FEHD Research
Standing at the forefront of research into education and psychology, the Faculty of Education and Human Development (FEHD) is the leading provider of educational courses and research at the Education University of Hong Kong (EdUHK). We conduct high-quality research that addresses educational, social, and policy issues of global significance.
Article of the Month
Discover this month’s featured article, showcasing cutting-edge research and impactful ideas from our Faculty’s experts across disciplines.
June 2026
Assessing Student Work in the Age of AI
The rise of Generative Artificial Intelligence (GenAI) has transformed higher education almost overnight. GenAI tools are increasingly integrated into how students brainstorm, draft, and refine their work. Yet this rapid shift has left universities grappling with urgent questions: What counts as a student’s “own” work? Should AI-assisted assignments receive lower grades? And what are grades meant to measure?
Two recent studies led by Dr Luo Jiahui Jess, Assistant Professor in the Department of Education Policy and Leadership at The Education University of Hong Kong (EdUHK), shed new light on these pressing issues. Published in Studies in Higher Education and Assessment and Evaluation in Higher Education, the research moves beyond the binary debate over whether AI use constitutes “cheating.” Instead, it examines how universities understand the “originality” of student work in policymaking and how teachers’ value judgements around AI shape grading decisions in practice. Across both studies, the findings suggest that higher education must shift its focus from policing AI use toward strengthening assessment validity and developing assessment policies that respond to evolving ways of learning and working in the AI era.
“Will I Get Lower Grades If I Declare Using AI to Help with My Assignment?”
In one study, Dr Luo and her team interviewed 33 university teachers using scenario-based discussions to explore how they would assess student work that might be mediated by AI. The findings show significant variations in how teachers made their grading decisions, with many non-academic factors nonetheless being prioritised. These factors tend to cluster around four key value orientations: person-oriented values (such as honesty and diligence), capability-oriented values (including independence and skill), relation-oriented values (trust between teacher and student), and justice-oriented values (fairness across the cohort).
The paper argues that these value judgements threaten the validity of grades because grades vary based on factors that are irrelevant to the outcomes being assessed. For example, marking a student’s work down because their declared use of AI was seen as a sign of poor effort threatens validity, unless effort is one of the outcomes being assessed. Conversely, giving full credit to students when AI did the substantive work of addressing the outcomes (perhaps because of beliefs about the student’s honesty) can also undermine validity, as the grades depend on the quality of the work produced by AI, not the student.
An Approach Grounded in Validity
The study points to a messy grading space full of tension and inconsistency. If left unaddressed, this will likely result in many unintended consequences, such as distrust from students, biased grading and weakened credibility of academic certifications.
To help teachers navigate these emerging challenges, the study highlights “validity” as a guiding concept. Validity centres grading on whether the assessment measures what it is designed to measure. Under this approach, grades should only be lowered if AI use prevents students from demonstrating the specific skills being assessed. Marks should not be deducted based on perceptions of laziness or effort unless those qualities are explicitly part of the learning outcomes.
“If I Use AI, Does That Make My Work Less Original?”
If classroom grading is shaped by hidden values, institutional policies reveal another layer of complexity. In a separate study, Dr Luo analysed GenAI policies from 20 world-leading universities to examine how institutions define the challenge.
She found that the dominant concern is the preservation of “originality” in student work. Many universities frame GenAI as a form of “external assistance” separate from the student’s independent efforts and intellectual contribution, thereby undermining the originality of their work. The paper cautions against this framing by showing how it fails to acknowledge the ways AI complicates the process of producing original work and the evolving meaning of originality at a time when knowledge production becomes increasingly distributed, collaborative and AI-mediated.
Dr Luo’s article “Exploring value judgements in grading: will teachers mark down student work assisted by GenAI, and should they?” was published in Studies in Higher Education in 2025 and can be accessed via DOI: 10.1080/03075079.2025.2552825. Her second paper, “A critical review of GenAI policies in higher education assessment: a call to reconsider the ‘originality’ of students’ work” was published in Assessment & Evaluation in Higher Education in 2024. The article is available via DOI: 10.1080/02602938.2024.2309963.
For enquiries about the research, please contact Dr Luo Jiahui Jess at jessluo@eduhk.hk.
.png)
