Track 2: Collaborative Research: Supporting Engineering Retention with Veteran Empowerment through AI-supported Mentoring

This project will develop an artificial intelligence (AI)-supported mentoring program for improving the retention of student veterans and service members (SVSMs) in engineering education. A lack of appropriate academic support often alienates and isolates SVSMs, leading to lower engagement, reduced academic performance, and higher attrition rates. This project will design and develop an AI-supported mentor for SVSMs that reflects the lived experiences, values, and communication styles of military veterans. This culturally congruent AI-supported mentor will provide SVSMs with tailored academic assistance during their transition from the structured, hierarchical environment of military service to the flexible, self-directed academic settings of engineering education. This project directly supports the National Science Foundation's mission to promote the progress of science and advance national prosperity by strengthening pathways into the engineering workforce. Our nation's military veterans and service members are a robust, highly trained population. Increasing the retention and academic success of SVSMs will contribute to a highly skilled engineering workforce, which is essential for addressing national challenges. The project will promote greater awareness of military culture within higher education institutions, bridge gaps between military and academic cultures, and generate new knowledge about the impact of culturally congruent mentoring on the retention of SVSMs. This project will provide a replicable model for improving the retention of engineering students through culturally congruent AI-supported academic assistance and enriching the engineering workforce. The project will investigate the effectiveness of culturally congruent mentoring for SVSMs in engineering education by leveraging AI techniques, including natural language processing, large language models, and machine learning. The project will address fundamental questions about how culturally congruent AI-supported mentoring can increase SVSMs' sense of belonging and improve their academic performance and retention in engineering education. The technical scope involves the design, development, implementation, and assessment of culturally congruent AI-supported mentoring. The AI-supported mentor will be engineered to (1) utilize military-aligned language to clarify complex academic norms, terminology, and structures, (2) provide military-related explanations and examples for engineering tutoring, and (3) help students translate their military technical skills into future engineering career paths. The project will employ a mixed-methods approach, incorporating surveys, interviews, interaction data, academic performance metrics, and retention rates to assess the effectiveness of culturally congruent AI-supported mentoring. The project will measure the impact of culturally congruent mentoring on students’ sense of belonging, academic performance, and retention rates using difference-in-differences, sentiment analysis, and narrative analysis. This project will engage campus veteran support offices, engineering faculty, and student organizations across two institutions to develop mentoring materials and text corpora of military culture, values, language, and experiences of SVSMs. The cross-institutional collaboration will ensure the cultural congruence, reliability, and usability of the AI-supported mentor and facilitate implementation. The project will contribute to foundational research on cultural incongruence in academia, which hinders SVSM’s smooth transition to engineering education. Also, the project will contribute to the growing body of research on AI-supported educational tools by examining whether AI can provide mentorship that is not only personalized but also culturally congruent with the unique military experiences of SVSMs. Dissemination efforts will include practitioner-oriented mentoring toolkits, digital storytelling platforms, and interactive onlin NSF Award ID: 2525266 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Sooin Kim | Institution: Wayne State University, DETROIT, MI | Award Amount: $300,000 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2525266 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2525266.html