Conference proceeding
Scaling Peer Feedback: AI-Assisted Analysis in Large Student Cohorts
World Engineering Education Forum & Global Engineering Deans Council Proceedings
WEEF& GEDC 2025 (Daegu, South Korea, 21/09/2025–25/09/2025)
12/2025
Metrics
1 Record Views
Abstract
Peer feedback plays a critical role in student learning by fostering reflection and constructive critique. In large cohorts, however, the volume of written peer feedback can be overwhelming and difficult to process effectively. This paper presents a practice-based approach implemented in the subject 41059 Mechanical Design Fundamentals Studio 1 at the University of Technology Sydney (UTS), where student teams engage in design reviews and receive peer feedback via an online form. Given the scale, over 1200 feedback entries across approximately 300 students, manual analysis proved impractical. To address this, Generative Artificial Intelligence (GenAI) was employed to synthesize feedback into concise, actionable summaries, highlighting strengths, areas for improvement, and tailored recommendations. Ethical considerations were addressed through informed consent and anonymization. The integration of GenAI significantly improved the clarity and timeliness of feedback, enabling students to leave review sessions with a clear understanding of their performance and areas for development. This study demonstrates that GenAI offers a scalable, efficient, and ethically sound solution for enhancing peer feedback in large educational settings.
Details
- Title
- Scaling Peer Feedback: AI-Assisted Analysis in Large Student Cohorts
- Creators
- Anna Lidfors Lindqvist - University of Technology SydneyKatrina Leung - University of Technology SydneyZachery Quince - Southern Cross University
- Publication Details
- World Engineering Education Forum & Global Engineering Deans Council Proceedings
- Conference
- WEEF& GEDC 2025 (Daegu, South Korea, 21/09/2025–25/09/2025)
- Publisher
- IEEE
- Identifiers
- 991013337989502368
- Copyright
- © 2025, IEEE.
- Academic Unit
- Centre for Teaching and Learning; Faculty of Science and Engineering
- Language
- English
- Resource Type
- Conference proceeding