ChatGPT in the Classroom: Boon or Bane for Physics Students' Academic Performance?
Project Overview
The document examines the role of generative AI, specifically ChatGPT, in education, focusing on its influence on student performance in a physics course. An experimental comparison of traditional teaching methods versus those incorporating ChatGPT showed a notable decline in academic performance among students using the AI tool. Although a subset of students reported that ChatGPT aided their comprehension of complex concepts, a significant number raised concerns about its impact on critical thinking and the risk of undermining independent learning skills. The findings suggest that while generative AI can be beneficial in enhancing understanding, it should be approached with caution and utilized as a supplementary resource alongside traditional educational methods, rather than as a replacement.
Key Applications
ChatGPT as an instructional aid
Context: Second-semester engineering physics course, targeting undergraduate students
Implementation: Integrated ChatGPT in the course, encouraging students to use it for inquiries and problem-solving.
Outcomes: Lower grades and negative conceptual understanding as measured by Hake factors for students using ChatGPT compared to the control group.
Challenges: Concerns about reduced critical thinking and independent learning; significant decrease in student performance.
Implementation Barriers
Educational Effectiveness and Student Engagement
ChatGPT may provide superficial understanding, reduce active engagement in learning, and lead students to rely on provided answers rather than think critically.
Proposed Solutions: Use ChatGPT as a supplement to traditional methods, encourage critical assessment of AI-provided information, and integrate ChatGPT in a way that promotes independent learning and critical thinking.
Project Team
Manuel G. Forero
Researcher
H. J. Herrera-Suárez
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Manuel G. Forero, H. J. Herrera-Suárez
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