Computational learning analytics that integrate statistical modeling, logistic regression, and natural language processing (NLP) offer scalable approaches for understanding student experiences in engineering education. This study examines first-year engineering students at a Hispanic-Serving Institution (HSI) to investigate attitudinal development and predictors of STEM persistence, a critical factor in improving retention and equity. A multi-phase analytical framework was employed. Paired-samples t-tests showed significant gains in engineering confidence and sense of belonging, with medium-to-large effect sizes. Sequential logistic regression demonstrated that post-course attitudes predict STEM persistence with substantially higher accuracy than demographics or pre-course measures, with engineering value and intent to major as the strongest predictors. NLP analysis of open-ended responses identified math integration as the most dominant and recurring theme across all questions. Sentiment analysis indicated primarily neutral and positive student feedback. Findings consistently show that course-driven attitudinal shifts rather than demographic factors drive persistence. This work contributes to a replicable, data-driven framework that integrates quantitative and qualitative analytics to generate actionable insights for improving first-year engineering experiences, particularly in HSI contexts.
http://orcid.org/0000-0002-9940-4405
University of Texas at El Paso
[biography]
The full paper will be available to logged in and registered conference attendees once the conference starts on June 21, 2026, and to all visitors after the conference ends on June 24, 2026