Unsettling Stats about Women in Science

Smithsonian Institution @ Flickr Commons

Smithsonian Institution @ Flickr Commons

In a recent post, I pointed out that:

“A 2012 study showed that the top 1% of the most highly cited ecologists were men (93%), American (65%), had an average of 32 years of experience in science, spent an average 11 hours per week writing grant proposals and publications, had a large lab of an average 11 people, and had average annual funding of $500,000. Similarly, the 2011 ESA survey I mentioned earlier also concluded:

“We can describe today’s ‘typical ecologist’ as a 55-year-old male professor who studies communities and ecosystems using field observations and experiments, and who finds professional satisfaction in his research. Twenty years ago, the description would have been a 38-year-old male professor studying aquatic communities.”

So, the keys to success are the ability to get funding and the ability to network and collaborate. It also apparently helps if you are a man.

Women make up about 50% of those who receive doctorates in science, but only 21% of tenured professors in science, said a publication in Nature in March. What is it about a woman’s experience during her scientific training that makes academia so unattractive? Let’s look at the numbers.

1. Women are paid less. Women in academic science are paid a salary that is lower by $24,000 than men. The increase in pay since the 1970s has been markedly slow for women, especially when compared to the increase in pay for men (citation). Given identical applications, science faculty (both men and women) are more likely to offer a man the job and would offer him $4,000 more in salary than a woman (citation). One of the common reasons noted for why women are not paid more is that they do not ask for a higher salary during negotiations. This is irksome because it makes it sound like it’s a woman’s fault that she’s not paid more. It’s not. This reasoning completely disregards the fact that if a potential employer already expects a person to be a pushover during negotiations, their negotiation tactics could be different, and their reaction to a person who is not a pushover when she is expected to be could also be unique. It also disregards the fact that if women are pervasively unable to negotiate for a fairer salary, they must be poorly prepared for negotiating their positions, which indicates a failure during training and/or biased behavior on the part of the employer.

2. Women are less likely to be funded, and, when they are, they are awarded ~$80,000 less than a man would be, according to a recent study. Women were awarded 30% of NIH research grants awarded in 2012, up a little from 24% in 2002. In a study just published in the British Medical Journal, only a quarter of the 6,000 research projects funded between 1997 and 2010 were awarded to a team led by a woman, and the amount of funding for woman-led research was 43 percent lower than research led by a man.

3. Most women leave academic science after finishing a doctorate. A 2012 publication in Proceedings of the Royal Society B, using NSF data from 1973 through 2009, crafted a model to test whether there are fewer women in science than would be predicted by the time lag of a historical lack of women. The authors found that a time lag explains some of the pattern but that it especially does not explain the loss of women from the field during the transition from graduate school to postdoctoral position. In 1979 women made up <40% of the undergraduate degrees in science and engineering but by 2006 made up almost 50%. In 1979 women made up about 30% of graduate degrees in science and engineering; by 2006, it was still under 40%. Most strikingly, women made up only about 10% of assistant professors and 5% of tenured professors in 1979, and those numbers only rose to ~10% and ~15%, respectively. Although they only gave numbers for postdocs for 1999 through 2006, the actual proportion of women in postdocs (~20%) was well under what they predicted would be the case due to the model’s built-in lag. In the biological sciences, where more than half of the undergraduate and graduate degrees are awarded to women, less than half of the women with PhDs ended up in tenured positions. (See the figures where I got the numbers here.)

4. Peer-review is gender biased. A recent study in PLoS One showed that women make up less than a third of those listed in prestigious authorship positions. Less than a third of first author positions, same for last author, and same for single-author publications. Although women are increasingly included as authors on scientific papers, they are not as likely to be in a prestigious position of authorship. When peer review is not double blind, meaning the reviewer can see who authored the paper that they are reviewing, women’s papers are less likely to be accepted than men’s (source here). Nature recently published an editorial on its attempts to increase the visibility of women in its pages, with some successes (more female authors) and failures (fewer female reviewers). However, an editorial in response seemed like a total slide backwards, as I discuss in #9 below.

5. Stereotypically feminine characteristics are at odds with the stereotypical characteristics of good intellectual leaders. One example assessment where we see this is in student evaluation outcomes. Although gender bias in student evaluations of teaching varies substantially between institutions and even between courses, the literature overall agrees that it exists (see here and here and here and here). Specifically, women are judged more harshly in large, lecture-based classes and in STEM fields, and male students in particular tend to rate professors who are female lower than their male counterparts. Professors are expected to be strong, smart leaders but women are expected to be sensitive and warm; male professors tend to be rated well if they show the characteristics of knowledgeability and leadership expected of a professor. Professors who are women must meet all expectations–students must perceive that they are sensitive and warm as well as knowledgeable, confident leaders. Assertive, ambitious women tend to receive lower evaluations if they are not also seen as warm and sensitive (source here).

As a recent article in the Harvard Business Review said about invisible barriers to women rising to leadership roles:

“Integrating leadership into one’s core identity is particularly challenging for women, who must establish credibility in a culture that is deeply conflicted about whether, when, and how they should exercise authority. Practices that equate leadership with behaviors considered more common in men suggest that women are simply not cut out to be leaders. Furthermore, the human tendency to gravitate to people like oneself leads powerful men to sponsor and advocate for other men when leadership opportunities arise…. The mismatch between conventionally feminine qualities and the qualities thought necessary for leadership puts female leaders in a double bind. Numerous studies have shown that women who excel in traditionally male domains are viewed as competent but less likable than their male counterparts. Behaviors that suggest self-confidence or assertiveness in men often appear arrogant or abrasive in women. Meanwhile, women in positions of authority who enact a conventionally feminine style may be liked but are not respected.”

6. Work culture favors those without families. Ecologists work on average 2.92 postdocs before settling in a more permanent, tenure-track position, as I discussed in a recent post. That’s a total of 22 years of moving around during your adult life (including college and grad school). The mobility expected within academic science in particular means those pursuing academic careers must compromise a partner’s career or endure a long-distance relationship (see postdoc Katie Mack’s commentary here), not to mention the difficulty endured if navigating this process with children. Postdocs and graduate students are not given the rights granted to most other people employed in the U.S., because they are deemed contractors rather than “real” employees. This problem is not limited to people in science. This oversight affects women especially–and in particular those who are recent mothers–because they are given neither maternity leave nor a private place to pump breastmilk, both required by U.S. law for “real” employees. Further, the average postdoc, regardless of gender, is paid a starting salary of ~$38,000. If this is a family’s main income, whether it comes from father or mother, it may be too low to pay for childcare, the cost of which is exorbitant in the U.S.

7. Metrics used to compare candidates based on research output inadvertently disadvantage women who have taken leave to care for family, despite their ultimate potential for high research output. A recent study used population modeling to extrapolate the effects of working part-time to care for family on research output over a career. As you can guess, working part-time affects science career metrics (e.g., h-index, impact factor) but not ultimate research output.

8. Sexual harassment and microagressions are pervasive in the workplace. University of Illinois anthropologist Kate Clancy undertook a survey of the experiences of women scientists at field sites. Over 60% reported sexual harassment; 20% reported sexual assault. Psychological and physical abuses based on gender were also reported. The male intellectual in need of female attention is a pervasive stereotype of men in academia, as well as other intellectual careers, as described here. Several Twitter conversations have tackled the topic, including this one, and #ripplesofdoubt, established after the sexual harassment scandal that led to the resignation of Scientific American‘s blogs editor Bora Zivkovic. The hashtag refers to the self-doubt instigated by acts of sexism and harassment, leading a woman to wonder whether she is valued for her sexuality and appearance rather than her talent and expertise, as explained here. Clancy also recently posted an anonymized email exchange exhibiting microagression of a male colleague.

9. Misogyny and sexism are often included as a valid “part of the conversation,” despite scientific and social evidence to the contrary. The most recent poster child of misogynist content published as a valid scientific perspective is a January 16th Nature editorial called “Research: Publish on the basis of quality, not gender,” in response to Nature’s November editorial “Gender progress (?).” It’s not the first time Nature has been called out for publishing sexist material. The January 16th editorial caused a storm of responses on social media. Science writer and editor Kelly Hills discussed the author’s lack of credentials  that normally go along with publishing in Nature (He has a BA in political science and no previous peer-reviewed work.) and his online trail of diatribes against “radical feminism.” University of Hawaii geobiologist Hope Jahren wrote an open letter to Nature about why she declined a Q&A with them after reading the editorial. With a final flourish of sarcasm, she offers to do a Q&A with the author of the editorial instead to keep from “leaving the editors in a lurch.” She, however, rescinds this offer when she realizes the author likes to Tweet about guns.

10. Gender bias is a problem in U.S. society in general, not just in science. The following stats are from the 2011 documentary Miss Representation: Women hold only 3% of higher-level positions in telecommunications, entertainment, publishing, and advertising. Only 16% of writers, directors, producers, cinematographers, and editors are women. Women comprise only 17% of Congress. The U.S. ranks 90th in the world in representation of women in federal legislatures, behind Cuba, China, Iraq, and Afghanistan. An equal number of 7-year-old boys and girls want to be president of the U.S. when they grow up, but a large gender gap in this ambition emerges by the time kids reach 15 years old.

SOLUTIONS

Gender bias is a problem for science, because a segment of the population of doctorates in science are leaving the field for reasons that have nothing to do with their potential for quality research. Here are the solutions that have been suggested:

Mentorship. Regardless of the gender of her PhD mentor, every woman studying for a PhD should receive advice from more advanced colleagues who are women, and her program should ensure that she is receiving such advice. The studies above show that a woman’s experience of building her career will be different than a man’s. Thus, she will be better advised about such differences by other women who have experienced it. Advice for women teaching, interviewing, and negotiating in academia can be found here and here and here and here and here. A recent article in Harvard Business Review states, “Without an understanding of second-generation gender bias, people are left with stereotypes to explain why women as a group have failed to achieve parity with men: If they can’t reach the top, it is because they ‘don’t ask,’ are ‘too nice,’ or simply ‘opt out.’ These messages tell women who have managed to succeed that they are exceptions and women who have experienced setbacks that it is their own fault for failing to be sufficiently aggressive or committed to the job.” In a recent NY Times article, Eileen Pollack talks about how many women underestimate their potential, and that lack of encouragement can make the difference between a woman advancing or leaving her career in science. Women judge themselves more harshly and tend to downplay their own value more than men do (sources here).

Building Allies. Having a supportive community of both men and women is essential to dealing with the difficulties mentioned above.  Biologist Andrew David Thaler wrote a blog post about how to be an ally to women as a male colleague. Gina Stewart wrote about the importance of camaraderie among women colleagues as they navigate their careers, as did chemist Carol V. Robinson in her commentary in Nature.

Better Postdoc Pay and Benefits. As graduate student Jennifer Bussell argued in her recent blog post, women would be more likely to stay in academia if postdocs were paid better. Lack of a paycheck that affords good childcare, (often) poorer healthcare benefits than staff and faculty, lack of paternity leave, and inflexibility in terms of location (i.e., generally not allowing working remotely) all play into the equation. As the Harvard Business Review article states: “Many entrenched organizational structures and work practices were designed to fit men’s lives and situations at a time when women made up only a very small portion of the workforce.” It’s high time for a paradigm shift in the way scientists are expected to work and build careers.

Lack of strong maternity/paternity leave policies is one of the biggest factors in evening out gender inequities in the workplace, as demonstrated by the data discussed in this Scientific American blog post:

“What’s left for countries like the U.S., which have closed gender gaps in education and health but fall short (in the case of the U.S., 23rd) in overall ranking? The big difference between top-ranking Finland and most other countries is the government’s unique level of commitment to helping women combine work and family life: With mandatory paternal leave, Zahidi says, comes more equal distribution of childcare and housework. This leaves women more able to rise to positions of power in the workplace, and has helped narrow the salary gap between men and women.”

Calling it out. Women and minorities are often advised to be silent about sexual harassment and discrimination in case their outspokenness costs them career advancement. Danielle Lee recently called B.S. in a Story Collider podcast:

“Even on my blog post [about an editor's calling me a "whore" after I politely declined a professional offer for unpaid work], someone was great enough to remind me how I was ruining my career potentially for speaking up and defending myself…. If that is what I must accept to be successful in this field, then I’ve lost my appetite for it…. I don’t believe that I have to accept mistreatment in order to just get a chance to sit at the table, particularly when we already know within these fields that they have diversity and retention issues. And if that is what we truly say we care about, and we got new people coming to the table, I think we are beyond time for a menu change.”

Science journalist Kathleen Raven advised in a recent post that if what one is about to say to a woman colleague is not something that one would also say to a man, perhaps it shouldn’t be said at all.

 

CONCLUSION

Women need to be educated about what constitutes inappropriate behavior, what subtle biases they may be up against, and how to deal with these obstacles professionally but directly. As the Harvard Business Review article points out: “More than 25 years ago the social psychologist Faye Crosby stumbled on a surprising phenomenon: Most women are unaware of having personally been victims of gender discrimination and deny it even when it is objectively true and they see women in general experience it.” The conversation in the science blogosphere surrounding the issues of sexual harassment and gender bias is candid and brave, and here are some examples. There’s a long way to go, but nevertheless I am heartened by the forthright conversation.


Comments

Unsettling Stats about Women in Science — 13 Comments

  1. You mention gender bias in peer review. Double-blind review wasn’t in the list of solutions (considering its effects on both submission and assessment of manuscripts). What can be done to improve the peer review process in a way that is fair(er)?

    • Excellent point, double-blind peer review belongs on the list. I hope other commenters will contribute more ideas for solutions; mine is certainly not comprehensive.

  2. This is a great overview of the situation. I’m sorry to say, though, that there’s even more. One example is in citation bias, which is related to but not quite the same as some of your points. When everything else is controlled for, the work of women is cited less than the work of men — although there seems to be quite a bit of variation between fields.
    I wrote up a few thoughts on that particular issue as “The great citation hoax: Proof that women are worse researchers than men” http://bit.ly/1g8gZuN
    along with much more on related issues at http://curt-rice.com/category/gender-equality/

  3. Pingback: Friday links: 7-year postdoc, every major’s terrible, citation Matthew Effect, KILL BLOGS WITH FIRE, and more | Dynamic Ecology

  4. There is an interesting commentary on this blog post happening on American Scientist’s Facebook page: https://www.facebook.com/AmericanScientist/posts/10101801417411888

    • This comment thread is pretty emblematic of my point that sexist commentary is included as a valid part of the conversation, despite much social and scientific evidence to the contrary.

  5. Thank you for this article. I’m leaving my graduate program with a MS mainly due to number 6, which you summed up succinctly. I have a spouse outside academia. We’ve done 3 years of long-distance already. My advisor asked why I was leaving, saying I was one of their best students. “I need to be able to get a job somewhere my spouse can get a job too”. “Ah, then a PhD doesn’t make any sense. Even as the best candidate, the job you want in the place you want is a 1/10 year event.”

    Dual career marriages require one or both people to be reasonably geographically flexible. Requiring post-docs to move across the country to a random location for a 1-2 year position, TWICE, has pushed a lot of coupled people out. I don’t know why this is more true for women than men; maybe because the timeline conflicts with fertility, maybe because on average their male partners will be earning more than them.

  6. Pingback: Nobody puts a Lady in the corner | Beyond The Codon

  7. Retention of women is an issue in many professions. On February 25 at noon Eastern Time, I will be presenting a webinar for the Association for Women in Science (AWIS) on Strategies to Retain Mid-Career Female Scientists.

    The geographic mobility issues adversely affect dual career couples in all disciplines. On March 25 at noon Eastern Time, I will be presenting a second AWIS webinar on The Dual Career Challenge.

  8. Thank you for this article/blog!! I felt like crying and yelling for joy at the same time. You have exposed it all in one logical, well-constructed piece.
    As a female entering graduate school in an almost exclusively male field in the 1960s, I expected some push-back. However, over the course of my career, I experienced essentially every one of the points you made in the article. I won support of male colleagues only by dint of persistence and a focus on quality. But things didn’t seem to get much better as I progressed through the ranks (I guess I wound up being one of the 5% of female full professors). I solved the authorship bias by simply using initials when submitting manuscripts; I received much less hassle from reviewers that way. A colleague who knew me only from the literature was surprised that I was a woman when we met at a conference.
    I took early retirement, feeling tired and under-appreciated after all the work I’d done during the previous 25 years–teaching, research, committee work, and course organization–for about $250,000 less than my male colleagues, although none of them had been willing to shoulder a comparable load.
    Women are almost always present at the grass roots of social reforms that have improved living standards and decreased tyranny. Until women are valued for their skills and contributions to human culture, our society will continue to be aggressive, destructive, and dysfunctional. Humanity cannot afford that cultural norm for much longer.

  9. Pingback: The Role of Men in “Women in Science” | WildlifeSNPits

  10. Pingback: Seksueel geweld houdt vrouwen uit de wetenschap | De Zesde Clan

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>