If not a paradox, then what? 7 alternative explanations for the inverse correlation between the Global Gender Gap Index and women’s tertiary degrees in STEM
Artist credit for all images in post: Brianna Weir, 2019
Author: Meredith Reiches
In their 2018 Psychological Sciences paper, Stoet and Geary assert that women in more gender-equal countries are less likely to get advanced degrees in science, technology, engineering, or math (STEM). They claim that this is because, when people are free to choose based on their preferences, they focus on their own relative academic strengths. Most women, they argue, are relatively weaker in STEM than they are in reading. Thus, when they have a choice, they are less likely to choose STEM careers.
Stoet and Geary anchor their argument to a significant negative correlation between two indicators. The first, their measure of gender equality, is the Global Gender Gap Index (GGGI).[i] The GGGI is a composite, nation-level measure of distance from gender parity. Their measure of women’s STEM participation is a ratio which, they claim, represents women’s propensity to complete an advanced degree in STEM.
Calling this measure a propensity encodes an explanation of Stoet and Geary’s findings that their study asserts[ii] but does not test: the differences in women’s and men’s relative abilities in and preferences for STEM fields are evolved, adaptive, and biological. We’ll call this the “women evolved not to prefer STEM” explanation. No alternative explanation for the relationship they demonstrate between the GGGI and their propensity measure is considered.
The goal of this post is to identify the assumptions underlying Stoet and Geary’s evolutionary explanation and to offer some equally strong, if not stronger, explanations that could be tested.
Let’s start with Stoet and Geary’s explanation:
“Women evolved not to prefer STEM”: Women have evolved to be relatively weaker in STEM than in other fields. Therefore, when women are free to choose an academic path, they are less likely to pursue an advanced degree in STEM fields.
None of the data Stoet and Geary use in their analysis concern evolution. Instead, Stoet and Geary extrapolate a fit between their data and the above explanation with the aid of several unacknowledged assumptions. These assumptions preempt alternative explanations for the relationship they find between their chosen measure of gender equality and their chosen measure of women’s tertiary degrees in STEM. Some key assumptions built into their interpretation of evidence include:
Assumption 1: Higher education systems globally offer the same range of options and apply standardized definitions of STEM in a similar manner.
How do countries categorize their degree programs and professions? Health professionals, including physicians, anesthesiologists, and pharmacy technicians, require significant training and competency in science and math. If some nations separate “health services” fields from STEM and others do not, this difference may affect the statistics on women’s representation in STEM. For example, in the United States, many universities do not offer pre-med majors, meaning that students who plan to pursue medicine may major in a STEM field to meet their science requirements. By contrast, some European education systems begin medical training at the undergraduate level. UNESCO uses standardized international definitions of educational fields, meaning that it may categorize these courses of study differently, as the universities themselves do. There may be consequences for how degrees that prepare people for health careers get counted, even if both have significant STEM content.
This effect could create gender bias in the way degrees are categorized. Globally, women are overrepresented relative to men in health professions. According to the United States Bureau of Labor Statistics, 78.4% of workers in health care and social assistance fields were women in 2018. Women get more advanced degrees than men in health services globally: UNESCO data show that, among 94 countries reporting graduates of tertiary level health and welfare programs in 2016, women made up more than 50% in all but six countries. Women earned more than two thirds of these degrees in 73 of the 94 countries.
Furthermore, within disciplines categorized as STEM, patterns of sex segregation vary from country to country. For instance, Charles and Bradley (2009) find that, in 1994, women were underrepresented in engineering, math, and the natural sciences in Finland. In Italy, however, they were strongly represented in math and the natural sciences but underrepresented in engineering.[iii] If you’re interested in women’s representation in STEM, the specific fields you pick can tell different stories--stories that may say more about the climate of a particular field in a particular location than about innate abilities or preferences. Finally, differences in representation across STEM fields can lead to a country scoring high on women’s representation in STEM globally, while in fact women are over-represented in some STEM fields but underrepresented in others.
Assumption 2: Low GGGI countries are low-income countries:
Stoet and Geary assume that women should be motivated to get advanced STEM degrees in low income countries because STEM degrees make possible a more secure economic future. Do countries with higher incomes always have higher GGGIs? While there is an overall positive relationship between GGGI and gross domestic product (GDP), there are exceptions. Take Rwanda, a high GGGI country: ranked 6th from the top out of 149 countries in the 2018 Global Gender Gap Report, it has a per capita GDP of only $1,854. The Russian Federation, meanwhile, with a higher per capita GDP of $24,766, ranks 75th in the GGGI.
Assumption 3: STEM degrees pay:
Does getting a STEM degree actually lead to greater economic security? Where women’s labor market participation is low, or where having a degree doesn’t lead to substantial remuneration, having a STEM degree may not pay. This may be the case in Italy, where university faculty, including those in STEM fields, make very low salaries. Might the same be true in other countries?
If you’ve been following along with the GenderSci Lab’s Commentary and blog series about Stoet and Geary’s 2018 article, you’ll know that we--and other scholars --have a lot of questions about what exactly Stoet and Geary are measuring and how they’re interpreting their results. The measures they used for gender equality and for women’s participation in STEM fields don’t necessarily capture either gender equality or women’s participation in STEM fields. The GGGI was not designed to be used in the kind of analysis that Stoet and Geary conduct. In the case of women’s tertiary degrees in STEM, as their Corrigendum demonstrates, Stoet and Geary did not measure women’s share of tertiary degrees in STEM, as they claimed. The analysis treats countries as independent, failing to take into account historical and cultural relationships among them that could affect the likelihood that women go into STEM fields. In fact, culture and history are absent from the hypothesis. Gender equality as indexed by the GGGI is treated as synonymous with conditions that empower women to pursue STEM education.
It could be, as we suggest in previous posts, that the correlation observed by Stoet and Geary is entirely spurious. Another possibility is that there does exist some type of causal relationship between measures of nation-level gender equality and women’s representation in STEM, but the drivers of that relationship are not women’s and men’s relative academic strengths or innate, evolved preferences. Rather, something else historical or cultural lurks behind the association, as researchers including Maria Charles and Karen Bradley have suggested. While the goal of previous posts has been to unpack conceptual and methodological challenges to Stoet and Geary’s analysis and interpretation, the aim here is less a “take down” than a “build up”: we want to think about what interpretations, other than Stoet and Geary’s assertion that “women evolved not to prefer STEM,” are possible, unexamined, in principle testable, and compatible with prior research. Why accept Stoet and Geary’s explanation without considering and testing alternatives? (The focus here is on Stoet and Geary (2018), but the same analytical playbook can be used to evaluate similar assertions of a paradox in the relationship between women’s and men’s equal opportunities and unequal outcomes on measures of personality or achievement.[iv])
Seven alternative explanations for the inverse correlation between the Global Gender Gap Index and women’s tertiary degrees in STEM
1. “People take gendered career paths in postindustrial service economies, where STEM is associated with men”
Let’s start with the leading alternative explanation on offer. It is articulated by sociologists like Maria Charles and not cited by Stoet and Geary. In post-industrial societies, characterized by histories of prosperity and predominantly service economies,[v] more people have access to higher education. The number and type of educational programs and career paths multiply. Under these labor and educational conditions, two cultural forces predispose men and women to diverge in their educational and career paths.
The first is gender-essentialist ideology, the idea that men and women are immutably different in characteristics like their aptitudes and preferences. Gender essentialist ideology is alive and well in prosperous countries.[vi] It persists even when the state approaches gender equity on measures of access to education and labor force participation.
Second, if immediate worries about subsistence aren’t pressing, people are encouraged to make educational and career choices that express their preferences. Historically, certain educational fields and career pathways were perceived as appropriate for and appealing to men, while others were seen as playing to the strengths and preferences of women. The history of higher education in prosperous countries reveals how such perceptions are built into the educational system, reinforcing these suppositions. In order to encourage women’s participation in higher education, certain programs were designed to target women, educating them to “make use of their abilities for the greatest good of society” and permitting them “to specialize in fields particularly suited to feminine aptitudes”.[vii]
Putting these two factors together, when people are incentivized throughout development to identify with their gender, they’re likely to choose educational programs and occupations consistent with their society’s understanding of what a person of their gender ought to want. This reinforces the system of stereotypes and aspirations described above.
In postindustrial societies, therefore, this theory predicts that women would be less likely to enroll in STEM degree programs not because they are biologically wired to be better at or to prefer other things, but because they identify with femininity. Femininity, in these societies, is not associated with STEM fields.[viii] Women are therefore less likely to land in STEM educational programs and occupations.
2. “Cultures vary in stereotypes related to gender and STEM ability”:
The idea that men are more inclined toward or better suited to STEM education and careers is not universal. The same countries that boast high GGGIs today may in fact have histories of associating STEM fields with men that are not shared by lower GGGI countries.
Overall, indices of gender equality are not strong predictors of contemporary attitudes towards women in STEM. Harvard researchers Adam Mastroianni and Dakota McCoy found no association between national indices of gender equality and implicit bias against women in science, meaning that there is no evidence that more gender equal countries have less bias towards women in science. In fact, research has demonstrated that countries like Sweden and Norway have higher scores of implicit bias against women in science than countries like Iran or Jordan.The level of bias predicted the involvement of women in science. Assuming that all nations share a common sex/gender role system and history--and that wealthier nations are globally more supportive of women in STEM-- is not only insulting but also inaccurate.
We can approach variation in the relationship between gender equality and attitudes towards women in STEM with curiosity, asking: what is the history of cultural beliefs about women’s suitability for science in countries like Algeria and Iran, where a high proportion of women’s advanced degrees are in STEM? Are these histories different in systematic ways from those of countries like Norway and South Africa, where a low proportion of women’s advanced degrees are in STEM? Intriguing data come from 2003, 2007, and 2011 surveys of eighth graders’ enjoyment of math. They find that girls report enjoying math more than boys in countries including Armenia, Botswana, the Czech Republic, Indonesia, Latvia, Malaysia, the Philippines, Romania, Russia, Serbia, Singapore, the Slovak Republic, Thailand, and Turkey.[ix] These data suggest that formative attitudes towards STEM--attitudes that can affect education and career options--vary globally.
3. “In some countries, most degree programs are in STEM”:
The UNESCO data on the distribution of advanced STEM degrees by sex tell us nothing about what degree programs are available in each country. If the majority of programs available in a given nation are in STEM, then people of all genders pursuing tertiary degrees in that nation should be more likely to enroll in STEM programs. In 2018, UNESCO data indicate that 40.77% of tertiary degrees awarded in Malaysia were in STEM disciplines, while only 11.23% of tertiary degrees were in STEM in Bangladesh. This is not a direct measure of program availability, but it suggests that STEM programs might be more prevalent in some countries than in others.
If, however, a nation gives students more diverse educational options, then cultural ideas about what kinds of learning and work men and women find fulfilling and are suited for may affect which degree programs they choose. Given this logic, it’s worth investigating whether there’s a relationship between a country’s GGGI or its wealth and the proportion of its advanced degree programs that are in STEM fields.
4. “Gender-biased travel to pursue STEM degrees changes the ratio of degrees awarded to men and women in countries of origin and destination”:
UNESCO data tell us how many students are awarded STEM degrees in a given country. They don’t tell us whether those students are citizens of that country, and they don’t tell us how many citizens go abroad to get degrees. If there are systematic and gendered patterns of migration for education, within-country data on who is getting STEM degrees could be affected. For example, if men from lower GGGI nations are more likely to go abroad for STEM degrees, then they will be artificially inflating men’s representation in STEM in their destination countries and inflating women’s representation in STEM in their home countries. Take the United States as a case study. Its National Science Foundation reported that, over the five year period from 2011 to 2015, 86% of PhDs awarded to non-citizens were in science and engineering, indicating that STEM degrees represented most of the degrees earned by students from outside the United States. Students from China, India, and South Korea accounted for more than half of PhDs awarded to temporary visa holders. The gender of degree recipients with temporary visas was not reported. However, the fact that high numbers of STEM candidates from some nations leave their countries of origin and get degrees abroad may tip the gender balance of STEM representation in both their countries of origin and destination.
5. “Backlash against women’s economic and political gains includes hostility to women in STEM”:
In countries where women make gains in economic participation and political representation, social science researchers have long documented resistance on the part of men who feel that their authority and power are compromised.[x] Among European countries, despite having some of the highest GEI scores, the Nordic countries are also known to have high rates of domestic violence (the so-called Nordic Paradox). No one is suggesting that this represents an innate preference for domestic violence revealed in high equity societies. Instead, research has examined the unique cultural circumstances shared among Nordic countries and the possibility of a cultural backlash in response to increased equity in other areas. This backlash, in the Nordic countries and elsewhere, can play out in private space (e.g,. in pressure to be unhealthily thin), but it can also affect the calculations women make about their professional lives. Might this include the choice to pursue an advanced degree in STEM in places where STEM is understood as masculine, or where STEM educational and professional climates are likely to be hostile to women?
6. “Attitudes towards women in STEM cannot be predicted by attitudes towards women in other public spheres”:
Is it possible that attitudes about women in STEM cannot be predicted by data about women in the workplace, in politics, in sports, or in other arenas? The fact that STEM representation seems to behave differently--that is, it produces different country rankings than markers of educational, labor force, and political participation--indicates that it could be driven by different factors. Rather than concluding that STEM participation and gender equality are negatively related, perhaps the conclusion should be that STEM achievement is its own domain, which needs to be added to measures such as the GGGI.
7. “Women in low GGGI countries represent the exception, not their country’s prevalent attitude towards women in STEM”:
In low GGGI countries with small, urban higher education systems, the students who attend those institutions disproportionately come from privileged classes with substantially different gender politics and norms for women’s education than the rest of the country. (Even in the United States, primary and secondary access to STEM education is stratified by social class.[xi]) Women in these classes pursue education at much higher rates than the rest of the country and, as elites, are empowered to resist gender norms that apply to other women.[xii] As educational opportunities expand to members of less privileged groups, more women may choose educational and career paths consonant with cultural ideas about gendered preferences and aptitudes.
In short, there are a lot of reasons why the GGGI might be negatively associated with women’s representation in STEM that have nothing to do with women’s evolved natures, how strong women are in STEM, or how much, all else being equal, they prefer it. Stoet and Geary allude to a single explanation in interpreting their results: they go right for evolved sex differences in STEM abilities and preferences. They do not consider the universe of possible explanations, including those already published and in circulation. A true evaluation of their explanation requires examining assumptions built into their treatment of the evidence, evaluating alternatives, and answering objections. Stoet and Geary have not established that a larger gender gap in STEM in more gender-equal countries is evidence that STEM abilities and preferences come with an X or a Y chromosome.
As we weigh their claim against alternative explanations, we bear in mind that we are in a moment of transformative change and enormous potential for women and gender minorities in STEM professions. The pronouncements of science are not merely academic; they impact the world view, life prospects, and aspirations of all people. We have a responsibility to hold scientific claims for innate difference to rigorous intellectual and ethical standards.
Authorship Statement:
This blog series on the Gender Equality Paradox emerged from collective GenderSci Lab discussions. Each author outlined and drafted their own piece. GenderSci Lab members offered comments and authors integrated these revisions. Brianna Weir developed original artwork for the series. Maria Charles authored and approved the final version of her interview answers and provided images and figures for our use. Tyler Vigen developed a “women in STEM” spurious correlations widget for us and provided permission for the use of his findings in this blog series. Juanis Becerra and Nicole Noll assisted with formatting the blogs for the website. Heather Shattuck-Heidorn oversaw the blog series development, review, and publishing process. For the Psychological Science paper, Sarah Richardson drafted the manuscript. Meredith Reiches and Joe Bruch performed the data analysis. All authors (Richardson, Reiches, Bruch, Boulicault, Noll, and Shattuck-Heidorn) provided critical revisions and approved the final version of the manuscript for submission. Action editor Tim Pleskac shepherded the Corrigendum and Commentary through the peer review process at Psychological Science. We thank the anonymous peer reviewers and Gijsbert Stoet and David Geary for their contributions.
Recommended Citation:
Reiches, Meredith. “If not a paradox, then what? 7 alternative explanations for the inverse correlation between the Global Gender Gap Index and women’s tertiary degrees in STEM,” GenderSci Blog, February 15, 2020, https://www.genderscilab.org/blog/if-not-a-paradox-then-what-7-alternative-explanations-for-the-inverse-correlation-between-the-global-gender-gap-index-and-womens-tertiary-degrees-in-stem
Endnotes:
[i] World Economic Forum. (2015). The Global Gender Gap Report 2015. Retrieved from http://reports.weforum.org/global-gender-gap-report-2015/
[ii] This assertion is in their citations supportive of evolved sex differences--e.g. Pinker, S. (2008) The sexual paradox: Men, women and the real gender gap. New York, NY: Simon & Schuster--and in their conclusion: “In closing, we are not arguing that sex differences in academic strengths or wider economic and life-risk issues are the only factors that influence the sex difference in the STEM pipeline. We are confirming the importance of the former (Wang et al., 2013) and showing that the extent to which these sex differences manifest varies consistently with wider social factors, including gender equality and life satisfaction.” The fact of sex differences is assumed. What varies is the extent to which they manifest.
[iii] Charles, M., & Bradley, K. (2009). Indulging Our Gendered Selves? Sex Segregation by Field of Study in 44 Countries. American Journal of Sociology, 114(4), 924–976, 942.
[iv] E.g., Falk, A., & Hermle, J. (2018). Relationship of gender differences in preferences to economic development and gender equality. Science, 362(6412) and Giolla, E. M., & Kajonius, P. J. (2018). Sex differences in personality are larger in gender equal countries: Replicating and extending a surprising finding. International Journal of Psychology.
[v] Bell, D. (1976). The coming of post-industrial society. New York: Basic Books.
[vi] Schmitt, D. (2012) When the difference is in the details: A Critique of Zentner and Mitura (2012) ―Stepping out of the Caveman’s Shadow: Nations’ Gender Gap Predicts Degree of Sex Differentiation in Mate Preferences. Evolutionary Psychology. 10(4):720-726.
[vii] UNESCO (United Nations Educational, Scientific, and Cultural Organization). (1953). Women and Education. Paris: UNESCO.
[viii] Charles, M., & Bradley, K. (2009).
[ix] Charles, M., B. Harr, E. Cech, A. Hendley (2014) Who likes math where? Gender differences in eighth-graders’ attitudes around the world. International Studies in Sociology of Education 24(1):85-112.
[x] E.g., Faludi, S. (2006). Backlash: The Undeclared War Against American Women. New York: Three Rivers Press.
[xi] Oakes et al. (1990) Multiplying Inequalities: The Effects of Race, Social Class, and Tracking on Opportunities to Learn. National Science Foundation Report NSF-R-3982. Santa Monica, CA: Rand Corp.
[xii] Fave, L. R. D. (1980). The Meek Shall Not Inherit the Earth: Self-Evaluation and the Legitimacy of Stratification. American Sociological Review, 45(6), 955; Gecas, V. (1991). The self-concept as a basis for a theory of motivation. In J. A. Howard and P. L. Callero (Ed.), The Self-Society Dynamic: Cognition, Emotion, and Action (pp. 171–187). New York, NY: Cambridge University Press.