BOARD # 390: Leveraging AI and Predictive Analytics for STEM Identity Development:Insights from the NSF S-STEM funded Engineering and Computer Science(ECS) Scholars Program

Michael W. Thompson; Anne Marie Spence; Nathan F Alleman; William A Booth; Sarah E Madsen; Taylor Wilby; Pacey Ham Mitchell

Download Paper | Permalink

Conference: 2025 ASEE Annual Conference & Exposition
Location: Montreal, Quebec, Canada
Publication Date: June 22, 2025
Start Date: June 22, 2025
End Date: August 15, 2025
Conference Session: NSF Grantees Poster Session II
Tagged Topics: Diversity and NSF Grantees Poster Session
Page Count: 6
DOI: 10.18260/1-2--55764
Permanent URL: https://peer.asee.org/55764
Download Count: 5

Paper Authors

biography

Michael W. Thompson Baylor University

visit author page

Michael Thompson received his BS, MS and PH.D. degrees in Electrical Engineering from Texas A&M University. He a professor in the Department of Electrical and Computer Engineering and previously served as the Associate Dean for Undergraduate Programs in the School of Engineering and Computer Science at Baylor University.

visit author page

biography

Anne Marie Spence Baylor University

visit author page

Clinical Professor
Mechanical Engineering

visit author page

author page

Executive Officer, Center for Enhanced Performance

visit author page

author page

Pacey Ham Mitchell Baylor University

Download Paper | Permalink

Abstract

This paper presents updated findings on the NSF S-STEM-funded ECS Scholars Program, which supports high-achieving, low-income students in Engineering and Computer Science. The program provides scholarships, faculty mentoring, research and internship opportunities, professional development, and social support, all aimed at promoting academic success and STEM identity formation.

Over three years, 31 students participated; nine left the program—six due to academic performance, two to pursue non-STEM majors, and one for another scholarship. Most departures occurred early, but retention improved significantly for those who continued beyond their first year. Currently, 19 of 22 students are on track for on-time graduation, with all expected to graduate within 4.5 years. In the first cohort, 10 of 11 students graduated within four years, all securing professional placement.

A critical part of our research involved EAB's Navigate platform, initially adopted for its predictive analytics to guide early interventions. Although the predictive analytics feature did not deliver the early, actionable data we anticipated, it provided insights for refining our approach. We shifted focus to tracking student engagement, which proved valuable for developing a strong STEM identity.

A key element in developing a data-driven approach for student success is creating methods that are easy for faculty and staff to use. Last year, we submitted an ASEE paper that proposed using generative AI to reduce challenges in data collection and analysis. This paper reports on what we feel is a significant contribution by describing a "proof of concept" system that leverages generative AI to track course-level attendance and student engagement providing instructors with summarized insights via email. The system is designed to be easily implemented for instructors, offering a practical tool for monitoring student engagement.

This approach illustrates how generative AI enables us to explore solutions that were previously challenging due to the complexities of coding and data management. By simplifying data collection, generative AI has reduced technical barriers, allowing us to focus on practical tools that support student success.

As the ECS Scholars Program concludes, we have effectively supported the academic and professional growth of our students. Emphasizing the development of engineering and computer science identity has been a key finding that will guide our future efforts to sustain the program. The combination of Navigate and generative AI shows promise in tracking student engagement and streamlining data collection, offering insights for future initiatives aimed at supporting underrepresented students in STEM.

Note: We submitted a paper last year. This paper will document that additional progress for the project. In our view, the results that will describes the specifics of how we used generative AI in and attendance monitoring system is a significant new contribution. I didn't reference the paper from last year because my understanding is that the process is supposed to be anonymous.

Citation
Format

Thompson, M. W., & Spence, A. M., & Alleman, N. F., & Booth, W. A., & Madsen, S. E., & Wilby, T., & Ham Mitchell, P. (2025, June), BOARD # 390: Leveraging AI and Predictive Analytics for STEM Identity Development:Insights from the NSF S-STEM funded Engineering and Computer Science(ECS) Scholars Program Paper presented at 2025 ASEE Annual Conference & Exposition , Montreal, Quebec, Canada . 10.18260/1-2--55764

TY - CPAPER
AB - This paper presents updated findings on the NSF S-STEM-funded ECS Scholars Program, which supports high-achieving, low-income students in Engineering and Computer Science. The program provides scholarships, faculty mentoring, research and internship opportunities, professional development, and social support, all aimed at promoting academic success and STEM identity formation.

Over three years, 31 students participated; nine left the program—six due to academic performance, two to pursue non-STEM majors, and one for another scholarship. Most departures occurred early, but retention improved significantly for those who continued beyond their first year. Currently, 19 of 22 students are on track for on-time graduation, with all expected to graduate within 4.5 years. In the first cohort, 10 of 11 students graduated within four years, all securing professional placement.

A critical part of our research involved EAB&#39;s Navigate platform, initially adopted for its predictive analytics to guide early interventions. Although the predictive analytics feature did not deliver the early, actionable data we anticipated, it provided insights for refining our approach. We shifted focus to tracking student engagement, which proved valuable for developing a strong STEM identity.

A key element in developing a data-driven approach for student success is creating methods that are easy for faculty and staff to use. Last year, we submitted an ASEE paper that proposed using generative AI to reduce challenges in data collection and analysis. This paper reports on what we feel is a significant contribution by describing a &quot;proof of concept&quot; system that leverages generative AI to track course-level attendance and student engagement providing instructors with summarized insights via email. The system is designed to be easily implemented for instructors, offering a practical tool for monitoring student engagement.

This approach illustrates how generative AI enables us to explore solutions that were previously challenging due to the complexities of coding and data management. By simplifying data collection, generative AI has reduced technical barriers, allowing us to focus on practical tools that support student success.

As the ECS Scholars Program concludes, we have effectively supported the academic and professional growth of our students. Emphasizing the development of engineering and computer science identity has been a key finding that will guide our future efforts to sustain the program. The combination of Navigate and generative AI shows promise in tracking student engagement and streamlining data collection, offering insights for future initiatives aimed at supporting underrepresented students in STEM.

Note: We submitted a paper last year. This paper will document that additional progress for the project. In our view, the results that will describes the specifics of how we used generative AI in and attendance monitoring system is a significant new contribution. I didn&#39;t reference the paper from last year because my understanding is that the process is supposed to be anonymous.
AU - Michael W. Thompson
AU - Anne Marie Spence
AU - Nathan F Alleman
AU - William A Booth
AU - Sarah E Madsen
AU - Taylor Wilby
AU - Pacey Ham Mitchell
CY - Montreal, Quebec, Canada
DA - 2025/06/22
PB - ASEE Conferences
TI - BOARD # 390: Leveraging AI and Predictive Analytics for STEM Identity Development:Insights from the NSF S-STEM funded Engineering and Computer Science(ECS) Scholars Program
UR - https://peer.asee.org/55764
DO - 10.18260/1-2--55764
ER -