Level Up Peer Review in Education: Investigating genAI-driven Gamification system and its influence on Peer Feedback Effectiveness
Project Overview
The document explores the use of generative AI within the peer feedback platform Socratique, designed to enhance peer review processes in software engineering education by incorporating gamification elements. A randomized controlled experiment revealed that students utilizing Socratique, supported by generative AI, produced a greater quantity and quality of feedback—particularly excelling in relevance and specificity—compared to those in a control group. These findings indicate that the integration of gamification with AI not only motivates students but also improves their feedback literacy, thereby bridging the gap between educational outcomes and industry expectations. Overall, the application of generative AI in this context demonstrates its potential to significantly enrich the educational experience and better prepare students for professional demands.
Key Applications
Socratique, a gamified peer-assessment platform integrated with Generative AI assistance.
Context: A functional programming course for master's students.
Implementation: The platform was piloted in Spring 2024, using a randomized controlled experiment comparing a gamified GenAI setup with a control group.
Outcomes: Students in the treatment group provided significantly more feedback and higher quality feedback in terms of clarity, relevance, and specificity.
Challenges: Low engagement in traditional peer feedback systems and the varying quality of student reviews.
Implementation Barriers
Engagement barrier
Low learning engagement in traditional peer feedback systems, with students often showing low enthusiasm and participation.
Proposed Solutions: Integrating gamification elements like leaderboards, badges, and rewards to increase motivation.
Quality barrier
The quality of peer reviews can be low and not trustworthy, making effective feedback difficult.
Proposed Solutions: Using Generative AI to provide automated, qualitative evaluations of feedback, thus improving review quality.
Project Team
Rafal Wlodarski
Researcher
Leonardo da Silva Sousa
Researcher
Allison Connell Pensky
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Rafal Wlodarski, Leonardo da Silva Sousa, Allison Connell Pensky
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai