There’s been a fair bit of research finding that negative stereotypes are part of what deters women and racial minorities from computer science and STEM in general (e.g. [1]). These stereotypes make it harder for women and minorities to personally identify with computer science, and amplify some of the biases that they face in CS. So for this post, I’ll be going over observed phenomena in social psychology and sociology that pertain to stereotype change.

Stereotypes

Stereotypes are really hard to change. They’re reinforced from many sources (media, individuals, groups, etc). But even more than that, stereotypes are schema: they are how we mentally organize information about social groups, and how we can determine whether we are “in” or “out” of a group. Schema allow us to process information effortlessly, and are pretty deeply ingrained once they’re there.

The human brain is not very good at changing schema. When we see evidence that contradicts our schema, our brains will do all sorts of mental gymnastics to avoid confronting or changing the incorrect schema. Most frequently, we forget that we saw it all. Sometimes our misconceptions even get stronger [2].

This happens with stereotypes. Betz and Sekaquaptewa did a study where they showed role models to young girls, to try to motivate the girls’ interest in STEM [3]. Role models were either gender-neutral, or feminine. The result? Gender-neutral role models boosted interest – and feminine (counterstereotypic) role models actually reduced girls’ interest in science. To these girls, the feminine scientist – a stereotype violator – is aberrant.

Stereotype violators are not viewed favourably by others. Indeed, in laboratory settings, people go out of their way to punish stereotype violators [4]. Stereotype violators are seen as less likable, and less competent. Not surprisingly, women in science are rated as less likable and less competent than otherwise identical men [5, 6].

Subtyping

So, let’s say instead of being exposed to just one woman scientist, you are exposed to a bunch of them. Regularly. That will change your schema, right? Nope.

The human brain does a thing that social psychologists call subtyping. Instead of changing your mental model of what a scientist is (white male), you instead create a new category: the woman scientist [7].

And the evidence is that this is what happens to female scientists, and to female engineers [3]. Furthermore, the stereotype of the woman scientist is of an unfeminine woman. The unfeminine label in of itself is costly: these women are seen as less likable, less attractive, less competent, and less confident [3].

Perceived Variability

So how can we change stereotypes, then? It turns out a thing called “perceived variability” is key: it’s how much variation we perceive in an out-group [7]. “Out-group” here refers to any group that a person does not identify with; an “in-group” is one that that person identifies with. Humans systematically underestimate the variability within an out-group, particularly in comparison to the variability within the in-group (e.g. men see women as more homogenous than they are; whites see aboriginals as more homogenous; etc).

This is known as the Out-group homogeneity effect. We mentally exaggerate the stereotypical qualities of outgroups (and outgroup members), and ignore the counterstereotypical qualities.

We stop paying attention to stereotypes when we perceive greater variability in the group that’s been stereotyped [7]. For example, it’s a lot harder to think about aboriginals in terms of generalizations and stereotypes when you’re used to thinking about the differences between Inuit, Metis and First Nations, and differences between the Haida, Salish, Blackfoot, Anishinaabe, Innu, Mi’kmaq, Dene, etc.

Subgrouping

So how can we increase the perceived variability of an outgroup? Subgrouping refers to the process in which both people members are brought together around common goals or interests, and can include both in-group and out-group members. For example, creating a study group in a computer science class in which both women and men are represented  – or joining a robotics club which has a mix of white, Asian, black, and hispanic students.

Subgrouping “allows for a more varied cognitive representations of group members” [7] – it leads you to start seeing the members of your subgroup around their membership in your common subgroup – rather than their membership in any in-group or out-group. Richards and Hewstone have a very nice literature review about subgrouping and subtyping, showing how dozens of studies have consistently found that subgrouping leads to increased perceived group variability, and stereotype change.

The Contact Hypothesis in sociology gets at subgrouping: the observed effect that being familiar with a member of an outgroup (eg. homosexuals) increases your acceptance of the outgroup. Having a friend, classmate or family member who is queer means you a share with a subgroup with them (friend group, class, family, etc).

For subgroups to form effectively, they need to have meaningful cohesion to those in the subgroup. One study by Park et al that is described by Richards and Hewstone found that they could not form a subgroup around all engineering students: “[they] were all hardworking and bright, but in very different ways. Some were motivated only by money, some by parental expectations, and some by larger environmental goals.” Instead, there subgroups of engineering students formed, around those three motivators. [7]

Similarly, Park et al found that they could not form subgroups around continuous variables (high/moderate/high) or arbitrary bases [7]. And other studies in the Richards and Hewstone review found that trying to form subgroups around having minority status (e.g. clubs for women in STEM, study groups for black students) either did not change stereotypes about their group, or intensified them [7].

Indeed, I wouldn’t be surprised that part of why instructional techniques such as Peer Instruction disproportionately helps female CS/physics students is because they encourage subgrouping. When you have your whole class together, collaborating in small groups for class activities, you’re having them bond as classmates – rather than as members of in-groups or out-groups.

Discussion

This is one reason I’m always a bit iffy about Women in CS/Science clubs: they don’t promote stereotype change, but instead promote subtyping. Instead of changing the notion of what a computer scientist is, they reinforce the subcategory of woman computer scientist.

But stereotypes aren’t the only thing that affect minorities. Part of why Women in CS clubs are so popular is that they provide a sense of community to these women. This is really important when you’re a minority member! The sense of isolation that many women experience in STEM is why many of them leave.

And, as always, is evidence that girls only schooling can be good for encouraging young girls’ interest in math and science [8]. It’s somewhat of a tragedy of the commons problem: putting all the women together in a club helps those individual women cope with a culture in which they are negatively stereotyped – but it doesn’t change the actual stereotype.

References:
[1] Cheryan, Sapna, et al. “The Stereotypical Computer Scientist: Gendered Media Representations as a Barrier to Inclusion for Women.” Sex roles 69.1-2 (2013): 58-71.
[2] McRaney. The Backfire Effect. http://youarenotsosmart.com/2011/06/10/the-backfire-effect/
[3] Betz, Diana E., and Denise Sekaquaptewa. “My fair physicist? Feminine math and science role models demotivate young girls.” Social Psychological and Personality Science 3.6 (2012): 738-746.
[4] Rudman, Laurie A., and Kimberly Fairchild. “Reactions to counterstereotypic behavior: the role of backlash in cultural stereotype maintenance.” Journal of personality and social psychology 87.2 (2004): 157.
[5] Steinpreis, Rhea E., Katie A. Anders, and Dawn Ritzke. “The impact of gender on the review of the curricula vitae of job applicants and tenure candidates: A national empirical study.” Sex roles 41.7-8 (1999): 509-528.
[6] Moss-Racusin, Corinne A., et al. “Science faculty’s subtle gender biases favor male students.” Proceedings of the National Academy of Sciences 109.41 (2012): 16474-16479.
[7] Richards, Zoë, and Miles Hewstone. “Subtyping and subgrouping: Processes for the prevention and promotion of stereotype change.” Personality and Social Psychology Review 5.1 (2001): 52-73.
[8] Barinaga, Marcia. “Surprises across the cultural divide.” Science 263.5152 (1994): 1468-1470.

The rates of female participation in CS – and STEM in general – vary wildly from culture to culture. In the US, women currently make up about 18% of undergraduate CS students [1], but over in Qatar, women make up about 70% of CS undergrads [2].

Women in STEM are better represented in countries such as Turkey, Hungary, Portugal, and the Philippines. In these countries, women make up approximately 50% of STEM undergrads [3]. Indeed, well-developed countries like Canada, the US, and the UK have some of the lowest levels of female participation in STEM.

So, what cultural factors lead to fewer or more women in STEM? Per the work of Barinaga, there are five factors [3]:

1. Recently developed science capabilities, resulting in an unentrenched scientific community
2. Perception of science as a low status career
3. Class issues that overshadow gender issues
4. Compulsory math and science education in secondary school
5. Large social support for raising families
   

New to Science

While it’s a bit surprising that Portugal and Mexico have better levels of female participation in science despite these countries not having well established scientific scenes, the evidence is actually that they have these better levels because of the newness of their scientific communities [3]. In countries like the US and the UK, the scientific communties have entrenched cultures. So called “old boys networks” were built up before women were allowed into the labour market; science has been firmly established as a masculine occupation. Portugal, for instance, begin its scientific and technological establishments in the 20th century, when society was more open to female participation.

It should be noted, however, that while countries like Portugal may have large numbers of women in science, few are making it to the top. Beatriz Ruivo, who studies female participation in Portugese science, has found that the
glass ceiling there is partly due to the lack of a strong women’s movement in Portugal [3]. We see an interesting parallel in the history of computer science. In the early days of computer programming (30s-60s), most programmers and coders were women [4]. It was later when stereotypes of programmers being nerds developed – and IT companies began specifically hiring those who were like the nerds in order to make up for a labour shortage in the late 60s – that programming became highly masuclinized.

Science As a Low-Status Occupation

It is fairly established in the sociology literature that, across cultures, the lower the status and pay an occupation, the more likely it is that women will be found there [3]. And not only are women more socially encouraged to stay in low-status occupations, but some occupations are reinforced as having low status due to the large numbers of women – forming “occupational ghettos”.

This was certainly the case in the early history of computer programming. Women were traditionally “computers” – those that did the hand computations, whereas men actually did the science [5]. When computers entered the mix,
it was the men who were to decide what the computers should calculate, and women were left as the low status “coders” to carry out the low-level work [4].

For countries with recently developed science communities, basic science is not highly connected to the production of goods and services. Science is hence seen only as an intellectual, cultural pursuit – not unlike how the humanities are regarded in the US and Canada. The humanities in North America are frequently (and unfortunately) derided as being “useless” – and have largely equal levels of women and men in modern day.

In computer science, it has been noted that male students often select careers in CS for the money. As computer science has become known as a lucrative field, more men have been specifically drawn to the field – and driving out their female colleagues.

A Matter of Privilege

In India, southern Europe, and Latin America, the social hierarchy puts high class women above low class men [3]. In these countries, education is often limited to the upper classes, resulting in a very different environment in academia than in the general population.

In North America, women from affluent communities, with parents in IT, were more likely to go into computer science themselves [6]. In short, the more privilege you have, the more likely you are to study CS – for instance, a White woman from a rich family and urban neighbourhood is more likely to have a job in STEM than than an Aboriginal man from a poor, rural family.

For computer science, the digital divide plays in to class issues [6]. The low classes not only are less likely to receive higher education, but also less likely to be connected to modern computing. Without a connection to computers, one would expect fewer of them to study computer science.

Compulsory Schooling – And Mindset

Former Soviet countries have higher rates of female participation in science, and Barinaga attributes this partly to the requirement that all secondary school students take multiple science courses and mathematics [3]. As a result, girls “can’t ‘chicken out’” of science and don’t close doors on themselves before they reach university’’. The policy of teaching all science subjects, in particular, is beneficial – when students can choose one science out of a list (as is the case in many Canadian provinces), female participation in physics is reduced.

The American approach of science being optional – and hence avoided by all but the gifted students – leads to a mentality to that you either have talent in science, or you don’t [3]. This fixed mindset approach to science has been consistently found detrimental both to individual success in science, as well as for minorities. In countries like Italy, where all sciences are mandatory, the communal mindset about science is a growth mindset: anybody can do it.

Support For Families

Forty percent of women who leave the workforce cite their husbands – and specifically, their husbands’ inability to pull their weight with housework and childcare – as their reason for leaving [8]. The United States was described by Barinaga’s international participants as “just a horrible place to try to raise a family and have a career’’. Without state-mandated parental leave, allowances for dads to stay home to look after children, and daycare, it is difficult for many women to manage both career and family.

Contributing to the problem is the Protestant work ethic for men, leading men to focus only on work and leave everything else to their wives. In northern Europe, Canada, and the US, fathers spend less time looking after their families [3]. Female science participation is higher in countries where childcare is a shared responsibility: not just between father and mother, but also with the extended family, and society at large.

This shared responsibility needs to be present in the workplace too; as one of Barinaga’s participants described: “if I missed a half-day of work [in the United States because] my kid had a temperature of 104, I was lectured on how this let down the [department]. In Israel there is 3 months paid maternity leave, day-care centers on every block, and if you don’t take off from work for your kid’s birthday party the department chairman will lecture you on how important these things are to kids and how he never missed one while his kids were little“ (Emphasis added).

A Final Note

Culture is a complex issue. None of the issues listed here can be a panacea for North American STEM. For example, even if we made CS obligatory in high school, it’s unlikely to have an effect for many racial minorities (Black/Hispanic Americans, Aboriginal Canadians, New Zealand Maori, etc), as these groups have low rates of high school completion [7]. By identifying these cross-cultural factors that promote women in STEM, we can better identify what factors (plural!) need addressing here in North America.

References:
[1] NCWIT By the Numbers. http://www.ncwit.org/resources/numbers
[2] Guzdial. Women in CS in Qatar: It’s Complicated. http://computinged.wordpress.com/2010/05/03/women-in-cs-in-qatar-its-complicated/
[3] Barinaga. “Surprises Across the Gender Divide”. Science 263, number 5152 (1994): 1486.
[4] Ensmenger. “The Computer Boys Take Over”.
[5] Rossiter. “Women scientists in America: Struggles and Strategies to 1940”, volume 1.
[6] Ashcraft, Eger and Friend. “Girls in IT: The Facts”. http://www.ncwit.org/resources/girls-it-facts
[7] Adams, Hazzan, Loftsson, and Young. “International Perspective of Women and Computer Science”. http://dl.acm.org/citation.cfm?id=611892.611897
[8] Stone. “Opting Out? Why Women Really Quit Careers and Head Home”. University of California Press, 2007.

In my last post, I described how the experiences of women in CS have changed historically. In this post, we saw that the academic side of computer science is a relatively recent thing. For this post, I’d like to focus some more on that aspect of the history. Like that last post, this post will be specifically focusing on North American CS (we’ve seen previously that female participation in CS is different outside the West!).

Generational differences exist between female scientists in academia. Etzkowitz et al in a 1994 paper found differences in experiences and values between the trailblazing “First Generation” of women in a field, and the subsequent “Second Generation”. As the paper is now 20 years old, it’s not too surprising that it feels a bit out of date – what comes after the Second Generation? (Another dated thing about the paper is that CS is described as being as female-friendly as biology.)

The Etzkowitz et al paper studied 30 academic science departments (biology, chemistry, physics, CS, and electrical engineering). They went into the study interested in the notion of critical mass – whether having enough women in a department would lead to a positive feedback cycle leading to gender equality. (Answer: it’s not that simple.) In the process of studying critical mass, they found the women who had entered the field before it was attained (First Gen) had fundamentally different experiences than the women who entered after.

The First Generation

The trailblazing women who entered CS – or similar disciplines – when there were no other women in their departments learned to cope with the culture by adopting the “male model” of a scientist. These women generally did not have families, and for those that did, it took a clear backseat to their scientific careers.

In departments without other women, these trailblazers often encountered blatant sexism and harassment. This open sexism did not abate until a critical mass of women was reached and women not only had “safety in numbers” but men were more aware that this behaviour was inappropriate. Etzkowitz et al describe the critical mass as a “strong minority of at least 15%”. Note that in this statistic, they are referring to how many women are faculty and graduate students in a department – this does not include undergrads.

These trailblazers were often uneasy about forming Women in Science type clubs, sometimes refusing to participate out of fear of stigmatization by their male colleagues. These women, having fought tooth and nail for any status and accomplishments they have, were sometimes afraid that association with women’s movements would devalue their achievements. Instead of being viewed on par with the other men, they worried they would be judged only in the “women’s track”. (There are certainly women who have gone against this – Maria Klawe would be a clear example of a First Gen computer scientist who has been promoting women in CS clubs and conferences.)

A quote from the paper really sums up the First Gen – one senior female scientist participating in the study described her generation: “The ones who did [science] were really tough cookies. Now it’s easier to get in. At one time it wasn’t even acceptable to start. So if you started back then you were tough to begin with. I have quivering women coming through who are very smart asking can they compete with men, and can they compete on a very competitive, fierce playing field. Of course they can. They just are not taught to be competitive. They don’t expect to win. The reason why I am successful is because I never felt this way.”

Competitiveness was a large source of tension between these women and the Second Generation. In the mind of the First Gen, women need to adapt themselves to the man’s word – and need to be competitive. Second Gen women have instead favoured trying to change the culture to allow women who meet cultural notions of femininity: making the culture more friendly and collaborative.

The Second Generation

Women who entered CS after critical mass was achieved had a very different experience coming into the field. Etzkowitz et al don’t provide timelines in their paper; from talking to female faculty in my department, I’d guess that this generation begins with the women who entered CS as undergraduates in the 80s.

These women tended to have high expectations about the (First Gen) female faculty in their departments, wanting their moral support and guidance for coping in a male-dominated culture. Often, they were disappointed. The Second Gen women wanted to have it all: to be women and scientists – and the First Gen women failed as role models in this regard.

For the Second Gen women who had First Gen women as advisors, there was tension. One Second Gen participant described: “[having a woman advisor] turned out to be somewhat of a mistake. I was under the impression that having a woman adviser would make life a bit easier… It turned out to be worse… Their motto is sink or swim… My adviser’s approach was to put it too far out of my grasp.“

First Gen women, as advisors, were extra hard on their female advisees, “to prepare them to meet the higher standards that they would be held to as women.” And as advisors, the First Gen women felt unable to help their advisees; as one participant put it, “They ask me when they should have children, can I take a part-time post-doc and then get back in? I don’t know [the answers]. I can’t help them.”

Most of the Women in CS/Science initiatives appear to have been started by Second Gen women, partly in response to the unhelpfulness of the First Gen women in terms of advising them about work-life balance and coping with a hostile, isolating work environment. And many Second Gen women left academia to look after their families, convinced that they would not be able to do both – if an academic career required conforming to the man’s world like the First Gen did, they decided they did not want to be a part of it.

Post-Etzkowitz et al: A Third Generation?

As I noted already, the Etzkowitz et al paper was published 20 years ago. I took my first CS course in 2007, and for me and my cohort it was a very different experience than that of the Second Gen women. Approximately 20% of the CVS faculty at my alma mater are women, predominantly women of the Second Generation. They have families and the Focus on Women in Computer Science club was (and still is) highly visible and active. Personally, I’ve received a lot of invaluable mentorship and advice from Second Gen women.

My generation is far removed from the overt sexism that the First Gen experienced, and we don’t appear as worried about balancing a career with family. For a lot of us, these feel like problems of the past. Occasionally I’ll hear friends comment about Women in CS events that “I feel like the women running this are trying to make up for what they didn’t have when they were our age rather than what our generation wants.” The best Women in CS events seem to be the ones that take generational differences into account.

Growing up, girls of my generation performed equally well in science and math as boys (sometimes outperforming). For a lot of us – though hardly all – there was no expectation setting foot in a CS class for the first time that it would be unfriendly to women. My experience of undergraduate CS was that of a collaborative field. Personally, it wasn’t until graduate school that I felt I encountered gender-based barriers.

But despite many improvements in the culture, female enrollment in CS hasn’t improved a whole lot since hitting that 15% critical mass. Despite an uptick in the mid-80s, the numbers are now down to around 18%. Clearly, critical mass isn’t enough on its own to get female participation to 50%.

For biology, however, the numbers have been increasing – 53% of biology doctorates in the US in 2009 were given to women (Zuk & O’Rourke). (I’ve posted previously about why biology has more women than CS.) But as Zuk and O’Rourke caution: “First, demography alone has not solved the problem [of gender inequality] in the past. We frequently make presentations about gender and science to young audiences; since perhaps the early 1990s, a common response from graduate students to the concern about lack of female professors is that “their” cohort had not yet gone through the system. In other words, the students optimistically suggested, all we needed to do was wait for them to move into the academic job market in equivalent proportion to their numbers. Unfortunately, that has not occurred over the past few decades, and it is not likely to happen now. Although the landmark majority of female biology Ph.D.’s was reached only recently, the number of women in undergraduate and graduate programs in the life sciences has been increasing for the past several decades.”

Subtle, social-psychological barriers still remain in the scientific community (see: Moss-Racusin et al, Steinpreis et al, Knobloch-Westerwick et al, Heilman et al). It’s unlikely that biology or any other science will get to having 50% female faculty until these barriers are gone. In a previous post I talked about how the key to changing stereotypes about women is to get people to see women as heterogeneous – generational differences are just one way that women in CS are heterogeneous.

Why is it that there are more women in biology than there are in computer science in North America? Women in the biomedical fields are now earning more than 50% of undergraduate degrees in the US [1].

Biology, like computer science, was once stereotyped as masculine. Medicine continues to be stereotyped as masculine, especially fields such as surgery. Why has biology attracted so many more women than computer science?

To answer this question, I’ll be synthesizing the findings of Cheryan’s “Understanding the Paradox in Math-Related Fields: Why Do Some Gender Gaps Remain While Others Do Not?” [2], Cohoon’s “Women in CS and Biology” [3], and Carter’s “Why students with an apparent aptitude for computer science don’t choose to major in computer science” [4].

Between these three papers, four themes emerge for why women choose one STEM field over another:

1. Exposure to the field
2. Expected value of the major
3. Lack of prejudice in the scientific culture
4. Prospects of raising a family in that scientific culture

Exposure

The ultimate finding of Carter’s “Why students with an apparent aptitude for computer science don’t choose to major in computer science” is that students simply didn’t know what CS is, or had misconceptions of the field [4].

Most undergraduates in North America never have to take any CS, and never saw any in high school. The big boon to biology enrollment is that biology is a course that pretty much everybody has to take in k-12. As we saw in comparing female representation in STEM between cultures, compulsory schooling plays a role in getting women into STEM.

But high school science isn’t the only way to expose young men and women to science. Women are better represented in astronomy and the earth sciences than they are in computer science, and neither of those fields are well-represented in k-12. Exposure can come from museums; television programmes and other documentaries; popular science books, magazines and blogs; public lectures; and science camps. Computer science does comparatively little public outreach.

Early exposure is also important. In Carter’s study, numerous students – disproportionately female – would only discover CS near the end of their degrees – too late to major or minor in the field. Multiple points of entry to CS majors, and multidisciplinary programmes, are hence recommended to increase female participation in CS [5].

Exposure at an early age also is useful. Girls who are given hands-on exposure to computers at an early age are more likely to wind up in CS [6]. Girls whose mothers are confident around computers are more likely to be confident around computers [6]. Girls who come from academic families are more likely to wind up in CS [6].

Finally, exposure is important for overcoming stereotypes about CS. Cheryan compared giving women descriptions of computer science as being a nerdy discipline – versus descriptions of computer science “not being like that” [7]. Women were statistically significantly more interested in computer science when given a non-stereotypic description of computer science.

Expected Value

Expectancy-value theory is one of the numerous theories out there used to model how undergraduates choose their majors. In a nutshell: undergrads are more likely to choose majors that they expect to align with their values and beliefs.

Cheryan argues that women are choosing biology over CS because they see in as more fulfilling: there is the promise of intellectual challenge combined with the promise of benefiting society [2].

But it’s not that simple – not all women have the same values, beliefs, and backgrounds. Margolis and Fisher found that women from racial minorities and international students who came to the US to study were motivated by the financial stability promised by a CS career [8]. And biology careers tend to appear more stable to undergraduates: biology faculty almost never turn over, whereas CS faculty will leave academia to go to industry. Maintaining a stable faculty in a department is good for gender representation [3].

Actual Openness

Another finding of Cohoon’s is that biology professors have better opinions of female students than CS professors do [3]. Biology professors also spend more time mentoring students than do CS professors [3].

Biology continues to have issues with prejudice. Women are less likely to be hired than equally competent men [9], will be offered lower salaries [9], and their work is viewed less favourably [10]. But the evidence indicates that biology is still more open to women to female scientists than CS is. And as we saw in the cross-cultural comparison, an unentrenched scientific community is conducive for minorities to enter the community.

Prospects of Raising a Family

There’s little evidence that women consciously choose majors based on how friendly the major is with respect to raising a future family. But in a society where women are socialized from a young age to expect to be the primary caretakers of their future offspring, it is not surprising that women are deterred from fields that seem unfriendly to raising a future family.

The process is more of an accumulation of red flags: The long hours in CS are only one red flag. As Carter found, CS is considered a volatile field without job security [4]. A lack of role models that are actively parenting adds to the notion of family-unfriendliness: they fail to provide evidence that women can have families and be in computer science. The relatability of role models is important: it is counterproductive in this regard to see female professors who have no families and are focused only on science [1].

The stereotypes about computer scientists are another red flag: computer scientists are seen as unattractive, singularly focused on technology, and asocial. Male computer scientists hence are unattractive as potential partners – and there’s plenty of evidence that humans are subconsciously drawn towards careers that are more conducive to meeting potential partners [11].

The sad evidence is that a fraction of white women are deterred from STEM because they do not want to be seen as unfeminine or intimidating to future partners [11]. Women who do go into STEM are more likely than non-STEM women to believe that men are unintimidated by their career choice, and they are more likely to have fathers, brothers and boyfriends that support this belief [11].

Overall, this lines up with what we saw in the cross-cultural comparison: women are more likely to go into STEM in cultures where raising a family is viewed as a communal responsibility.

References:
[1] Ashcraft, Eger and Friend. “Girls in IT: The Facts”. http://www.ncwit.org/resources/girls-it-facts
[2] Cheryan. “Understanding the Paradox in Math-Related Fields: Why Do Some Gender Gaps Remain While Others Do Not?” 10.1007/s11199-011-0060-z, Sex Roles 66 (3 2012): 184–190. issn: 0360-0025. http://dx.doi.org/10.1007/s11199-011-0060-z.
[3] Cohoon. “Women in CS and biology.” SIGCSE Bull. (New York, NY, USA) 34, number 1 (February 2002): 82–86. issn: 0097-8418. doi:10.1145/563517.563370. http://doi.acm.org/10.1145/563517.563370.
[4] Carter. “Why students with an apparent aptitude for computer science don’t choose to major in computer science.” SIGCSE Bull. (New York, NY, USA) 38, number 1 (March 2006): 27–31. issn: 0097-8418. doi:10.1145/1124706.1121352. http://doi.acm.org/10.1145/1124706.1121352.
[5] Cohoon. “Recruiting and retaining women in undergraduate computing majors.” SIGCSE Bull. (New York, NY, USA) 34, number 2 (June 2002): 48–52. issn: 0097-8418. doi:10.1145/543812.543829. http://doi.acm.org/10.1145/543812.543829.
[7] Cheryan, Plaut and Handron. “The Stereotypical Computer Scientist: Gendered Media Representations as a Barrier to Inclusion for Women.” Sex roles (2013): 1–14.
[8] Margolis and Fisher. Unlocking the clubhouse: Women in computing. MIT press, 2003.
[9] Moss-Racusin, Dovidio, Brescoll, Graham and Handelsman. “Science
faculty’s subtle gender biases favor male students.” Proceedings of the National Academy of Sciences 109, number 41 (2012): 16474–16479. doi:10.1073/pnas.1211286109. eprint: http://www.pnas.org/content/
109/41/16474.full.pdf+html. http://www.pnas.org/content/109/41/16474.abstract.
[10] Knobloch-Westerwick, Glynn and Huge. “The Matilda Effect in Science Communication: An Experiment on Gender Bias in Publication Quality Perceptions and Collaboration Interest.” Science Communication (2013).
[11] Hawley. “Perceptions of male models of femininity related to career choice.” Journal of Counseling Psychology 19, number 4 (1972): 308.

Female participation in computer science in North America has varied a great deal over time. Women were the original “computers” before the days of computing machines – and then were hired as the low-status “coders” to run those machines. Over time, coding/programming was more widely recognized to be difficult – and it was shifted from being “women’s work” to “men’s work”.

When computer science emerged as an academic discipline in the 70s and 80s, women were well-represented (30-40%). As enrollments in CS programmes exceeded what departments could manage, they tightly restricted the paths one could take into a CS major – unintentionally pushing non-traditional students like women out of the field. A big lesson from that period is that non-traditional students come from non-traditional paths – many of these women were starting in majors such as psychology or linguistics, or transferring from community colleges, and hence did not follow the “standard” path into computing careers.

Women as Computers: from the 1820s to the 1910s

Our view of women in computer science begins with the history of women in academia. The 19th century marked the rise of women’s colleges in the United States [1] as policies barring women from education were loosened. Women campaining for access to higher education did so on an argument that it would “produce better wives and mothers’’ for Americans [1]. For women of privilege in American society, a basic understanding of science and math in turn became “necessary for motherhood.’’

It should be emphasized that this was a trend for white women of privilege – most women who studied science in the 19th century were the daughters of scientists and other intellectuals.

For the women scientists that emerged from these colleges, there were few job opportunities. Teaching at the women’s colleges was the main possibility [1]. Working as a “computer’’ was another possibility. Women pursuing PhDs or faculty positions were expected to be single or “in no danger of marrying’’; marriage meant resigning from the programme or their job [1]. As time progressed and society progressed, women in these positions began to feel they could be both wives and scientists – when they resisted the norm of resigning upon marriage, they were met with opposition: they were threatened and usually fired [1].

1870-1900 marked an era of slow infiltration: women began entering doctorate programmes at traditional (male) institutions in countries such as the US and Germany [1]. Most universities were hesitant to allow the women into the PhD programmes, but would instead admit them as “special students’’ and give them additional bachelor’s degrees at the end of their studies. While by 1910 women were starting a presence in science at traditional institutions, there was no equality in employment, and jobs remained deeply sex typed.

With the slow rise of women in science came the corresponding rise of “women’s work‘’ in science. So-called women’s jobs typically were “assistants’’ to scientists, or working as computers for larger groups. These women were systematically ignored in the larger scientific community, left out of lists of scientists, conferences, and histories [1]. Indeed, from 1911 onward there were overt efforts to reduce the numbers of women in science, even with their roles undervalued [1].

It should be emphasized that computation was considered “women’s work’’ in the 19th and early 20th century. Looking at the history of the biological and social sciences in this time, quantitative methods were considered “low’’ enough that women could do them – but qualitative methods required “the intellect of a man’’ [2]. The reversal of the status (and gendering) of quantitative vs. qualitative work in the social and biological sciences happened well into the 20th century (sometime between the 30s-50s) [2].

The expansion of “Women’s Work”: 1920s to 40s

By the 1920s, women in academia were still largely kept to the women’s colleges [1]. The colleges, however, allowed a place to organize campaigns for change. Women began fighting for access to education using evidence from psychology and anthropology that women too were capable of science and math [1].

The 20s and 30s marked an expansion of government-employed scientists, who were assigned “women’s work’’ (assistants, computers, etc) and were grossly underpaid and undervalued [1]. The World Wars increased the scope of “women’s work’’ as labour shortages necessitated it. By 1938, the numbers of women working in scientific and technological roles for the US government had dramatically increased – despite overtly hostile job conditions [1].

The World Wars also marked the birth of digital computing. Computing machines were devised in the UK for cryptographic purposes. These machines, and the hand computations done in the wars throughout the world, were commonly performed by women. ENIAC, arguably the first real computer, was announced in 1946. The plan to run the ENIAC was such: a male scientist would be the planner, deciding what was to be computed – and a low-rank, female “coder’’ would do the actual machine coding [3]. These “Eniac Girls” and the other female machine operators of their time have been frequently forgotten in the history of science; at the time they were not seen as important and it is really only in recent decades that their work has been recognized.

Grace Hopper, who worked on the ENIAC, later described programming as “it’s just like planning a dinner. You have to plan ahead and schedule everything so it’s ready when you need it. Programming requires patience and the ability to handle detail. Women are ‘naturals’ at computer programming.” [4]

The Continual IT Labour Crisis: the 50s through 70s

What was not anticipated was that the coding would actually be difficult [3]. As computers began being used for commercial purposes in the 50s, a labour shortage emerged. The status of being a programmer rose; as the difficulty of its task was recognized, the assumption that it should be done by men took over. Computing in the 50s and 60s can be characterized by a large, shotgun approach to recruiting “good programmers’’ with little knowledge of what a “good programmer’’ was [3]. Programming began to be seen as a “dark art’’, and programmers began to be seen as asocial [3].

As computer programming rose in prominence, it became masculinized. Women were still allowed entry to the jobs due to the desperation for quality labour. However, lazy hiring practices that focused on spurious aptitude and personality tests hurt female participation in the industry [3]. Inconsistent professionalization efforts also hurt female participation by restricting what it mean to be a programmer [3]. The men running the show simply did not consider how their hiring practices discriminated against women.

Computer programming stayed largely independent from academic computer science. In the 50s and 60s, computer science was conducted through other departments, typically as a hobby or side-project [3]. The first CS classes were offered in the 60s, as the discipline struggled to assert itself as a discipline of its own [3].

By 1969, MIT had opened an undergraduate programme in CS – and the 70s marked the beginning of bachelor’s degrees in CS offered typically through electrical engineering or mathematics [3]. It would not be until the 80s, though, that CS programmes moved into their own departments.

From the start, computer science seemed like a “grab bag of various topics’’ related to computers [3] and attempts to define the discipline were inconsistent. Was computer science about information? Analysis? Algorithms? No consistent narrative was established, though algorithms eventually became dominant. This inconsistent narrative continues to be a difficulty in public outreach for computer science.

Academic CS: cyclical enrollments from the 80s to present

The opening of CS departments in the 80s provided a fertile ground for women. Women were increasingly studying the sciences in the 80s [5] – and academic CS had a relatively unentrenched culture. Women of the time flocked to CS in what is now seen as a golden age of female participation in the field. 37% of American CS degrees in 1985 were awarded to women [5]. In my next post, I’ll talk about how the experiences of these were different than the previous generations of women in CS. (Edit: the generational differences post is here)

The early 80s were also a boom-time for student enrollment in CS [6], which was linked to the rise of the personal computer. Personal computers had not been available until the late 70s; prior to then, computer science was hence only pertinent to academia, military, and business.

However, by the late-80s, enrollments began dropping – and disproportionately so for women [7]. The decline was “largely the result of explicit steps taken by academic institutions to reduce computer science enrollments when it became impossible to hire sufficient faculty to meet the demand.’’ [7] Steps included adding new GPA requirements for entering CS programmes, requiring more prerequisites, and retooling first-year CS as a weeder course. These actions disproportionately hurt not only female participation in the field, but participation of racial minorities as well. These “non-traditional’’ students had disproportionately come to CS via non-traditional paths (such as via psychology or linguistics) and disproportionately lacked the prerequisites as a result. The retooling of first-year CS as a weeder course also resulted in a competitive atmosphere that deterred many women.

The personal computer also led to further masculinization of computing [8]. Five reasons thought to have reduced female participation in the 90s were: the rise of video games, subsequent changes in stereotypes/perceptions of computing, the encouragement of boys to go into the field and not girls, an inhospitable social environment for women, and a lack of female role models [8].

The birth of the World Wide Web in the 90s and its spread beyond academic/military use led to a second bubble in CS enrolments. The hype of the dot-com bubble and the promise that a CS degree would lead to easy prosperity
led to a resurgence in enrollments in the late 90s. The dot-com bubble burst in 2000 – and enrollment with it a few years later [6]. Indeed, the NASDAQ has been found to be a predictor of CS enrolment at Stanford [9]. The perception of CS jobs as being volatile has also been implicated as a reason why women are deterred from CS careers [10].

The boom-time in the late 90s and early 00s led to a return of strict enrolment controls and a spree of hiring more CS faculty [6]. These boom-times also reduced the amount of service teaching: with CS programmes overburdened, CS departments had few resources and little motivation to teach non-CS students. At some universities, departments such as physics or math began offering their own CS classes to their own students – leading to CS becoming increasingly isolated from the other sciences – and from non-traditional students.

When the bubble burst, the “get-rich-quicker’’s disappeared – and CS departments were left trying to get more “bums in seats’’. Enrolments did not recover again until the mid 00s – and have been on the rise since [6]. Overall, a pattern of cyclical enrolment emerges. Boom times lead to more students, then more enrolment controls; bust times lead to more outreach. Bust times also result in disproportionately many women leaving the field, or not going in at all [6] – indeed, as of 2011, 18% of CS students are female [5].

Enrollments in CS are now skyrocketing again: the 2012 Taulbee Survey found that CS enrollments have risen for the fifth straight year [10]. Facing packed classrooms and overburdened teaching resources, some CS departments
are once again considering cutting their interdisciplinary programmes and service courses. Hopefully this time around we’ll have learnt from the past.

References

1. Rossiter, Margaret W. Women scientists in America: Struggles and strategies to 1940. JHU Press, 1982.
2. Luker, Kristin. Salsa dancing into the social sciences: Research in an age of info-glut. Harvard University Press, 2008.
3. Ensmenger, Nathan. The computer boys take over: Computers, programmers, and the politics of technical expertise. MIT Press, 2010.
4. Normalizing Female Computer Programmers in the ’60s
5. Ashcraft, Catherine, Elizabeth Eger, and Michelle Friend. Girls in IT: The Facts, 2012.
6. Slonim, Jacob, Sam Scully, and Michael McAllister. Outlook on Enrolments in Computer Science in Canadian Universities. Information / Communications Technology Council, 2008.7. Roberts, Eric S, Marina Kassianidou, and Lilly Irani. “Encouraging women in computer science.” ACM SIGCSE Bulletin 34, number 2 (2002): 84–88.
7. Camp, Tracy, and D Gurer. “Women in computer science: where have we been and where are we going?” In Technology and Society, 1999. Women and Technology: Historical, Societal, and Professional Perspectives. Proceedings. 1999 International Symposium on, 242–244. IEEE, 1999.
8. McGettrick, Andrew, Eric Roberts, Daniel D. Garcia, and Chris Stevenson. “Rediscovering the passion, beauty, joy and awe: making computing fun again.” In Proceedings of the 39th SIGCSE technical symposium on Computer science education, 217–218. SIGCSE ’08. Portland, OR, USA: ACM, 2008. isbn: 978-1-59593-799-5. doi:10.1145/1352135.1352213. http://doi.acm.org/10.1145/1352135.1352213.
9. Cohoon, J. McGrath. “Women in CS and biology.” SIGCSE Bull. (New York, NY, USA) 34, number 1 (February 2002): 82–86. issn: 0097-8418. doi:10.1145/563517.563370. http://doi.acm.org/10.1145/563517.563370.
10. McGettrick, Andrew, and Yan Timanovsky. “Digest of ACM educational activities.” ACM Inroads 3, number 2 (2012): 24–27.