Patterns of Choice Among Students Who Switch out of STEM Majors

• In Event: STEM Switching Reconsidered: Why Do Undergraduates Leave STEM Majors and Where Do They Go?

Abstract

Numerous studies of undergraduate students use factors such as GPA, test scores, academic performance and demographics to predict persistence in STEM majors (Chen, 2015; Ackerman et al., 2013; Ehrenberg, 2010). Many of these studies examine STEM persistence as a generalized outcome across all STEM majors. However, other studies have sought to make finer distinctions by addressing differences in rates of attrition among disciplines (Chen, 2013) and among types of institution (Griffith, 2010). Fewer studies examine which majors are chosen by those leaving STEM, with the notable exception of Bettinger (2010), who tracks the frequency of different non-STEM destination majors by discipline.

The current study examined how destination majors differ in terms of students’ prio STEM discipline. We wanted to know: 1) Do students in different STEM disciplines switch to particular types of majors at higher rates than expected? 2) What are the primary non-STEM destination majors of students switching out of STEM? and, 3) Do patterns in students’ choice of non-STEM destination majors differ by gender, ethnicity or GPA?

We examine patterns of switching among 2,127 students who switched out of STEM majors at six institutions (four large public, two private). Student majors from semesters directly before and after switching out of STEM were compared to discover general switching patterns. Comparisons were made using broad STEM and non-STEM disciplines. STEM categories included: biology/life science, physical sciences, agricultural sciences, environmental sciences, mathematics, engineering, and computer science. Non-STEM categories were business, psychology, social science (excluding-psychology), humanities, art, foreign languages, education, communication/journalism, and smaller disciplinary categories, such as architecture.

We assessed whether students were more or less likely to switch to specific non-STEM majors through cross-tabulation of all before and after disciplinary categories. Observed frequencies were compared to expected frequencies given the marginal row and column frequency totals for each cell. Adjusted standardized residuals were calculated for each cell and assessed for statistical significance (Sharpe, 2015). A second analysis compared adjusted residuals between gender, students of color, and GPA categories.

Our analysis found significant patterns in students’ choice of destination major based on prior STEM discipline. Biology/life science students were more likely to choose psychology and education majors and less likely to choose majors in business and humanities. Engineering majors switched to business majors in greater numbers than expected; they were also less likely to choose psychology or education. Math majors were much more likely to switch to business than any other category of non-STEM major. Perhaps surprisingly, computer science students switched to art majors and students in agricultural sciences switched into education at higher than expected rates.

Demographic comparisons for major pathways indicate that women were significantly more likely to change from biology to psychology majors than men and less likely to switch to business from mathematics. Students of color were much less likely to switch from engineering to business than white students. Students with low GPAs switching out of engineering were more likely to switch to the humanities, while those with high GPA’s were more likely to switch to business.

Author

• Tim J. Weston, University of Colorado - Boulder