Exploring the Efficacy of ChatGPT in an Undergraduate Electronics Course through Comparative Analysis of Circuit Problem Solving

This study explored the efficacy of using ChatGPT, a generative AI application, in a sophomore-year undergraduate engineering electronics course focusing on students’ learning of fundamental concepts and circuit problem-solving. As a part of this study, a group of sophomore-year engineering students were assigned a set of electronics circuit problems to solve first using traditional methods, followed by repeating the process using ChatGPT. Their experience was captured in a questionnaire-based survey, which attempted to gauge the perceived efficacy of ChatGPT or similar generative AI in electronics circuit problem-solving against the traditional problem-solving method under five criteria: accuracy, efficiency, learning retention, user experience, and assistance in critical thinking, using a 5-scale evaluation method. Additionally, some open-ended questions were included in the study for a qualitative assessment of ChatGPT’s perceived efficacy. The outcome of this exploratory research prompted a critical examination of the application of generative AI and similar assistive tools in the electronics course. Through the qualitative assessment, this study also identified several potential factors contributing to the observed performance differences between traditional and AI-assisted learning and problem-solving. Results from this study highlighted ChatGPT’s inherent strengths in efficiency and speed, while frequently compromising accuracy by being prone to errors and misinformation. Participants also emphasized the human factor of learning and advocated for the active involvement of instructors in training students in the proper and effective use of AI and similar assistive tools. Insights from this study offer a nuanced understanding of the role of generative AI in engineering education, encouraging further research into optimizing the use of technology in the classroom.