Institutional Context Matters: Linking Characteristics of Universities to the Gender Composition of Engineering and Computer Science Programs

Le résumé fourni par la source

Our research focuses on assessing how institutional factors shape the gendered composition of engineering and computer science degrees.We use data from the Integrated Postsecondary Education Data System (IPEDS) to identify how institutional-level (rather than individuallevel) parameters shape this outcome.The IPEDS is an annual survey of all U.S. postsecondary educational institutions and contains data on institutional characteristics such as student body diversity (e.g., race/ethnicity), institutional selectivity (e.g., SAT, Pell grant recipients), Carnegie research classifications, student-to-faculty ratio and institutional size.We analyze completion rates by gender for a sample of four-year institutions (N=525), specifically those with more than 5,000 students awarding degrees in at least one of 19 computer science and engineering programs.Our sample mirrors widely reported nationallevel trends: Women comprise approximately 16 percent of degree earners in computer science and 18 percent in engineering programs.Because our outcome variables are measured as proportions, we use ordinary least squares (multivariate) regression and employ multiple imputation using chained equations (MICE) to account for missing data.Analyses show that institutional characteristics are associated with gendered completion rates in both fields.In computer science, a higher student-to-faculty ratio is associated with completion rates that exacerbate the underrepresentation of women, whereas HBCU's and institutions with a higher proportion of African American students play a key role in boosting the representation of women in computer science.In contrast, for engineering programs, we find that private and highly selective institutions exacerbate the underrepresentation of women.For both fields, the proportion of Hispanic students is associated with boosting the representation of women, whereas women remain particularly underrepresented at institutions with a higher proportion of Pell-grant funded students.We situate this combination of findings in extant research suggesting that postsecondary educational institutions (or at least the computer science and engineering units) constitute an example of gendered organizations.We then discuss seemingly unintended consequences of diversity and inclusion efforts and outline potential implications for STEM recruitment and retention, in an effort to inform meaningful interventions that can advance women's representation in these two male-dominated fields.We conclude with suggestions for future research.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.