Wednesday, 17 July 2024

There's something about Wellesley

The gender gap in economics has drawn a lot of attention in recent years (see the list of links at the end of this post), and there are many efforts underway to address the gender gap. Given the pervasiveness of the gender gap, and the difficulties the profession is facing in addressing it, it is interesting to note examples that provide some (maybe any) evidence of substantial success in reducing the gender gap.

So, I was interested to read this new article by Kristin Butcher, Patrick McEwan and Akila Weerapana (all Wellesley College), published in the journal Feminist Economics (ungated earlier version here). They look at all admitted applicants to Wellesley College, a private liberal arts college in Massachusetts, and essentially compare admitted applicants who followed through and enrolled at Wellesley with admitted applicants who did not (and enrolled at some other university or college). That comparison tries to answer the question of, what is the impact of enrolling at Wellesley College? The key outcome variable is whether the students enrol in an economics major, but Butcher et al. also look at enrolment in a number of other majors.

Butcher et al. have a sample of over 15,000 admitted applicants over the period from 1999 to 2013, which they gather graduation data from the National Student Clearinghouse for, and then link to high school data for each student. Their preferred analysis controls for a range of student-level characteristics (including admission board rating) and school-level characteristics. Overall, they find that:

...Wellesley College enrollees are 6.2 percentage points more likely to major in Economics, relative to a comparison-group mean of 7.7 percent. The estimate increases to 7.2 percentage points with a full set of applicant and high school controls.

That is an enormous impact, implying that enrolling at Wellesley College nearly doubles the probability that a student enrols in an economics major. Looking at other majors, Butcher et al. are able to show that students enrolling at Wellesley are not less likely to take STEM majors, so economics is not poaching students from other quantitative disciplines - at least not at Wellesley. There is evidence that enrolments in any other maths-intensive business major (business/managerial economics, finance, or management sciences) and engineering decrease. However, this should not be surprising because neither of these options are offered at Wellesley.

The enrolment effects also show through in other variables, such as:

Only 0.27 percent of non-enrollees complete a graduate degree in Economics, but Wellesley enrollment more than doubles the probability by 0.32 percentage points.

They also find that Wellesley enrollees are more likely to receive a National Science Foundation Graduate Fellowship. However, this is a rare event, and the results are not statistically significant. 

Now, given that all students at Wellesley are female students, the massive increase in economics majors (and subsequently in graduate degrees) is a potentially important finding that is worth exploring further. So, Butcher et al. try to look at what factors might explain this large enrolment effect. They use a decomposition analysis to look at the contribution of different factors to the overall effect. This reveals that 60 percent of the greater enrolment in economics majors is explained by observable 'gender-related and non-gender-related college variables'. Focusing only on the gender-related variables, Butcher et al. note that:

Gender-related college variables include an indicator of women’s colleges, the proportion of women students, and the proportion of women faculty in the Economics Department and college-wide... The variables capture the varied effects of exposure to women classmates, mentors, and instructors. Collectively, the gender-related college variables explain –0.032 of the reduction, which is statistically distinguishable from zero at conventional levels. This is 44 percent of the estimated gap of 7.2 percentage points between Wellesley enrollees and non-enrollees.

The other 16 percent of the effect that is explained by college variables is made up of:

...the public or private status of the college, its focus on undergraduate versus graduate education, and the presence of higher- or lower-achieving peers of any gender, as proxied by math SAT scores...

That still leaves 40 percent of the effect unexplained by observable differences between Wellesley College and other colleges and universities. That is a lot to just be left as the 'unexplained effect of Wellesley College on students enrolling in economics majors'. The disappointing thing is that this is a substantial impact, and we really want to know what is causing it. Butcher et al. offer some speculations, including a greater focus on teaching in tenure and promotion decisions, and an anti-grade-inflation policy that was implemented in 2004. I didn't find those speculations very persuasive (and to be fair, they probably weren't meant to be). This is a case where some additional qualitative work could be used to investigate students' decision-making on major choice at Wellesley, to try and tease out what drives them to choose economics in such greater numbers. That might help other universities to learn from their example.

[HT: Marginal Revolution, last year]

Read more:

Posted by at
Labels:

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)