2026 ASEE Annual Conference & Exposition

Integrating AI-Assisted Learning Analytics for Personalized STEM Instruction

Presented at DSAI-Session 5: Student Perceptions, Trust, and AI Use in Coursework

Generative artificial intelligence (GenAI) tools such as ChatGPT and Copilot are increasingly shaping how engineering students approach coursework, study routines, and conceptual learning. This study investigates undergraduate engineering students’ AI tool usage patterns and perceptions of AI-assisted learning, with an emphasis on learning support, critical thinking concerns, trust, and responsible use. Survey data were collected from 185 engineering students across multiple majors and academic levels. Descriptive analysis examined AI use frequency, purposes for AI use, and item-level attitudes measured using Likert-scale questions. A Composite AI Learning Attitudes Index was constructed from four aligned perception items, with reliability assessment performed using Cronbach’s alpha. Inferential testing included one-way ANOVA to compare composite attitudes across AI use frequency groups and multiple linear regression controlling for academic characteristics. Results indicate that AI tool use is widespread, with most students reporting at least occasional usage, primarily for homework support and exam preparation. Students generally agreed that AI tools improve understanding of course concepts, while many also expressed concern that overreliance could weaken critical-thinking skills. Trust in AI accuracy and ethical reliability remained moderate to low; however, strong support was observed for responsible AI integration in coursework. ANOVA and regression results showed that AI use frequency was a significant predictor of composite attitudes, with frequent users reporting substantially higher overall AI learning attitude scores than non-users. Findings suggest that engineering programs should adopt structured strategies for responsible GenAI integration that emphasize verification, critical thinking reinforcement, and AI literacy to support learning outcomes.

Authors

Dr. Eazaz Sadeghvaziri Mercer University [biography]

Dr. Eazaz Sadeghvaziri is an Assistant Professor at Mercer University with over a decade of experience in engineering education and Transportation Engineering. He holds a B.Sc. in Civil Engineering, an M.Sc. in Transportation Planning, and a Ph.D. in Transportation Engineering. His research interests include engineering education, travel patterns, smart mobility, and transportation safety. He conducts studies utilizing advanced data analysis techniques and geographic information systems to understand and improve urban mobility, promoting inclusive and equitable transportation systems through data analysis and planning. Dr. Sadeghvaziri has led various research projects and has a solid record of presenting and publishing in peer-reviewed conferences and journals. With prior private sector experience, Dr. Sadeghvaziri bridges the gap between academia and industry, applying theoretical knowledge to real-world challenges and fostering innovative solutions. As an educator, Dr. Sadeghvaziri mentors the next generation of engineers. He supervises research projects, promoting an environment of learning and innovation. His teaching emphasizes critical thinking and hands-on experience, preparing students to address complex transportation issues with creative solutions. Dr. Sadeghvaziri’s dedication to the field is evident through his active involvement in professional organizations, TRB committees, paper reviews, and contributions to advancing engineering education and transportation engineering.
Ramina Javid http://orcid.org/0000-0002-5487-1560 Morgan State University [biography]

Ramina Javid is a PhD candidate at Morgan State University. Her work focuses on leveraging Geographic Information Systems (GIS), travel behavior analysis, and advanced data science methods to address pressing challenges in mobility, traffic safety, and emerging technologies such as micromobility and automated vehicles. She has published and presented widely at conferences such as the Transportation Research Board, with projects ranging from evaluating Complete Streets through driving simulation to studying accessibility, crash severity, and the adoption of new mobility modes. Alongside her academic work, she is passionate about mentoring students, teaching, and creating innovative, data-driven solutions that contribute to a safer transportation system.
Keller Andrea Wilson Mercer University
Parham Atoofi Texas A&M University
Dr. Arash Afshar Mercer University [biography]

Dr. Arash Afshar is currently an associate professor in the School of Engineering at Mercer University. He earned his M.S in systems and design and Ph.D. in solid mechanics from the State University of New York at Stony Brook. He also received his B.S and M.S in Solid Mechanics from Amirkabir University of Technology in Tehran, Iran. His teaching and research interests are in the area of solid mechanics including composite materials, finite element analysis, mechanical design and machine learning. Prior to joining Mercer University, he taught at Saginaw Valley State University and worked as a design engineer in oil and gas and injection molding industries.

Download paper (870 KB)

Are you a researcher? Would you like to cite this paper? Visit the ASEE document repository at peer.asee.org for more tools and easy citations.

» Download paper

« View session