Is it really just sexism? An alternative argument for why women leave STEM

Karen J. Morenz

Feb 1, 2019 · 16 min read

Everyone knows that you’re not supposed to start your argument with ‘everyone knows,’ but in this case, I think we ought to make an exception:

Everyone knows that STEM (Science, Technology, Engineering and Mathematics) has a problem retaining women (see, for example Jean, Payne, and Thompson 2015). We pour money into attracting girls and women to STEM fields. We pour money into recruiting women, training women, and addressing sexism, both overt and subconscious. In 2011, the United States spent nearly $3 billion tax dollars on STEM education, of which roughly one third was spent supporting and encouraging underrepresented groups to enter STEM (including women). And yet, women are still leaving at alarming rates.

Alarming? Isn’t that a little, I don’t know, alarmist? Well, let’s look at some stats.

A recent report by the National Science Foundation (2011) found that women received 20.3% of the bachelor’s degrees and 18.6% of the PhD degrees in physics in 2008. In chemistry, women earned 49.95% of the bachelor’s degrees but only 36.1% of the doctoral degrees. By comparison, in biology women received 59.8% of the bachelor’s degrees and 50.6% of the doctoral degrees. A recent article in Chemical and Engineering News showed a chart based on a survey of life sciences workers by Liftstream and MassBio demonstrating how women are vastly underrepresented in science leadership despite earning degrees at similar rates, which I’ve copied below. The story is the same in academia, as you can see on the second chart — from comparable or even larger number of women at the student level, we move towards a significantly larger proportion of men at the more and more advanced stages of an academic career.

Although 74% of women in STEM report “loving their work,” half (56%, in fact) leave over the course of their career — largely at the “mid-level” point, when the loss of their talent is most costly as they have just completed training and begun to contribute maximally to the work force.

A study by Dr. Flaherty found that women who obtain faculty position in astronomy spent on average 1 year less than their male counterparts between completing their PhD and obtaining their position — but he concluded that this is because women leave the field at a rate 3 to 4 times greater than men, and in particular, if they do not obtain a faculty position quickly, will simply move to another career. So, women and men are hired at about the same rate during the early years of their post docs, but women stop applying to academic positions and drop out of the field as time goes on, pulling down the average time to hiring for women.

There are many more studies to this effect. At this point, the assertion that women leave STEM at an alarming rate after obtaining PhDs is nothing short of an established fact. In fact, it’s actually a problem across all academic disciplines, as you can see in this matching chart showing the same phenomenon in humanities, social sciences, and education. The phenomenon has been affectionately dubbed the “leaky pipeline.”

But hang on a second, maybe there just aren’t enough women qualified for the top levels of STEM? Maybe it’ll all get better in a few years if we just wait around doing nothing?

Nope, sorry. This study says that 41% of highly qualified STEM people are female. And also, it’s clear from the previous charts and stats that a significantly larger number of women are getting PhDs than going on the be professors, in comparison to their male counterparts. Dr. Laurie Glimcher, when she started her professorship at Harvard University in the early 1980s, remembers seeing very few women in leadership positions. “I thought, ‘Oh, this is really going to change dramatically,’ ” she says. But 30 years later, “it’s not where I expected it to be.” Her experiences are similar to those of other leading female faculty.

So what gives? Why are all the STEM women leaving?

It is widely believed that sexism is the leading problem. A quick google search of “sexism in STEM” will turn up a veritable cornucopia of articles to that effect. And indeed, around 60% of women report experiencing some form of sexism in the last year (Robnett 2016). So, that’s clearly not good.

And yet, if you ask leading women researchers like Nobel Laureate in Physics 2018, Professor Donna Strickland, or Canada Research Chair in Advanced Functional Materials (Chemistry), Professor Eugenia Kumacheva, they say that sexism was not a barrier in their careers. Moreover, extensive research has shown that sexism has overall decreased since Professors Strickland and Kumacheva (for example) were starting their careers. Even more interestingly, Dr. Rachael Robnett showed that more mathematical fields such as Physics have a greater problem with sexism than less mathematical fields, such as Chemistry, a finding which rings true with the subjective experience of many women I know in Chemistry and Physics. However, as we saw above, women leave the field of Chemistry in greater proportions following their BSc than they leave Physics. On top of that, although 22% of women report experiencing sexual harassment at work, the proportion is the same among STEM and non-STEM careers, and yet women leave STEM careers at a much higher rate than non-STEM careers.

So, it seems that sexism can not fully explain why women with STEM PhDs are leaving STEM. At the point when women have earned a PhD, for the most part they have already survived the worst of the sexism. They’ve already proven themselves to be generally thick-skinned and, as anyone with a PhD can attest, very stubborn in the face of overwhelming difficulties. Sexism is frustrating, and it can limit advancement, but it doesn’t fully explain why we have so many women obtaining PhDs in STEM, and then leaving. In fact, at least in the U of T chemistry department, faculty hires are directly proportional to the applicant pool —although the exact number of applicants are not made public, from public information we can see that approximately one in four interview invitees are women, and approximately one in four hires are women. Our hiring committees have received bias training, and it seems that it has been largely successful. That’s not to say that we’re done, but it’s time to start looking elsewhere to explain why there are so few women sticking around.

So why don’t more women apply?

Well, one truly brilliant researcher had the groundbreaking idea of asking women why they left the field. When you ask women why they left, the number one reason they cite is balancing work/life responsibilities — which as far as I can tell is a euphemism for family concerns.

The research is in on this. Women who stay in academia expect to marry later, and delay or completely forego having children, and if they do have children, plan to have fewer than their non-STEM counterparts (Sassler et al 2016, Owens 2012). Men in STEM have no such difference compared to their non-STEM counterparts; they marry and have children about the same ages and rates as their non-STEM counterparts (Sassler et al 2016). Women leave STEM in droves in their early to mid thirties (Funk and Parker 2018) — the time when women’s fertility begins to decrease, and risks of childbirth complications begin to skyrocket for both mother and child. Men don’t see an effect on their fertility until their mid forties. Of the 56% of women who leave STEM, 50% wind up self-employed or using their training in a not for profit or government, 30% leave to a non-STEM more ‘family friendly’ career, and 20% leave to be stay-at-home moms (Ashcraft and Blithe 2002). Meanwhile, institutions with better childcare and maternity leave policies have twice(!) the number of female faculty in STEM (Troeger 2018). In analogy to the affectionately named “leaky pipeline,” the challenge of balancing motherhood and career has been titled the “maternal wall.”

To understand the so-called maternal wall better, let’s take a quick look at the sketch of a typical academic career.

For the sake of this exercise, let’s all pretend to be me. I’m a talented 25 year old PhD candidate studying Physical Chemistry — I use laser spectroscopy to try to understand atypical energy transfer processes in innovative materials that I hope will one day be used to make vastly more efficient solar panels. I got my BSc in Chemistry and Mathematics at the age of 22, and have published 4 scientific papers in two different fields already (Astrophysics and Environmental Chemistry). I’ve got a big scholarship, and a lot of people supporting me to give me the best shot at an academic career — a career I dearly want. But, I also want a family — maybe two or three kids. Here’s what I can expect if I pursue an academic career:

With any luck, 2–3 years from now I’ll graduate with a PhD, at the age of 27. Academics are expected to travel a lot, and to move a lot, especially in their 20s and early 30s — all of the key childbearing years. I’m planning to go on exchange next year, and then the year after that I’ll need to work hard to wrap up research, write a thesis, and travel to several conferences to showcase my work. After I finish my PhD, I’ll need to undertake one or two post doctoral fellowships, lasting one or two years each, probably in completely different places. During that time, I’ll start to apply for professorships. In order to do this, I’ll travel around to conferences to advertise my work and to meet important leaders in my field, and then, if I am invited for interviews, I’ll travel around to different universities for two or three days at a time to undertake these interviews. This usually occurs in a person’s early 30s — our helpful astronomy guy, Dr. Flaherty, found the average time to hiring was 5 years, so let’s say I’m 32 at this point. If offered a position, I’ll spend the next year or two renovating and building a lab, buying equipment, recruiting talented graduate students, and designing and teaching courses. People work really, really hard during this time and have essentially no leisure time. Now I’m 34. Within usually 5 years I’ll need to apply for tenure. This means that by the time I’m 36, I’ll need to be making significant contributions in my field, and then in the final year before applying for tenure, I will once more need to travel to many conferences to promote my work, in order to secure tenure — if I fail to do so, my position at the university would probably be terminated. Although many universities offer a “tenure extension” in cases where an assistant professor has had a child, this does not solve all of the problems. Taking a year off during that critical 5 or 6 year period often means that the research “goes bad” — students flounder, projects that were promising get “scooped” by competitors at other institutions, and sometimes, in biology and chemistry especially, experiments literally go bad. You wind up needing to rebuild much more than just a year’s worth of effort.

At no point during this time do I appear stable enough, career-wise, to take even six months off to be pregnant and care for a newborn. Hypothetical future-me is travelling around, or even moving, conducting and promoting my own independent research and training students. As you’re likely aware, very pregnant people and newborns don’t travel well. And academia has a very individualistic and meritocratic culture. Starting at the graduate level, huge emphasis is based on independent research, and independent contributions, rather than valuing team efforts. This feature of academia is both a blessing and a curse. The individualistic culture means that people have the independence and the freedom to pursue whatever research interests them — in fact this is the main draw for me personally. But it also means that there is often no one to fall back on when you need extra support, and because of biological constraints, this winds up impacting women more than men.

At this point, I need to make sure that you’re aware of some basics of female reproductive biology. According to Wikipedia, the unquestionable source of all reliable knowledge, at age 25, my risk of conceiving a baby with chromosomal abnormalities (including Down’s Syndrome) is 1 in about 1400. By 35, that risk more than quadruples to 1 in 340. At 30, I have a 75% chance of a successful birth in one year, but by 35 it has dropped to 66%, and by 40 it’s down to 44%. Meanwhile, 87 to 94% of women report at least 1 health problem immediately after birth, and 1.5% of mothers have a severe health problem, while 31% have long-term persistent health problems as a result of pregnancy (defined as lasting more than six months after delivery). Furthermore, mothers over the age of 35 are at higher risk for pregnancy complications like preterm delivery, hypertension, superimposed preeclampsia, severe preeclampsia (Cavazos-Rehg et al 2016). Because of factors like these, pregnancies in women over 35 are known as “geriatric pregnancies” due to the drastically increased risk of complications. This tight timeline for births is often called the “biological clock” — if women want a family, they basically need to start before 35. Now, that’s not to say it’s impossible to have a child later on, and in fact some studies show that it has positive impacts on the child’s mental health. But it is riskier.

So, women with a PhD in STEM know that they have the capability to make interesting contributions to STEM, and to make plenty of money doing it. They usually marry someone who also has or expects to make a high salary as well. But this isn’t the only consideration. Such highly educated women are usually aware of the biological clock and the risks associated with pregnancy, and are confident in their understanding of statistical risks.

The Irish say, “The common challenge facing young women is achieving a satisfactory work-life balance, especially when children are small. From a career perspective, this period of parenthood (which after all is relatively short compared to an entire working life) tends to coincide exactly with the critical point at which an individual’s career may or may not take off. […] All the evidence shows that it is at this point that women either drop out of the workforce altogether, switch to part-time working or move to more family-friendly jobs, which may be less demanding and which do not always utilise their full skillset.”

And in the Netherlands, “The research project in Tilburg also showed that women academics have more often no children or fewer children than women outside academia.” Meanwhile in Italy “On a personal level, the data show that for a significant number of women there is a trade-off between family and work: a large share of female economists in Italy do not live with a partner and do not have children”

Most jobs available to women with STEM PhDs offer greater stability and a larger salary earlier in the career. Moreover, most non-academic careers have less emphasis on independent research, meaning that employees usually work within the scope of a larger team, and so if a person has to take some time off, there are others who can help cover their workload. By and large, women leave to go to a career where they will be stable, well funded, and well supported, even if it doesn’t fulfill their passion for STEM — or they leave to be stay-at-home moms or self-employed.

I would presume that if we made academia a more feasible place for a woman with a family to work, we could keep almost all of those 20% of leavers who leave to just stay at home, almost all of the 30% who leave to self-employment, and all of those 30% who leave to more family friendly careers (after all, if academia were made to be as family friendly as other careers, there would be no incentive to leave). Of course, there is nothing wrong with being a stay at home parent — it’s an admirable choice and contributes greatly to our society. One estimate valued the equivalent salary benefit of stay-at-home parenthood at about $160,000/year. Moreover, children with a stay-at-home parent show long term benefits such as better school performance — something that most academic women would want for their children. But a lot of people only choose it out of necessity — about half of stay-at-home moms would prefer to be working (Ciciolla, Curlee, & Luthar 2017). When the reality is that your salary is barely more than the cost of daycare, then a lot of people wind up giving up and staying home with their kids rather than paying for daycare. In a heterosexual couple it will usually be the woman that winds up staying home since she is the one who needs to do things like breast feed anyways. And so we lose these women from the workforce.

And yet, somehow, during this informal research adventure of mine, most scholars and policy makers seem to be advising that we try to encourage young girls to be interested in STEM, and to address sexism in the workplace, with the implication that this will fix the high attrition rate in STEM women. But from what I’ve found, the stats don’t back up sexism as the main reason women leave. There is sexism, and that is a problem, and women do leave STEM because of it — but it’s a problem that we’re already dealing with pretty successfully, and it’s not why the majority of women who have already obtained STEM PhDs opt to leave the field. The whole family planning thing is huge and for some reason, almost totally swept under the rug — mostly because we’re too shy to talk about it, I think.

In fact, I think that the plethora of articles suggesting that the problem is sexism actually contribute to our unwillingness to talk about the family planning problem, because it reinforces the perception that that men in power will not hire a woman for fear that she’ll get pregnant and take time off. Why would anyone talk about how they want to have a family when they keep hearing that even the mere suggestion of such a thing will limit their chances of being hired? I personally know women who have avoided bringing up the topic with colleagues or supervisors for fear of professional repercussions. So we spend all this time and energy talking about how sexism is really bad, and very little time trying to address the family planning challenge, because, I guess, as the stats show, if women are serious enough about science then they just give up on the family (except for the really, really exceptional ones who can handle the stresses of both simultaneously).

To be very clear, I’m not saying that sexism is not a problem. What I am saying is that, thanks to the sustained efforts of a large number of people over a long period of time, we’ve reduced the sexism problem to the point where, at least at the graduate level, it is no longer the largest major barrier to women’s advancement in STEM. Hurray! That does not mean that we should stop paying attention to the issue of sexism, but does mean that it’s time to start paying more attention to other issues, like how to properly support women who want to raise a family while also maintaining a career in STEM.

So what can we do to better support STEM women who want families?

A couple of solutions have been tentatively tested. From a study mentioned above, it’s clear that providing free and conveniently located childcare makes a colossal difference to women’s choices of whether or not to stay in STEM, alongside extended and paid maternity leave. Another popular and successful strategy was implemented by a leading woman in STEM, Laurie Glimcher, a past Harvard Professor in Immunology and now CEO of Dana-Farber Cancer Institute. While working at NIH, Dr. Glimcher designed a program to provide primary caregivers (usually women) with an assistant or lab technician to help manage their laboratories while they cared for children. Now, at Dana-Farber Cancer Institute, she has created a similar program to pay for a technician or postdoctoral researcher for assistant professors. In the academic setting, Dr. Glimcher’s strategies are key for helping to alleviate the challenges associated with the individualistic culture of academia without compromising women’s research and leadership potential.

For me personally, I’m in the ideal situation for an academic woman. I graduated my BSc with high honours in four years, and with many awards. I’ve already had success in research and have published several peer reviewed papers. I’ve faced some mild sexism from peers and a couple of TAs, but nothing that’s seriously held me back. My supervisors have all been extremely supportive and feminist, and all of the people that I work with on a daily basis are equally wonderful. Despite all of this support, I’m looking at the timelines of an academic career, and the time constraints of female reproduction, and honestly, I don’t see how I can feasible expect to stay in academia and have the family life I want. And since I’m in the privileged position of being surrounded by supportive and feminist colleagues, I can say it: I’m considering leaving academia, if something doesn’t change, because even though I love it, I don’t see how it can fit in to my family plans.

But wait! All of these interventions are really expensive. Money doesn’t just grow on trees, you know!

It doesn’t in general, but in this case it kind of does — well, actually, we already grew it. We spend billions of dollars training women in STEM. By not making full use of their skills, if we look at only the american economy, we are wasting about $1.5 billion USD per year in economic benefits they would have produced if they stayed in STEM. So here’s a business proposal: let’s spend half of that on better family support and scientific assistants for primary caregivers, and keep the other half in profit. Heck, let’s spend 99% — $1.485 billion (in the states alone) on better support. That should put a dent in the support bill, and I’d sure pick up $15 million if I saw it lying around. Wouldn’t you?

By demonstrating that we will support women in STEM who choose to have a family, we will encourage more women with PhDs to apply for the academic positions that they are eminently qualified for. Our institutions will benefit from the wider applicant pool, and our whole society will benefit from having the skills of these highly trained and intelligent women put to use innovating new solutions to our modern day challenges.