The double-edged sword of gender equality

Posted on by Randy Olson

Yesterday, I charted the trends in Bachelor’s degrees conferred to women in the U.S. using the NCES 2013 Digest of Education Statistics, and found that there has been a glaring gender disparity in Computer Science and Engineering majors for more than 40 years. In that post, I focused on the gender disparity in CS&E, but my readers were quick to point out that there are also glaring gender disparities in favor of women in certain majors as well.

To better visualize these gender disparities, I inverted yesterday’s chart and plotted the percentage of Bachelor’s degrees conferred to men over time below. You can download the data set used for this chart here.

percent-bachelors-degrees-men-usa

The four most female-dominated majors in the U.S. are:

  1. Health Professions (85% women): nursing assistant, veterinary assistant, dental assistant, etc.
  2. Public Administration (82%): social work, public policy, etc.
  3. Education (79%): pre-K, K-12, higher education, etc.
  4. Psychology (77%): cognitive psychology, clinical psychology, etc.

These four majors alone comprise 25% of all college degrees conferred nowadays, yet most conversations surrounding the degree gender gap focus on the relatively unpopular Computer Science and Engineering degrees (only 6% of all college degrees today, combined).

Why do we disproportionately focus on majors where women are underrepresented, when there’s a much larger set of majors where men are vastly outnumbered by women? Especially given the fact that women earn 60% of all college degrees in the U.S., there seems to be some severe reporting bias going on here.

For example, male educators are often barred from being left alone in a room with a student, for fear that he will sexually harass the student when no one is watching. (Note: Female educators are allowed to be alone in a room with a student without issue.) That’s akin to preventing a female computer programmer from logging into her laptop without supervision, for fear that she’ll go watch porn instead of work. How is that anything but gender discrimination?

In the Health Professions, men face an altogether different kind of gender discrimination. Perhaps best highlighted by the 2000 film Meet the Parents, male nurses are often the butt of a joke because they’re working in what is considered a “traditionally female” profession. Societal pressures steer men away from these “traditionally female” professions, just as they do for women in professions that are “traditionally male.” Yet how often do we hear about the plight of male nurses?

Gender equality is a double-edged sword. If we want to work toward gender equality for women, we need to work toward gender equality for men too. Women aren’t alone in their struggle against discrimination. It’s about time we acknowledged that.

About

Randy is a PhD candidate in Michigan State University's Computer Science program. As a member of Dr. Chris Adami's research lab, he studies biologically-inspired artificial intelligence and evolutionary processes.

Posted in data visualization Tagged with: , , , , , , , , ,
  • Matt

    Thanks for flipping the study. I read the other one and I was surprised at how few professions men could actually be in and be equal.

    • http://www.jake4maths.com/grapher/ olivia

      Great topic, In NZ last year 30% of girls got high grades in their last year of school compared to 6% of boys…. Not sure about the states, but I know the reason behind that disparity is because education has been adapted to suit females more than males. Not sure if this is a good model for society though?

  • Carol

    Nice to see the study flipped, and gender inequality shown on both sides. Sexism is shit for everyone!
    I would be interested in a comparison of the salaries of the careers that these majors often lead to, between the two genders. Perhaps this is why the computer science/ engineering is more focused on as these can lead to higher paying jobs?

  • [email protected]

    It’s pretty obvious – right now tech workers shape the world and have better pay. Besides, if men were interested in the professions you mention above, I’m pretty sure they would be dominated by men.

  • Erik

    Hi Randy,

    This has been a really interesting set of articles and it’s really good that you are focusing on both women’s and men’s inequalities. I wasn’t aware, for example, that women were earning 60% of undergraduate degrees. I earned a degree in biomedical engineering and my class was definitely ~60% female, but like you pointed out in your previous article, most engineering fields are seriously skewed the other way (as evidenced by the gender ratios in any of the general engineering classes that I took).

    TL;DR for below: Men who take positions in fields such as nursing or secretarial work are the butt of jokes because they are taking roles that are somehow beneath them. Women who get pushed out of science/tech/engineering are pushed out because they are told they can’t hack it. We should be pushing for men to stop being stigmatized for doing jobs that require empathy and may have less financial gain. We should be pushing for women to be treated as equally capable. Ultimately the hope is that “men” and “women” in the last two sentences can be replaced with the word “people”.

    ———

    I will however point out that your example of male nurses is patently different from your example of discrimination in the Eng/Tech fields. Women are generally pushed out of Eng/Tech because of attitudes (fervently held by an unfortunately vocal minority and implicitly accepted by an unfortunately larger general population) that women are somehow not cut out for these difficult fields.

    Men can also be told that they are not talented enough–and they frequently may be told that–but generally not because of their gender. The exception to this would tend to relate to ability to nurture or empathize and that does touch on your example about men in the health professions. However, doctors also need to be empathetic and men don’t get shit for becoming a physician. I’ve certainly never gotten any in my pursuit of an MD degree. Men (most often) get mocked for becoming male nurses because of a cultural conception that nursing is somehow beneath them. The jokes don’t suggest that male nurses are bad at their jobs, but rather that because they could be good nurses that they must be less manly. And this is a serious problem. But it is a fundamentally different problem. Men are (most often) mocked for taking less prestigious positions. Women are (most often) told that they can’t hack it in difficult fields.

    I think this actually relates strongly to the trends you’ve noted about which fields have been and continue to be dominated by one or the other gender. If you look the fields from most male-dominated to least, you get the following: Engineering, Computer Science, Physics, Architecture, Math and Statistics, Business, Social Sciences and History, Agriculture, Biology, Art and Performance, Communications and Journalism, English, Foreign Languages, Psychology, Education, Public Administration, Health Professions. Note that there is a rough correlation here with prestige, potential for financial gain and perceived difficulty. The primary exception is Health Professions. However, health professions encompasses all of that support staff like nurses and technicians. If you look at just physicians, females actually account for around ~50% of medical school matriculants in 2012 [1]. So the female dominance that you see in the health professions field is primarily accounted for by those who enter the support-type roles.

    I was able to confirm that the 60% female undergraduate matriculant rate is approximately accurate [2], but I wasn’t able to find anything on the gender breakdown of the applicant pool so I can’t reliably say anything about whether or not there is bias in the application process with regards to gender. I can only note that that the male-female ratio in the 19-24 age bracket in the US is ~51:49 percent so if the applicant pool is reflective of the overall population ratio then I would think that the app process is female biased. However, it could also be that fewer men are applying and then the onus would be to determine what cultural or K-12 factors are preventing men from applying.

    Given that there are 60% females in undergraduate college, if you make the assumption that men and women are innately equally capable of succeeding in a given field, then the balance point at which a field favors women is when there are >60% women in an education discipline. Conversely, this can be stated as when there are <40% men. Looking above, that means that the fields which favor men are Physics, Architecture, Math and Statistics, Business, Social Sciences and History, Agriculture, and Biology. The fields that favor women are Art and Performance, Communications and Journalism, English, Foreign Languages, Psychology, Education, Public Administration and Health Professions. That split really speaks even more highly to the relation of prestige, financial gain and perceived difficulty to a discipline's gender orientation. One could potentially posit that the less financially focused fields are more female dominated also because men have to think about supporting their families and have that unfair pressure that women do not. I will note, however, that this mindset stems again from the idea that men are somehow more capable and that women would not be able to provide as well as men could.

    I apologize for the wall of text, but this is a topic that I have thought about a lot and means a lot to me. As a guy with a lot of female friends in the technical fields, I have seen some of the pervasive attitudes that they have had to endure and it deeply upsets me.

    Thanks for listening, have a picture of a cat: http://imgur.com/gallery/20NsE
    Erik

    [1] https://www.aamc.org/download/277026/data/aibvol12_no1.pdf
    [2] http://www.forbes.com/sites/ccap/2012/02/16/the-male-female-ratio-in-college/

    • onething

      So I agree with nearly everything you said. But one thing toward the end irk’d me.

      To say that a particular degree is biased if it is putting through (64:36) but a 20% difference is touching on bias (60:40) -> (68:32) )

      • onething

        what? it cut out the bulk of the middle. nevermind.

  • Pingback: Percentage of Bachelor's degrees conferred to women, by major (1970-2012) | Randal S. Olson()

  • Doug

    Why is the conclusion that everyone’s discriminated against? Why isn’t the conclusion that men and women make different career choices? Why is it a foregone conclusion that if there were no discrimination, men and women would make up exactly half of each area of study?

    • http://drabbiticus.com Erik

      I agree that men and women most likely do make different career choices. Yet, choices aren’t made in a vacuum.

      We have a culture that reinforces ideas of innate gender roles. Little boys are told every day that they need to be tough when they get hurt, that they shouldn’t be a crybaby. Little girls are told that they can’t handle rough housing, that math is too hard for them. Little girls are told they should be friendly and “nice”, that they should play well with others and build communities, mediate between friends. Little boys are told that they need to be responsible and be able to take care of people.

      These attitudes are shifting. I am glad that they are shifting. When we tell boys and men that they have to be tough, we create a culture that makes it harder for men who are struggling with things things like depression and addiction to seek help or even admit that they have a problem to their friends. When we tell girls and women that they are weaker and less capable in certain fields, they begin to internalize those values. They begin to believe that they really aren’t cut out for it, that it is too hard.

      There is a lot more that can be said on this topic, and to some degree it can be debated about just how “innately” different men and women are. However, the problem is that this viewpoint considers things in the aggregate. In life, you are an individual. That means that you should be able to deviate from what “men” are like or what “women” are like just because that is the way you are. So why do we tell little girls and little boys that they should be one way? If there is a difference between men and women, then we should be able to just let that difference exist, without trying to shape people to fit into our view of what those differences might be.

      In general, a humans, we like to categorize things. This has been an immensely useful skill. It underlies most of our science and most of understanding of the world. It is fundamental to computer science since it allows us to do things like DRY (Don’t Repeat Yourself) and OO (Object-oriented). The problem is that when we see fields that are largely male dominated or largely female dominated, many people draw conclusions about what men and women are like. Relatively few stop to consider what societal pressures might have shaped the decisions of the people who enter these fields. They chalk it up to innate differences between men and women. And then they question people who don’t fit into those molds that they have created in their own minds.

      Think back to those male nurses and female coders that we keep talking about. They don’t seem to fit into a mindset that says that men prefer to do things like code and women prefer to do things like caregiving. People express surprise and sometimes even censure because these people don’t fit. If we push men out of nursing, we have an even more female-dominated nursing population. We reinforce the idea that women are caregivers. Because why else would so many nurses be women? We’ve created a cycle. This is why I get concerned when the fields are so disproportionate. It’s not just that men and women are making different choices. It’s that we have created attitudes that reinforce those choices and make it harder for people to make different choices. It makes it harder for people to succeed when they make a choice that strays from our mental guidelines.

      Again, sorry for the wall of text. Have another kitten: http://imgur.com/gallery/KWDadE1

  • tomas

    Well it’s nice to see this discussion and have a chance to get through to someone who is competent at statistics, I think you will understand if you investigate the following concern yourself.

    Part of the problem is that women were NOT 30% or more of computer science majors at any point in time. That claim is simply false. The source data you’re looking at is completely fraudulent. It’s been debunked before.

    Look at the source data.

    For one, tens to hundreds of thousands of students “disappear” from other majors. Look at the absolute numbers. Since no one has ever noticed or accounted for this, that isn’t actually what happened; in producing these data other categories were misclassified. Correspondingly, one can notice there are far more majors in certain CS and engineering fields for a few years in the mid-1980s than there are in the surrounding years before and after. If you logically considered at this, it’s already ridiculous that there were far more majors in certain fields in the 1980s than close to the present day, the 2000s, with far more modern technology and booms in engineering students. There is especially no reason for there to be sharp discontinuities on short timeframes – no reason for a year like 1985 to have double the numbers of 1980 or 1990. There is no legitimate reason for this except for reporting error.

    What clearly happened, in the most generous interpretation, is some sort of typo or coding error in producing their database tables. Wrong data is presented for a few years in the 1980s, when in all likelihood college major trends (like absolute number of students, along with the sex ratios you are attempting to discern) probably mostly proceeded in a monotonic fashion from say 1980 to 1990.

    • http://www.randalolson.com Randy Olson

      tomas, while your claim may be possible, I haven’t seen any thorough studies showing that this data set is in error. It’s a data set that has been carefully curated by the National Center for Education Statistics (NCES), which is the primary federal entity for collecting and analyzing data related to education in the U.S. This isn’t a data set curated by Joe Schmoe down the street in an Excel sheet; collecting this data and making sure it is correct is these people’s primary job.

      • tomas

        Have you looked at the data, again? Not just the derived data for proportion of students by sex if you got that either from a script or from a separate table without seeing the raw numbers.

        http://nces.ed.gov/programs/digest/d10/tables/dt10_317.asp
        http://nces.ed.gov/programs/digest/d10/tables/dt10_323.asp
        http://nces.ed.gov/programs/digest/d10/tables/dt10_324.asp
        http://nces.ed.gov/programs/digest/d10/tables/dt10_314.asp
        http://nces.ed.gov/programs/digest/d10/tables/dt10_316.asp

        The problem is not with the overall calculation of sex ratios (nor necessarily in all fields, STEM and non-STEM alike, though there might be other errors), but that the raw numbers are totally borked for computer science. This does make the derived male-female percentages unreliable but that’s not even the real story.

        It’s true we can’t know exactly what is wrong. It could be along the lines of Southern Horse College reported 10 bachelor’s degrees granted in one discipline and that was somehow entered into a spreadsheet as 10,000. It could be any number of more complicated code glitches. In principle someone might have the raw data needed to resolve exactly what we see but the evidence is on the side of error.

        What we see from the raw data is relatively minor and smooth transfers, growth, or decline in raw numbers in other related disciplines that are actually plausible as time varying trends. Sure there will be fluctuations and noise, but nothing to really account for the short time scale trend in the CS field that nobody has claimed to have actually experienced. Effects we can account for from what we do know actually happened historically and external to any one field of study are also evident. The effect from the end of the Baby Boom is understandable and natural, where in the 1990s Generation X cohorts were smaller in population size in general despite the still continuing trend of increasing proportional enrollment in college. This is reflected in pretty much every other STEM field except for our anomalous exception.

        If you actually believe the information is legitimate and not a glitch, then you have more a real story to focus on anyway. The real story is what happened to tens of thousands of students for a few years in the 1980s, why some limited number of majors saw explosive growth and declines in a few years and what that did to those disciplines.

        The serious question for the field would not be about the gender balance so much as why for both males and females numbers massively collapsed and it took until the tech bubble at the turn of the millenium and the generally increased total student population today to recover. (Except this didn’t happen, and no one noticed because the error is not in reality but in one faulty statistical series put out by some bureaucrat. Once can also notice trends in masters and doctoral degrees as expected are much more monotonic – and for those postgraduate degrees that disproportionately includes non-resident aliens being brought into departments anyway. This again matches every other similar science and technology field but is compounding evidence of error. If there weren’t large departments, plentiful professors to teach students, or higher degrees being granted were all these phantom degrees in the field really being produced?)

        • http://gasstationwithoutpumps.wordpress.com gasstationwithoutpumps

          The fluctuation in the numbers of CS majors has been huge: particularly during the dot-com bubble. While other majors have seen gradual changes in the numbers of students, CS has been subject to boom-and-bust cycles. The numbers in the graph are consistent with the what CS professors have observed in their own institutions.

          The bursting of the dot-com bubble caused a lot of women to leave CS (or not join it in the first place), accounting in part for the rapid change around 2000.

  • sweatpants

    What about Home Economics? Had that been phased out by 1970 so as to be insignificant?

  • http://www.endofwomen.blogspot.in namae nanka

    “These four majors alone comprise 25% of all college degrees conferred nowadays, yet most conversations surrounding the degree gender gap focus on the relatively unpopular Computer Science and Engineering degrees (only 6% of all college degrees, combined).”

    The same is with the maths gender gap(which favors boys) while reading gender gap(which favors girls), which is a fundamental skill, still gets scant mention and efforts considering it’s about thrice the maths gap(on PISA).

    The reading skills gap also affects the maths scores, if you can’t read you can’t do anything, maths included. So the conundrum for gender-equality is that if reading gap is closed the maths gap would grow wider.

    “What was it about the 1970s and early 1980s that made Computer Science more welcoming to women? And what changed?”

    Perhaps the field itself changed from 90s onward.

    • tomas

      No, again, the statistics are simply completely false.

      If the blog author had made a post saying statistics show that raccoons abduct millions of American babies from hospitals every year I’m sure some of you people would be defending it. Commenters would show up with their own pet theories to justify why raccoons absconded with millions of babies every year instead of seeing that the information in the first place is false.

      There were never tens of thousands more computer science majors than graduates in all other engineering fields. The source data here has some unknown coding error it, that much is obvious.

      • Luke

        “USA Today has come out with a new survey—apparently, three out of every four people make up 75% of the population.”

        – David Letterman

    • David

      I noticed that as well. Interesting that the author pointed to CS degrees, but made no comment about that.
      The graphic doesn’t tell that story. Maybe it was combined degrees (CS + Music, or CS + Art, or CS + History). Maybe it is simply an uptick for women across other degrees.

  • Joe

    Hi,

    thanks for your work and for bringing in your statistician’s understanding!

    What struck me immediately when I saw the graph was that there seems to be a general trend towards more women in most majors, “Computer Science” being the strongest exception. How do the numbers account for the overall growing percentage of women in university education? I.e., are women more likely to choose Engineering or are there simply more women now?

  • David

    Based on most of the comments (grind those axes), I’d think the graphic could be simplified to show 1970 vs 2010. Now and Then. Very little discussion about what is in between. It just adds to the clutter.

    • andrea

      I think it’s useful to show that there haven’t been any huge surges or troughs, with the exception of computer science around 1980.

  • Pingback: Bachelor Degrees to Women (and Inversely, to Men) Over Time | Exploring the Challenges and Opportunities for Women in Business, Technology and Leadership()

  • Marie desJardins

    Tomas, as far as I can tell you don’t have any evidence that “the numbers are wrong” other than your intuition that “the numbers don’t look right to me.” There are a lot of potential reasons why CS/information systems have showed different patterns than other fields. One big reason for the explosive growth in the mid 80s is that that’s when many universities *created* computer science degrees. So you couldn’t major in that field at many places before 1985 or so. Then we had a tech bubble in the 1980s, followed by a tech downturn, and another bubble in the 1990s, followed by a downturn. You see exactly that economic pattern reflected in the number of majors. This just isn’t very surprising and it certainly doesn’t mean that there is some kind of massive error in the numbers. (These numbers are consistent with CRA’s Taulbee data, too — do you think they’re *both* wrong?)

    The decreasing percentage of women in computer science is very much a fact. Yes, there really were about 1/3 female CS majors in the late 1980s. (I was there. I remember.) And yes, there are really about 12-15% female CS majors now. (I’m still here. I see it.) It’s very depressing. It is NOT just because “women don’t like CS.” The sociocultural pressures reinforcing the male dominance of the field are incredibly powerful. It’s a complex phenomenon and one that a lot of people have thought about, researched, and written about. You could learn more if you would like — it’s not that hard to find more information about these issues on the Internet.

  • Pingback: Somewhere else, part 143 | Freakonometrics()

  • Anna

    “In the Health Professions, men face an altogether different kind of gender discrimination. Perhaps best highlighted by the 2000 film Meet the Parents, male nurses are often the butt of a joke because they’re working in what is considered a “traditionally female” profession. Societal pressures steer men away from these “traditionally female” professions, just as they do for women in professions that are “traditionally male.”

    Men are steered from these traditionally female professions because, as you said, they are mocked for being in a feminine–subordinate–role. Women are not steered away from traditionally male professions for the same reason, as they’re just seen as plain old incompetent/unable to fill the man’s role. You really can’t call both those things gender discrimination on the same level when men are pushed away from certain careers because it’s a shame to be seen as feminine.

    If you want to really get to the bottom of why the genders are dispersed like this, you have to go back even further historically, and you will see that, again, much of it stems from the same source: a gender hierarchy which places women as subordinate to men, not equal gender discrimination against both sexes. Jobs like social work, nursing, etc. were deemed as appropriate jobs for women based on their assumed nurturing nature. However, when engineering became more of an established field, there was no concept of women being able to do those jobs because they were seen as incapable, too ‘emotional,’ ‘irrational’, etc., so they were kept out.

    Also, another essential point is that the professions men go into more make a lot more money and are, generally, more respected than woman-dominated professions. This is why we focus more on the lack of women in those fields because their underrepresentation is a huge part of why there is such a wage gap between the sexes. I think, too, that we should work towards legitimizing and appreciating the necessity of women-dominated jobs, along with encouraging more women to go into STEM fields.

    While I do think we should even out the genders in all fields, I really don’t agree that all the gaps are due to gender discrimination against both sexes. Really, it has more to do with a gender hierarchy, rather than evenly distributed discrimination. If women and their traditional jobs weren’t subordinated, than male nurses would not be seen as a butt of a joke.

About this blog

The data visualizations on this blog are the result of my “data tinkering” hobby, where I tackle a new data analysis problem every week. If I find something interesting, I report my findings here to share with the world.

If you would like to use one of my graphs on your website or in a publication, please email me.

Archives

Subscribe

Enter your email address to subscribe to this blog and receive notifications of new posts by email.