HSI Implementation and Evaluation Project: Increasing Computer Science Undergraduate Retention through Predictive Modeling and Early, Personalized Academic Interventions

With support from the Improving Undergraduate STEM Education: Hispanic-Serving Institutions (HSI Program), this Track 2 project aims to increase the current undergraduate retention rates in Computer Science, especially for students from underrepresented/underserved communities. This systematic problem has long prevailed primarily due to its complex nature. Academic retention, especially in demanding STEM fields like Computer Science, is influenced by many factors, including academic preparedness, socioeconomic backgrounds, mental health, and campus support systems. Additionally, students’ diverse experiences and needs make a one-size-fits-all solution ineffective. Institutions need to identify at-risk students early to provide prompt support tailored to students’ needs. In addition, as technology and educational methods continue to develop, it is essential to adjust retention strategies to keep up with these changes constantly. Addressing these challenges requires an innovative, data-driven, and student-centric approach. This project will use predictive modeling to identify students at risk of struggling academically. By identifying students’ potential adversarial factors early, the project will evaluate prompt and personalized evidence-based interventions to support students in overcoming these challenges. The proposed interventions include small group math tutoring, wellness workshops, and peer and faculty mentorship. This comprehensive approach is expected to improve academic performance and well-being among participating students thereby increasing retention rates. The resulting design, implementation, and measured outcomes can guide future interventions to improve student retention, thus increasing diversity in STEM programs. This proposed project has three specific aims. Firstly, it aims to assess the impact of evidence-based interventions on students who are at risk of struggling in key areas such as math, mental health, wellness, and self-efficacy. Machine learning algorithms will identify first-year students at risk of probation based on student-reported academic and demographic data. Interventions will be tailored to address identified risk factors, and a control group will be included for comparison purposes. The effectiveness of each intervention will be measured using quantitative surveys, course grades, and qualitative interviews and analyzed using statistical methods such as ANOVA and regression models. Secondly, using a mixed-method approach that combines quantitative and qualitative analyses, this project will refine and improve interventions and predictive models based on participant feedback on the academic program, vocational interests, and concerns about professional development. Finally, the project will evaluate these interventions' sustainability and broader impact, contributing to developing refined methods for future application. All software artifacts and findings will be open-source and available online. The PIs will present their progress and disseminate results on and off campus through presentations, workshops, and publications at relevant conferences. The success of this project is expected to lead to an expanded adoption of these interventions across other academic programs and institutions, thus improving retention and academic success in STEM fields at HSIs. The HSI Program aims to enhance undergraduate STEM education and build capacity at HSIs. Projects supported by the HSI Program will also generate new knowledge on how to achieve these aims. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.