A national shortage of workers in Science, Technology, Engineering, and Math (STEM) occupations has led to efforts to identify why people leave these fields. Lower persistence rates in STEM for females than for males have also led to examinations of features that cause females to leave STEM fields. The current study examines individual- and school-level features that influence undergraduate students' decisions to leave STEM majors, focusing on potential explanations for why females are more likely than males to leave. Persistence in STEM was examined in three samples: (a) persistence through the second year of college in a sample of high school seniors interested in STEM majors; (b) persistence through the fourth year of college in a sample of second year undergraduate STEM majors; and (c) persistence through the second, third, and fourth years of college in a sample of high school seniors interested in STEM majors. Differences between persistence in male-dominated and non-male-dominated STEM majors were also examined. In all samples, gender differences were found for most individual-level predictors, with males tending to score higher than females on measures such as SAT-Math, self-rated STEM ability, and high school extracurricular activities and awards in STEM. On the other hand, females earned better high school grades and had stronger relative non-STEM ability and achievement than males. Bivariate analyses indicated that those who persisted in STEM majors typically had higher scores than those who did not persist for SAT-Math, high school achievement, STEM course taking, undergraduate STEM grades, self-rated STEM ability, interest in STEM, extracurricular activities and awards in STEM, degree goals, and socioeconomic status. Multivariate analyses identified SAT-Math as one of the best predictors of persistence in high school samples, and undergraduate STEM GPA was one of the best predictors in the samples of second year undergraduates. In several samples, a significant cross-level interaction was found between gender and undergraduate females' college-level proportional representation in STEM; however, the effects were inconsistent across samples. Even when controlling for various individual- and school-level predictors, gender effects tended to remain significant, with females in most samples leaving STEM majors at higher rates than males.
University of Minnesota Ph.D. dissertation. August 2013. Advisor: Paul Sackett. Major: Psychology. 1 computer file (PDF); xxi, 318 pages.
Koch, Amanda Joy.
Predicting Undergraduates' Persistence in Science, Technology, Engineering, and Math fields.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.