복수전공은 누가, 어떻게 선택하는가? 젠더, 사회경제적 지위, 제도적 자원의 상호작용 = Who chooses to double major and how? Interactions between gender, socioeconomic status, and institutional resources

다국어 초록 (Multilingual Abstract)

In recent years, there has been a growing interest in, and need for, various programs that allow students to select additional majors during their undergraduate careers in addition to the one they declared at the time they entered college. Existing studies have mostly analyzed the effectiveness of these programs, but few studies have examined how students come to choose a major beyond a single major. Using data from the Graduate Occupational Mobility Survey and sequential logit models, we examine how the decision to double major and choose a high-reward major varies by gender, family socioeconomic status, and prior achievement status (original major and college characteristics). Our findings are fourfold. First, children of highly educated parents and those who attended selective colleges were more likely to double major. Second, while non-STEM majors were more likely to double major, the likelihood of choosing a high-reward major when double majoring was higher for STEM majors and high-reward majors. Third, overall, family capital (parental education, household income) and educational capital (selective college) helped students use double majoring more aggressively to compensate for an unfavorable primary major. Fourth, women are more likely to double major, and this is especially true for women in STEM majors. On the other hand, women in STEM majors at selective universities are significantly less likely to choose a high-reward major. These results suggest that the choice of a double major is stratified by the motivation to compensate for pre-existing disadvantage, with better family capital and educational capital contributing significantly to inequality. The results also raise the possibility that double majoring is used as an exit route for women in STEM majors at selective universities into non-STEM majors with lower expected rewards.