Predictors of first-year retention among undergraduate engineering students who earn a c in their first-semester math course

Abstract

This Complete Research Paper examines non-cognitive predictors of first-year engineering retention for students who received a C in their first semester mathematics course at the University of Louisville. Scores across eight non-cognitive measures served as model predictors, obtained at the beginning of the first year, including: value interest in engineering, perceived effort, opportunity, and psychological costs, perceived belonging uncertainty, contingencies of self-worth: academic competence, test anxiety, and self-efficacy. Using least absolute shrinkage and selection operator regression, we found that value interest and test anxiety were the strongest predictors of C-student retention. The results from this study inform research on the decision-making of students that have potential for graduation but are at risk of leaving engineering. Our results indicate that a strategic intervention in increasing interest in engineering may lead to strong gains in engineering retention at this university, and potentially others as well.