Designing an Integrated Multi-Dimensional Assessment Framework for AI-supported Academic Writing
Type
conference paper
Date Issued
2025
Author(s)
Abstract
The widespread adoption of generative AI is transforming academic writing in higher education, rendering traditional, product-focused assessment models obsolete. These methods fail to capture the iterative and tool-mediated nature of modern writing processes, creating an urgent need for new evaluation approaches. This paper addresses this gap by proposing an integrated, multi-dimensional assessment framework. Grounded in genre pedagogy, self-regulated learning, and writing analytics, our model conceptualizes assessment as a holistic process. It evaluates foundational skills for AI use (Computational Thinking and Genre-Knowledge), the quality of text revision, final writing performance, and long-term development. By aligning these dimensions with formative, summative, and diagnostic purposes, the framework fosters transparency, metacognitive engagement, and responsible AI use. We are preparing a design-based research project to pilot and iteratively refine key elements of the framework within a mastery learning programme for 1,800 first-year students. The goal is to explore how the model can be implemented in practice and refined through iterative cycles of design, enactment, and evaluation.
Language
English
Keywords
Generative AI
Academic Writing
Writing Analytics
Evidence-Centered Design (ECD)
Computational Thinking (CT)
Event Title
Cognition and Exploratory Learning in Digital Age (CELDA)
Event Location
Porto, Portugal