Demographic Trends in the Philosophy Major Might Be Mostly Due to Pre-College Factors (guest post)


This guest post* looks at two questions related to demographic trends among philosophy majors. First, are women disproportionately less interested in the philosophy major at the beginning of their first year of study? And second, is the recent apparent increase in interest in philosophy reflected in first-year intention to major? 

Authored by Eric Schwitzgebel (UC Riverside), Morgan Thompson (Pittsburgh), and Eric Winsberg (South Florida), the post first appeared at The Splintered Mind.

Félix Armand Heullant, “Edle junge Dame im Salon”

Demographic Trends in the Philosophy Major Might Be Mostly Due to Pre-College Factors
by Eric Schwitzgebel, Morgan Thompson, and Eric Winsberg

As we mentioned last month, we recently obtained data from the Higher Education Research Institute (HERI) on intention to major in philosophy among first-year students in the U.S.

Today we will explore two questions.

First, it’s well known that undergraduate philosophy majors in the U.S. are disproportionately men. For example, recent data from the National Center for Education Statistics show 36% of graduating Philosophy majors in the U.S. to be women, compared to 57% of graduating majors overall. Our first question is this: Are women also disproportionately less interested in the Philosophy major at the beginning of their first year of study?

The answer to this question is crucial to understanding the causes of the low proportion of women among graduating philosophy majors. If women begin their studies with less interest in philosophy than undergraduates as a whole, then the causes of disproportion trace back to something prior to college enrollment. In contrast, if women begin their studies with approximately proportionate interest in the Philosophy major, then their underrepresentation among Bachelor’s recipients in Philosophy suggests that something in students’ college experience is driving the disproportion.

Second, as Eric Schwitzgebel noticed last fall, the Philosophy major seems to be back on the rise in popularity while other humanities majors continue to fall. We wanted to see if the recent apparent increase in interest in Philosophy was also reflected in first-year intention to major. This is relevant to evaluating both the causes of and the likely persistence of the trend that Eric S. noticed last year.

First-Year Intention to Major by Sex, 2000-2016

Every fall, HERI gathers information from first-year undergraduates at a sample of U.S. colleges and universities, with about 200,000-400,000 respondents per year. One question asks respondents’ sex, with response categories “male” and “female”. About half of one percent of respondents decline to state. Another question asks for intended major, with “Philosophy” as one among dozens of choices.

This graph shows the percentage answering “female” among first-year students, both overall and in Philosophy, excluding students who declined to state.

As you can see from the figure, first-year student respondents were about 56%-59% female across all majors throughout the period (54%-55% if nonresponse bias is taken into account; see below). From 2000-2012, 32% to 36% of first-year student respondents intending to major in philosophy were female. This compares with about 30-34% of women among graduating majors in Philosophy in the same period. Thus, female students appear to be disproportionately less interested in the Philosophy major from the beginning of their undergraduate studies. These results match with some earlier analyses of the HERI database by Christopher Dobbs and Philippe Lemoine.

There may be some further loss of interest among women—about 2% in absolute percentage terms (32-36% vs 30-34%)—between first year intention to major and completion of the major, but due to differences in methodology between HERI and NCES it’s difficult to be confident about effects of this size, and we note that “female” and “woman”, though approximately comparable, are not identical categories.[1]

The second striking feature of this graph is the recent increase in percentage of respondents intending to major in Philosophy who reported being female: 40%-43% in 2013-2016. This suggests that the increased percentage of women among Philosophy BA recipients that appeared in the NCES data from 2018, which we noticed last fall, may not be a blip but might be the beginning of a trend that showed up in first-year students in 2013. In fact, the timing is perfect. With a national average of five years to Bachelor’s degree, a change in first-year students in the 2013-2014 academic year should be reflected in a change in graduating majors in the 2017-2018 academic year.

The change could be explained either by an increase in female students’ interest in the Philosophy major or a decrease in male students’ interest or both. This is a slightly complicated question which will first require us to address changes over time in the Philosophy major in general.

One big methodological caveat here is that the HERI data have some nonresponse and sampling problems: Not all colleges are included, with lower prestige public colleges especially undersampled, and not all students respond, and this skews the HERI demographic data.[1] Furthermore, the number of participating colleges declined substantially over the period in question. Some preliminary analyses we’ve tried suggest that nonresponse and over/undersampling might be an especially big issue with student race (which we hope to analyze in a future post), but only a minor issue with sex.

HERI provides researchers with a calculated variable “Student Weight”, which represents their best attempt to overweight the responses of students from underrepresented portions of the sample and underweight the responses of students from overrepresented portions of the sample, with the hope that the weighted responses are representative of first-year students in the U.S. as a whole. (The NCES data, in contrast, are reported by administrators and are approximately complete.)

The results above are based on raw responses. We attempted to correct for sampling and nonresponse bias by multiplying all responses by HERI’s Student Weight variable, but statistical noise became a problem. For example, using this method, estimates of the percentage of philosophy majors who were female jumped implausibly from 27% to 37% from 2013 to 2014. Since the Student Weight variable weights some students’ responses several times more than others, it should be expected to amplify noise, and given the small numbers of female philosophy major respondents (207 in 2013), it’s unsurprising that noise might be a limiting factor.

Overall, all trends reported in this post are confirmed when data are weighted by HERI’s Student Weight variable. However, the percentage of philosophy majors overall might actually be somewhat lower than reported (due to disproportionate representation of elite schools, where Philosophy is more commonly chosen as a major) and the percentage of female students might be slightly lower (due to slightly higher response rates among female students at the included schools).

While History and English Continue to Fall, Philosophy Has Partly Recovered

In 2017, Eric S. noted sharp declines in completed Philosophy, History, and Language majors in the NCES database, followed the next year by a slight recovery or stabilization in Philosophy, while the other big humanities majors continued to decline.

We were curious to see if this would also reflected in the HERI data on first year intention to major. As with the data on sex, examination of the HERI patterns could give us insight into mechanisms (are these changes due to something happening before college or in college?) and also perhaps some basis for projection into the future.

This chart shows rise and decline in intention to major, normed to the year 2000.

As you can see, the percentage of students majoring in History and English is about 2/3 of what it was in 2000. Philosophy showed an equally sharp decline in the early 2010s but seems to have partly recovered and is now at 86% of 2000 levels, while History and English continue to fall. As with gender, the timing shows a nice offset between HERI and NCES: The decline in first-year intention to major started in about 2010, while in the decline in completed Bachelor’s degrees started in about 2014 or 2015.

As with sex, the timing offset and similar pattern in the HERI and NCES data suggest that the primary factors behind these demographic trends are pre-college.

The decline and partial recovery of interest in the philosophy major interacts with sex, as shown in this figure:

As you can see, the percentage of female first-year students’ intending to major in Philosophy has recovered fully to 2000 levels, but not so for male first-year students.

We conclude that those of us who are interested in exploring the causes of demographic trends in the philosophy major should look more carefully than is usually done at factors that might be influencing students’ perceptions and intentions even before they enroll in college.


notes

[1] Most non-women in the undergraduate population are men, but a small percentage will be non-binary. We are unaware of any good data source on the rates at which non-binary students choose to major in philosophy. This dataset from HERI does not ask for gender, so it is possible that many respondents are answering with gender rather than sex. HERI’s Freshman Survey did not revise the question to be about one’s gender identity until 2018 and did not add a question about whether the student is trans or cis until 2019. Unfortunately, we were unable to access those data due to temporary embargoes on more recent years’ data.

[2] Unlike the NCES data, which is reported to the U.S. government by adminstrators at each institution, HERI collects data by selling U.S. universities and colleges the results of their survey for that particular institution. Wealthier institutions appear to be more likely to pay for this data collection and thus more likely to be represented in the HERI Freshman survey dataset than lower prestige colleges. The “Student Weight” variable discussed below is partly intended to help correct for demographic differences between wealthier, higher-prestige institutions and lower prestige public colleges.


There are 12 comments

Your email address will not be published. Required fields are marked *

  
Please enter an e-mail address