specified no AI participation and carried no AI-related criterion in either iteration. Although generative AI was taught in class and demonstrated for specified creative operations, documented declaration rates were 5.8% and 28.6% across the two undergraduate iterations, and of 182 submissions none evidenced revision or disciplinary justification of generated output. Mean marks in the focal undergraduate module did not rise, but no AI-related criterion existed from which an effect of specification could be evaluated. The case is diagnostic rather than evaluative. The study proposes specified latitude, the task-stage-specific mapping of permitted AI operations to the evidence of judgement they must leave, operationalised through CRAFT as a design response to the documented gap.
Paper Information
Conference: ECE2026Stream: Design
Published: Wednesday, September 23, 2026
This paper is part of the ECE2026 Conference Proceedings (View)
Full Paper
View / Download the full paper in a new tab/window
To cite this article:
Wu Y., Ross A., Stewart M., & Alam F. (2026) Designing AI-Supported Assessment in Marketing Education: Specified Latitude and the CRAFT Framework ISSN: 2188-1162 The European Conference on Education 2026: Official Conference Proceedings (pp. 637-658) https://doi.org/10.22492/issn.2188-1162.2026.49
To link to this article: https://doi.org/10.22492/issn.2188-1162.2026.49
Comments
Powered by WP LinkPress