Skip to main content Skip to navigation

Scalable Two-Minute Feedback: Digital, Lecture-Accompanying Survey as a Continuous Feedback Instrument

Project Overview

The document explores the integration of generative AI, specifically large language models (LLMs) such as ChatGPT, in higher education through a digital formative feedback method known as the Two-Minute Feedback (2MF) survey. It emphasizes the significance of continuous feedback in enhancing teaching quality, particularly in large classroom settings. By utilizing AI to analyze and summarize student feedback, the study reveals how this approach can facilitate timely and actionable insights for educators. The findings indicate that while the use of AI in educational feedback processes can streamline data handling and improve responsiveness, it also presents challenges, including the need for careful implementation and potential biases in AI interpretations. Overall, the document demonstrates that generative AI has the potential to transform feedback mechanisms in education, leading to improved learning experiences and teaching effectiveness, while also necessitating ongoing evaluation of its impacts and limitations.

Key Applications

Two-Minute Feedback (2MF) survey with AI support for summarizing feedback

Context: Used in large introductory courses at two educational institutions, targeting both students in computer science and mathematics.

Implementation: The survey was administered weekly, collecting both quantitative and qualitative feedback through a digital platform, and AI was used to summarize the feedback.

Outcomes: Improved ability to analyze a large volume of student feedback quickly, leading to better insights for course adjustments and teaching improvements.

Challenges: Participation rates were low, raising concerns about the representativeness of feedback; ethical implications of handling sensitive student data were also discussed.

Implementation Barriers

Participation Barrier

Low participation rates in feedback surveys hinder the collection of comprehensive data.

Proposed Solutions: Incentivizing participation through gamification or providing dashboards for self-reflection may encourage more students to engage.

Ethical Barrier

Dealing with health-related data raises ethical concerns regarding intervention and privacy.

Proposed Solutions: Develop clear policies for handling sensitive data and use established questionnaires while ensuring compliance with data protection regulations.

Project Team

Armin Egetenmeier

Researcher

Sven Strickroth

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Armin Egetenmeier, Sven Strickroth

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

Let us know you agree to cookies