By Rumman Chowdhury
[Correction: a previous version of this post stated that Danica McKellar has a PhD in Mathematics. Ms. McKellar has a BS in Mathematics from UCLA.]
The following post is written by Rumman Chowdhury, a graduate student in political science at the University of California at San Diego. Chowdhury received an undergraduate degree in political science and management science from MIT, and a masters degree in quantitative methods in social sciences from Columbia University.
Graduate students, regardless of field of study, gender, race, or any other distinguishing characteristic, generally feel isolated from their peers, petrified of passing comprehensive exams and dissertation defenses, and intimidated by the academic job market looming ahead. The insecurity of being in a PhD program is compounded by being a minority. Women, as the previous posts have established, are a minority within political science, and especially within methods.
It is daunting to discuss why there are not enough women in quantitative fields. It requires identifying macro-level societal issues, honing in on micro-level individual behaviors, but couching it in the framework of a larger discussion of what it means to be a woman and have a successful career. In an attempt to simplify, I focus on two selection mechanisms which may pull women away from successful and rewarding careers in quantitative fields: selection due to early-development socialization and selection due to gender-differential social pressures.
Early development socialization
A significant amount of work has explored the gender and math question. In short, the literature illustrates a weeding-out process, whereby girls grow up in a society that explicitly and implicitly deters them from quantitative fields. We inadvertently groom your young girls to give up when faced with difficult problems, and compound that with messages that math isn’t for them.
Psychologist Carol Dweck’s Mindset establishes two types of learners: fixed and growth. Individuals of a fixed mindset believe that qualities like intelligence or ability are innate traits that can be refined, but not significantly improved on. In contrast, those with a growth mindset believe that most skills are a function of hard work and dedication, not simply talent. These mindsets are an important and distinguishing characteristic when the individual is faced with difficulties. Those with a fixed mindset are more averse to challenges, as they fear failure will define them as “not smart,” whereas those with a growth mindset are more capable of dealing with failure, as it is perceived as part of the learning process. While this mindset is seen in both genders, women are particularly susceptible to adopting a fixed mindset in math. [1] This causal mechanism has three parts.
At an early age, young girls are more developmentally advanced than young boys. Their brains develop language faster which results in young girls who are more expressive than their male counterparts [2]. This linguistic ability can be attributed to intelligence. Accordingly we provide results-based compliments to our precocious little girls, saying things like “You’re so clever!” for actions that come naturally to them. Parents also offer words of encouragement to their children who are not as capable, and little boys may hear action-based compliments, like “You’re working really hard on that!” Dweck establishes that this form of reinforcement creates a fixed mindset. Paradoxically, more capable children can internalize that intelligence is intrinsic while their less able counterparts associate ability with hard work.
The second part of the mechanism that leads young girls to believe that they are not good at math is the tired, but still extant, stereotype that math is for boys. [3] [4] Cevencek et al observe this math-gender stereotype expressed in Implicit Association Test results for children ages 6 to 10. Children associated math with male, and little boys identified more strongly with math than little girls.
The literature on mindsets is particularly salient with regard to math and science. A two-year panel study of 7th graders found that children with growth mindsets significantly outperformed children with fixed mindsets, even though they both entered the analysis period with equal prior math achievement. Growth mindsets predicted success in college-level organic chemistry, when controlling for prior math ability. Similarly, women primed with a fixed mindset treatment performed significantly worse than women provided with a growth mindset treatment [5].
Society’s attempts to remedy this situation runs the spectrum of offensive to brilliant. The poster child of well-intended, but rather ridiculous, “math for girls” is Danica McKellar’s series “Math Doesn’t Suck.” Ms. McKellar (most famously known as Winnie Cooper from The Wonder Years), who has a BS in math and her name on a theorem, writes a series of books in pastel colored, curly fonts that sport covers that look like Cosmo. I’m surprised the I’s aren’t dotted with hearts. Her hair is always perfect, her shirt unbuttoned just so, and her head its tilted at a come-hither (to math, of course) 45-degree angle. At the other end of the spectrum, we have the promising line of girl engineering toys, GoldieBlox, created via Kickstarter by Debra Sterling, a Stanford engineer. GoldieBlox has received much press for their innovative line of fun, creative toys that encourage little girls to use applied simple math and engineering skills to solve problems – and their clever use of a Beastie Boys song.
As anyone who has ever shopped for young girls will know, GoldieBlox is the exception, rather than the rule. Girls who like math and grow up to be women in math-oriented fields are considered to be anomalies. What does this mean for women who do enter these fields, whether in academia or in the professional world? It means that we sit in classrooms where we are the only female. It means we sit in client meetings where we are mistaken for the secretary. It means our male advisors and supervisors occasionally get the wink-nudge “I see why you’ve got her working for you.” At this year’s MPSA, I was a panelist in a nearly-packed room. As I got up to speak, it occurred to me that I was one of only four women in the room, and the only person of color. We have few role models and little encouragement. Promising women are deterred from productive careers in quantitative fields, based on socialization rather than ability.
Gender-differential social pressures
Graduate school is best described as a monastic experience. We lead a sparse life, consumed by esoteric information-gathering that is appreciated by a small group. While most people respect a PhD, they generally have little understanding of what it is we do, exactly. Most of us manage to eke out a personal life, but it is generally limited by the all-encompassing nature of research and by the lack of disposable income. We continue to live in the pizza and free beer world while our non-academic counterparts move on to fancier affairs.
What does this mean for a woman in graduate school? In short, we are alienated. Most graduate students are in their twenties. The average completion time for a Political Science PhD is 6.5 years, according to the National Research Council. The average age that Americans get married is 27 for women and 29 for men – skewing slightly downward for women and slightly upward for men. The pressure for women in their twenties to get married and have children is intense. The social expectations for women in their twenties can be at odds with what it means to be a successful graduate student.
How are we alienated by society? First, we are constantly bombarded with messages and images of what it means to be a successful woman (hint: it has little to do with R skills). A glance at any 20-something’s Facebook feed of female friends is a menagerie of engagements, first dances, and tastefully planned flower arrangements. This progresses to baby bumps, knitting projects, and first steps. Most societies applaud these accomplishments. Recently, a cousin of mine was married. In true South Asian style, the celebration was a week of elaborate parties, delicious food, and an obscene amount of gold. At the same time, her sister ranked first in her class at a competitive pharmacy graduate program. Even in my often-ostentatious society, there’s no comparable reward for scholastic achievement.
How are we alienated by our graduate programs? The monastery analogy extends further to encompass family. Maternity leave has become a recent allowance at graduate programs. While a promising boost, what many schools offer is simply an extension of in-absentia status. This can mean no pay, and the potential of being removed from student health insurance, losing visa status, or reinstating suspended student loan payments. Expanded maternity allowance can help with the practical nature of juggling academics and family, but it does little to address the culture and perception of women who are pregnant or have children during their graduate career. Women who get married or become pregnant may be perceived as less serious, and an advisor with a limited amount of time and resources may choose to focus on more “promising” advisees. Similarly, the lack of emphasis on paternity leave reinforces the idea that the woman should be at home caring for children.
To better understand the culture, I suggest reading Anne-Marie Slaughter’s 2012 article Why Women Still Can’t Have it All. Her assessments of a work environment for mothers is parallel to the female academic experience. I often hear male colleagues complimented for being great dads was because they shoulder child care while their wives work. In contrast, a female professor once confided to me that at her first post-maternity leave faculty meeting, one of the male professors jokingly asked how her vacation was. Similarly, my female RA was hesitant to reveal her pregnancy to potential recommendation-writers, because she felt it would be counted against her. This article in Inside Higher Ed specifically addresses this issue in political science.
It is unfair to target only graduate programs for this stigma. As an undergraduate senior at MIT, many of us had the opportunity to interview with some of the top firms in any field. I remember being provided a word of advice from an alumna who had launched a successful investment banking career. “When you’re asked the five-year question,” she said, “never say you plan on being married. They’ll see you as a liability.” She was referring to the generic interview question of “Where do you expect to be in five years, professionally and personally?” For men in their early twenties, the advice was the opposite. Young men who have a goal of being married are viewed as reliable and stable, while women who expected to get married were viewed as a waste of resources.
What can we do?
Graduate women in quantitative fields have consciously chosen an alternative path that is explicitly and implicitly discouraged by our environment. The issues I point out in this article – mindset development and social pressures – are daunting but not impossible to overcome. As a woman who has weaved her way through the male-dominated environments of MIT, The Conference Board, analytic forecasting, and now quantitative methods within Political Science, I offer the following advice to women pursuing or considering pursuing graduate education in the male-dominated quantitative fields:
1. Observe your mindset. The best way to do this is to pay attention to your language. Women often attribute our success to luck and our failures to a lack of ability (interestingly, this is the opposite in men)[6]. This is magnified in quantitative fields, where our negative thoughts are validated by the actions and words of others. If you are a woman in a PhD program, you did not get there by chance. You are not “lucky” you got into a top-tier program. You are there because you belong there. To think otherwise is an insult to you and your hard work.
2. Hold your own. It is inevitable that you will be in situations that are uncomfortable for women, even if the males in the room don’t see it that way. A colleague of mine related that she is the TA for a graduate class composed of military mid-career professionals. While she emphasized that her students are very respectful to her, it is still a fragile situation. In neither this, nor my MPSA experience, is the environment overtly disparaging or negative. But that is irrelevant; being the one that doesn’t belong only amplifies any insecurities and self-doubts. Notice the situation, acknowledge it, and own it.
One of my favorite stories to tell is from a conference where I was invited to dinner with senior academics in the field. It was me, eight tenured white male academics in their 60s and 70s, and one girlfriend of one of the academics in a rather small booth. It was an informal dinner where the food and wine flowed freely. I was the only sober individual while the others ranged from raucous to falling asleep at the table. By the end of the night, the waiter was deferring to me as the authority at the table, since I was the only person able to answer his questions clearly.
3. Be friends with other women. One of the most self-defeating things that women do is alienate other women. By doing so we reinforce negative stereotypes of groups of women as catty, gossipy, and unproductive. We also cut off a resource for ourselves by internalizing our problems or airing them to individuals who cannot relate and we put ourselves in a situation where we have to go it alone, by eliminating those who have done it already.
4. Be a mentor. Graduate students are often barely out of college. It is hard to us to view ourselves as a mentor to anyone, yet we are in a unique position where we have authority but can still relate to our students. Use that to help your promising female students. Make it a point to ask them how they are doing or to provide action-based compliments that develop growth intelligence, such as, “I can see you’re working really hard on this problem.”
Quantitative methods can be a free-form field. While there is a hurdle of learning a programming language and the basics of statistics, the rest of our learning is often project-specific. If methodologists have a problem to solve, we google packages, read vignettes, find github accounts and snag some code. We then hack away at our problem until the code works. We screw up quite a bit, and, at some point, screw up a little bit less. There is an degree of self-confidence that is required to tackle a problem in that manner.
Even that last sentence is a loaded statement. Women who have advanced in these fields usually do so in spite of their socialization and their environment. The rebuttal of the “confidence gap” literature is that these concepts of the qualities of a good leader (or university professor) are predicated upon the path that has been forged by men.[7] While women may not be as aggressive self-salesmen, that does not make us less qualified as methodologists. What departments can do to improve their environment for women could (and should) be the subject of another blog post.
Due to the broad accessibility of information, advances in technology and statistical abilities and the growth of applied data science, the walls of the ivory tower are crumbling. For political methodology to remain relevant, interesting, and in order to advance the field in a meaningful way, we must embrace diversity. It is a detriment to the field that qualified and capable women are being turned away, and there is much we can do to draw the best and brightest, regardless of gender.
Notes
*I would like to thank Elaine Denny and Kathryn Dove for their advice in crafting this article.
[1] Henderson, V., & Dweck, C. S. (1991). Adolescence and achievement. In S. Feldman & G. Elliott (Eds.), At the threshold: Adolescent development (pp.197-216). Cambridge, MA: Harvard University Press.
[2] Burman, Douglas, Bitan, Tali, & Booth, J. B. (2008). Sex differences in neural processing of language among children. Neuropsychologia 46 (5) p. 1349-62.
[3] Cvencek, Dario, Meltzoff, Andrew N., Greenwald, Anthony G. (2011) Math–Gender Stereotypes in Elementary School Children. Child Development 82 (3), p.1467-8624.
[4] Blackwell, L., Trzesniewski, K., & Dweck, C.S. (2007). Implicit Theories of Intelligence Predict Achievement Across an Adolescent Transition: A Longitudinal Study and an Intervention. Child Development, 78, 246-263.
[5] Dar-Nimrod, I., & Heine, S.J. (2006). Exposure to scientific theories affects women’s math performance. Science, 314, 435.
[6] http://www.nber.org/digest/aug08/w13922.html
[7] http://www.theatlantic.com/features/archive/2014/04/the-confidence-gap/359815/