Changing course: the gender gap in college selectivity and opportunities to learn in the high school curriculum

Abstract

Gender gaps in learning and education outcomes have changed dramatically over the last few years. However, researchers have not adequately assessed how the high school learning environment differentially affects boys and girls. An important component of the learning environment in US secondary school is the opportunity to learn in an Advanced Placement (AP) curriculum, which allows high school students to do college-level work. Using the US National Education Longitudinal Study 1988–2000, we explain how high school AP curriculum interacts with gender to predict the selectivity of colleges that students attend. The results show that girls and boys who attend high schools with a larger percentage of students in AP curriculum attend more selective colleges (that require higher standardised scores for admissions); yet the positive effect of the opportunity to learn in an AP curriculum is greater for girls than for boys. This research furthers the debate about the effects of school structure on gender stratification.

Keywords:

• sociology
• secondary education
• higher education
• girls

Acknowledgements

The authors acknowledge the helpful comments of Roslyn Mickelson.

Notes

1 We do not focus on IB tests because they are less widely utilised across schools.

2 Of course, some scholars question the long-term implications of attending selective colleges (Black and Smith 2004; Brand and Halaby 2006).

3 It is important to note that while concerns over access to college are minimal, there are still major concerns over college major. Boys are still substantially more likely than girls to pursue science and math-oriented majors. This paper does not address that topic.

4 Feminists would also challenge the supposition that gender is no longer relevant to academic success because recent studies have found that within schools traditional gender relations emerge in peer groups, even if simultaneously accompanied by more egalitarian norms (Reay 2001). Furthermore, even if schools provide equal opportunities, girls might be actively discouraged from acting on these opportunities by the actions of boys looking to maintain the status quo.

5 It is important to note that the exclusion of specialty colleges and foreign colleges means that our results are not generalizable to the population of students attending those colleges.

6 The number of students who took AP tests is also not available in the NELS database.

7 If grades from one of these categories were missing, we took the average of the three remaining categories.

8 We utilise data on achievement tests rather than the scholastic aptitude test because we lose three times as many cases when we utilise SAT data. Furthermore, SAT data introduce a form of selection bias because students not planning to attend college do not usually take these tests. This would lower our sample. Thus, following Blau, Moller, and Jones (2003) we rely on achievement data.

9 These covariates allow us to control for Type A effects associated with school choice (i.e. school context; Raudenbush and Willms 1995).

10 Rigorous academic track reflects the kind of courses that the student took in school. Within schools in the USA, students can take very different courses. Some students take remedial courses, some take courses that focus on trades, such as cosmetology, and some take more rigorous, college-bound courses, such as AP courses and honours courses.

11 In separate analyses (not shown), we include an interaction between SES and gender. We find that the interaction is not significant. The differences between boys and girls in college selectivity do not vary by parental SES.

12 In separate analyses (not shown), we test an interaction between gender and whether the student takes AP classes. The interaction term is not significant. Girls who take AP courses are not more likely to attend moderately selective colleges than boys who take AP courses. In addition, there is no difference in the type of college attended between girls who do and do not take AP courses, net of other factors. Thus, the effect of per cent AP is not simply an artifact of an individual's course load. This supports, or at least does not contradict, our proposition that per cent AP reflects broader opportunities to learn in a rigorous curriculum.