Examining Precollege Gender Differences and Similarities among Fourth-Year Undergraduate Students, with a Focus on STEM

Email research.policy@act.org for more information. © 2018 by ACT, Inc. All rights reserved.www.act.org/researchThis study provides precollege profiles of fourth-year STEM students based on ACT Composite scores, High School GPAs (HSGPAs), and Interest Inventory scores (shown as Work Task Dimension scores) to estimate gender differences in precollege academic achievement and measured interests.Even after accounting for students’ academic majors, male and female students often differed in their precollege academic achievement levels and measured interests. ACT Research & PolicyTechnical BriefJanuary 2018Examining Precollege Gender Differences and Similarities among Fourth-Year Undergraduate Students, with a Focus on STEMPaul WesTrick, PhDBackgroundStudents’ academic abilities and interests play prominent roles in their choices of academic majors (Allen & Robbins, 2008; Le, Robbins, & Westrick, 2014; Leuwerke, Robbins, Sawyer, & Hovland, 2004). Consequently, evidence of group differences in abilities and interests may help explain group differences in choices of academic majors. Gender differences in students’ choices of academic majors has been well documented, and the underrepresentation of females within science, technology, engineering, and mathematics (STEM) fields has been a subject of national interest (Ceci & Williams, 2007; Ceci, Williams, & Barnett, 2009; Shaw, Kobrin, Patterson, & Mattern, 2012). Male students tend to score higher than females on standardized tests used for admission purposes, but female students tend to earn higher class grades, including math and science courses (Halpern, Benbow, Geary, Gur, Hyde, & Gernsbacher, 2007).One explanation for why males outnumber females in math-intensive majors is that students tend to select academic majors that match their strengths. Many males with high math ability have lower verbal abilities, so this gap becomes important if male students conclude that they are only good at math (Lubinski & Benbow, 2007). In contrast, many females with high math ability also have high verbal abilities (Achter, Lubinski, Benbow, & Eftekhari-Sanjani, 1999; Wai, Lubinski, & Benbow, 2005). This smaller gap between SAT Quantitative and SAT Verbal scores may give females more career options (Ceci et al., 2009), so even if they are academically prepared for STEM work, they are also academically prepared to succeed in non-STEM fields.Another explanation given for the STEM gap is that males tend to prefer working with things and females tend to prefer working with people, which is supported by a recent meta-analysis of vocational interests (Su, Rounds, & Armstrong, 2009). Decades of research by ACT (1995; 2009) has also indicated that fields such as engineering are filled by people whose People/Things work task dimension scores, calculated using respondents’ ACT Interest Inventory scores, are tilted toward things rather than people.Paul Westrick is a research scientist in Statistical and Applied Research specializing in postsecondary outcomes research and validity evidence for the ACT test.R1668 2 ACT Research & Policy Examining Precollege Gender Differences & Similarities among Fourth-Year Undergraduate Students, with a Focus on STEMIn light of these findings, the next logical step would be to examine the differences between males and females within the same academic major, especially STEM majors where females tend to be underrepresented. As there is evidence that academic majors can be differentiated by the average measured interest profile of the students within each major (ACT, 1995; 2009) and their precollege academic achievement profiles of their prospective student members (ACT, 2016; College Board, 2016), persisting male and female students within a given academic major should be quite similar in regard to their precollege academic achievement levels and measured interests. Moreover, if students’ choice of academic major is influenced by their relative academic strengths, within an academic major there should be observable differences in students’ academic achievement levels across high school subject grades and ACT® test scores.The current study aimed to answer the following questions:1. Do male and female students within academic majors differ as much in their precollege academic achievement levels and their measured interests as the overall population of ACT-tested students?2. Are precollege relative academic strengths associated with students’ declared academic major among fourth-year undergraduates?Data and AnalysesData for this study came from 120,612 students who initially enrolled at 26 four-year institutions. To increase the likelihood that students had settled on an academic major and had demonstrated satisfactory academic performance within their areas of study, students had to be enrolled at their institution for eight consecutive semesters. Students were categorized by their two-digit classification of instructional program (CIP) codes in the se