The Myth of Female Software Developers

One of the recurring debates in the software development community is the lack of female software developers and the underlying reasons. Some of the arguments we've heard recently include:

  • Working with computers is a "solitary" activity which attracts more men than women
  • Females and males have abilities in different areas and it's only natural for certain jobs to be dominated by a particular gender.
  • The male-dominated culture of software development teams is putting females off

However very rarely is any attempt to back up these arguments with hard data, which is unfortunate as when we look at the data it turns out none of these arguments hold true.

We're going to offer a data based analysis and look at the data tracking students from school level (age 14) to university in the UK to see where precisely the gender drop-off occurs.

To get onto a good Computer Science degree you generally need to have a strong grade at A-Level Maths (which students study between the ages of 16-18), and in order to study A-Level Maths you need to typically have grade A/A* at GCSE Maths (at age 14-16). The reason behind these requirements is that performance at Maths A-Level is the best indicator of performance at degree level Computer Science. 

So let's have a look at the gender breakdown at each of these stages and also the data for Maths degrees which have similar requirements:

 

 

 

(Data shows gender breakdown for GCSE and A-Level for students getting A/A* grades)

So as is clearly visible far fewer females study Computer Science than Maths at either A-Level or degree level. So clearly it's not an issue of ability or a failure to meet the requirements to get onto the degree, but rather one of choice. Clearly at age 17 female students who are qualified to get onto a Computer Science course are choosing not to do so.

We can however see that this choice is being made even earlier if we look at A-Level choices between the genders:

 

 

At age 16 girls aren't choosing Computing, but they are choosing IT and Maths, so it's not the technology basis nor the theoretical basis that's putting off female students, but perhaps it could be the combination of the two. IT gives students immediate practical skills, while Maths gives students the theoretical basis needed for studying many other subjects. The theoretical basis of Computing on the other hand can come across as not having any practical applications.

It's also possible that Computing is considered too "high-risk", students who pick it have no idea if they're going to be good at it (as they've never had a chance to do it before) and don't want to risk getting a bad grade that will hurt their chances of getting into a good university. This risk would presumably be worth taking for students planning to study CS at university (i.e. predominantly males).

 

Why is gender equality important for software development?

Inevitably in these debates someone will ask "Why does it matter?" - we strongly believe it matters not only for the software development industry, but for society as whole.

There's a huge shortage of software developers in the UK, the number of software development roles is increasing by approximately 10,000/year. The UK just isn't producing enough talented software developers to meet the demand. When it comes to expanding the talent pool it makes sense to target the groups who are most under-represented.

But it is also an issue for society as whole. Having a 90:10 ratio of males to females for software development should be as shocking as having a 90:10 ratio for literacy. Software development is fast becoming one of the fundamental skills of the 21st century as technology starts to dominate every industry. Of the 26 billionaires the web has produced in the last decade, only one has been female. Only a tiny fraction of technology companies started today are started by female technologists. By neglecting female adoption we're creating the potential for huge disenfranchisement issues in the future.

How can we fix it?

Students by-and-large don't study Astronomy because their love of Dopplr shift equations, but rather because their love of space. Similarly we should make clearer the link between theoretical Computer Science and the practical side of web and mobile applications that many students use on a daily basis.

Movies like The Social Network are just as important as outreach programs, making that link part of societal knowledge. We need to project the image that Computer Science lets you build cool projects.

Introducing CS at any earlier age would help too, both with increasing interest and allowing students to try it without risking damaging their future career prospects. Last year saw the introduction of GCSE Computing which means students can now chose to study it at age 14. We won't see statistics on uptake for a couple of years but we can be hopeful that it will mean an increase in the number of female students studying the subject.

36 responses
So your solution is more techy movies? that's silly. You almost had it right at the beginning. But It's not that there is too many men in the field, turning women off, it's the type of men in the field that tends to turn women off. No one wants to deal with the obvious because it means pointing the finger at oneself and saying "im undesirable".
...as when we look at the data it turns out none of these arguments hold true.
Where is the rest of my comment? I typed two more lines.... What I intended to say is: you've disproven the second of those three arguments with your very nice analysis. Do you also have data to disprove either of the other two?
The data shows that there is a dropoff and when the drop off occurs. But it does not say why it occurs.
Your analysis says that kids see computing as "high risk" but you do not analyze why boys chose a high-risk career and girls do not. As a former boy, I knew that I would be working my whole life. For most, stay-at-home-dad is not an option. If computing is risky (even if it just perceived that way by students), why are boys choosing programming more than girls?
I would have never tried a Computer Engineering major, if I haven't had learned how to program from highschool, and found it fulfilling to spend hours and days with it. And still, from a class with 50% girls, 10 years from graduation, the ratio of girl programmers today is 20% of total. (Anyways, a class of 30, hardly classifies as statistical data, this is just my own experience of studying in Romania).

Highschool of informatics profile, ~50% girls.
University of Computer Engineering, ~30% girls.
Industry, my company's local branch today, I work at Microsoft Copenhagen, between 15% and 20% girls.

What happens between highschool and industry, I have no idea.

Females are really under-represented in the programming industry, and whatever can be done to correct this should be. I'm interested to see the discussion that this article generates - thanks! :)
There are many ways to make computer programming appealing to young women. My project may seem a bit like pandering to the girly-girls, but I've received more interest from men! Please read my interview, you might come up with other ways to make programming a means to reach a goal, a goal that girls want. But apparently, guys want this too. So, go figure out something cool and develop it. Not to teach teachers how to teach STEM, but to teach students to do something artistic or sculptural (like a series of lessons and programs, diagrams, etc. for making kinetic sculpture in art class with arduinos, not just how to use an Arduino in an after school robotics club). Don't teach the teacher, teach the kids.

http://www.linuxpromagazine.com/Online/News/Tau-Meta-Tau-Physica-Bringing-Ope...

I am a female studying Computer Science, and your first diagram certainly tallies with my experience.

Although your data seems to disprove it, I think that the second point is still quite a plausible reason as to why women don't have more presence within CS degrees.

Also, there is a big difference between the abstract concepts of mathematics and the technical concepts of computing, both of which are fundamental aspects of CS. Many women I know simply aren't interested enough to want to understand the technicalities of computers and how they work, but they still have the ability to understand complex mathematical concepts.

I think young women believe that programming will result in working on missile projects, corporate infrastructure and databases, and all manner of really sterile and unfulfilling work. About as attractive as assembly line work just with code.
I'm a female developer and wrote a blog post about my thoughts on the matter a little while ago - http://www.codefrenzy.net/2009/03/01/girls-in-it/ :)
Reminds me of a study done a few years ago (this is absolutely true) statics to find out why teenage boys masterbate. Talk about a waste of time and money!
Didn't need a study for that either.
I don't know why I don't want a job in the fashion industry, but I know I don't. I don't know why most heterosexual males don't care about fashion, but we don't.
Not sure why people waste time and money to quantify something they already know.
I do know things like this don't matter and are not an issue until someone makes it an issue.
People are different and they do what interests them, it's not a mystery.
It's important to make working with computers real. Lots of heterosexual guys are interested in my fashion design project, most of them are half my age. It's an issue of personal choice. What a person is presented with in early life determines their personal choice. So let's get fab labs installed in schools with hands on interesting ways to use computers with art or writing games or something with a point to it that is appealing to adolescents. More fathers share mechanical, electrical, computer, and engineering skills to their sons than to their daughters. This is basis of the problem. Guns vs. dolls isn't the problem. It's what you can do with computers that appeals to the gun-loving kid and the doll-loving kid. We've been ignoring the doll lovers.
So as is clearly visible far fewer females study Computer Science than Maths at either A-Level or degree level. So clearly it's not an issue of ability or a failure to meet the requirements to get onto the degree, but rather one of choice. Clearly at age 17 female students who are qualified to get onto a Computer Science course are choosing not to do so.

Major logic fail.

You've ignored the possibility that they're being rejected after applying.

As a United States mathematics undergraduate degree holder who is now a software engineer, and who later sought and attained a graduate degree in computer science, let the record show that studying math doesn't necessarily preclude one from choosing a career in software development.
16% of CS majors are female, and make up only 9% of the programming work force.
The phenomenon that most males recently graduated with CS majors are offered jobs in programming (which makes them programmers) whereas most females recently graduated with CS majors are offered jobs in Quality Assurance or Desktop/User support (which makes them IT) may have something to do with it. Then these females drop out of the computing business because QA and User Support has no future, they're treated like computer janitors, and they can't transfer over to programming, despite assurances at interview time that they'll be able to do so in the future.
@handelaar acceptance rate for Computer Science is roughly the same for both genders. Most students apply to a range of universities with different levels of entry requirements, so they'll be sure to get in somewhere. So most CS applicant do get in, although not always to their first choice university.
Could programming be correlated with low fertility in women? I. e. that women who are good at these things have on average, worse fertility than the average woman? In that case it would make sense for a programming-apt woman to not advertise this skill and instead pursue a career that is associated with better fertility. Studies would need to be done of course.
Your argument is incoherent. People *do* study astronomy because Doppler shift equations rock. People study fields for the hard and interesting challenges that they offer. Sadly, computer science isn't upfront about how cool it gets when you get to studying things like, say, SAT algorithms, compilation, formal language theory, etc: all the intro courses are about repetitive, sterile programming tasks. Without cultural norms exposing women to the fun parts of CS outside of those courses, they have no motivation to stick around.

Also, CS undergrads are by and large socially inept chauvinists. I know, because I am one (socially inept, and probably a chauvinist).

So it isn't that we should do "outreach," it's that we should try and get the most interesting parts of the subject into intro courses, and be less personally unpleasant.

All of the arguments about young people doing "risk assessments" about future career paths and women deciding against computing because its risky or they want to have babies is incorrect in my opinion.

People become programmers because they discover that programming is a fabulously interested, fun an rewarding hobby. And then they need to get a job. All the really good programmers I've ever met (I've met lots) program for fun in evenings and weekends and have their own side projects outside of work. People without that sort of passion can't compete in the professional world and end up moving into 9-5 systems, support or management jobs or dropping out of IT altogether. Nothing wrong with that, its just what happens.

So the why question comes down to, why don't young women enjoy programming in their spare time, the path that most assuredly correlates with programming as a career? I don't know, but thats the right question to be asking if you're one of these people that worries about these things.

As for recruitment biases. I would prefer more women in the software teams I work in but I'd always hire the person who codes for fun in the weekends and loves coming in to work over the 9-5er every time.

> There's a huge shortage of software developers in the UK, the number of software development roles is increasing by approximately 10,000/year. The UK just isn't producing enough talented software developers to meet the demand.

There is lot of talented programmers in Poland, many many beautiful womens among them ;) Outsourcing to Western Europe has advantages.

What is career path for freshly started coder? Usually there are two options, became manager or some sort of geeky ubercoder. I think women would usually go with management path.
Boys approach technology and IT since childhood, with a multitude of toys.
Girls, on the contrary, are kept away from that kind of entertainment because the market (and society, etc.) puts a gender connotation on toys. This is such that there are "girl toys" and "boy toys", and for some reason, girl toys rarely include science/IT/tech oriented ones (apart from those awful pinky sparkling gadgets).
Imho, giving girls (and boys) the chance to pick up what to play with, without gender restriction, would be a good start to change things.
Here a nice representation of the concept: http://www.smbc-comics.com/index.php?db=comics&id=1883 :P
"To get onto a good Computer Science degree you generally need to have a strong grade at A-Level Maths"

This is a common and unfortunate misbelief in the computer degrees. You only need the maths for scientific IT, and this is actually very often related more to the scientific aspects than it is to IT by itself. I mean, I still need to see when a network admin, a DBA or a business programmer do mathematics. For the modern IT (i.e. nowadays), you need pure logic, not the maths.

Now, for the general topic of the discussion, I believe a good career in IT is very hard to combine with the responsibilities the women have after work. I don't know how you can keep your competencies up-to-date in this extremly fast moving business if you have to cook, clean, wash, take care of the children, ... after work; rather than be reading books and learning new skills.

I think that in analyzing the factors that impact female enrollment in computer software programs, it is helpful to classify these programs as "Software Engineering." The factors that keep women away from Engineering are almost identical across the board, from Civil and Mechanical to Software and Computer. I can attest as a female engineering student that the most discouraging aspects of engineering was the blatant sexism. When a 17 year old boy says he wants to be an engineer people asks questions like "Where do you want to go?" and "What type?" When a 17 year old girl makes the same statement she is asked "Why?" and "Are you sure?"

The integration of women into traditionally male dominated fields is up against a vicious cycle of sexism. In which girls are constantly told they can, but cautioned that they will have to work harder and prove themselves more often. That no matter how competent they are there will always be men who disrespect their work on the basis of their sex. That no matter how hard they try, they will be passed over for promotion in favor of less senior men. That no matter what they accomplish, there will still always be someone who firmly believes that they got an "easy ride" because they are a woman.

Until women are common in the field, there will be sexism. And until the sexism stops, women will be discouraged from choosing that field. It is a vicious cycle solved only by time and those few of us who hear these warnings and are brave enough to say "I'm going to do it anyways."

I am a female coder and have been for a number of years, with skills in various languages. In my current position, I find myself doing more systems administrative and application support work than coding. Mostly because my boss is passing most of the coding tasks to other staff (all men) and giving me support work; though originally I was told at my interview that I would be doing coding and development work. I believe this is the case for a number of women who have skills in software development but are simply not given the opportunity to further develop these and instead find themself side-lined into support roles. It's like there is a perception/misconception held by bosses that if you are female you can't code well.
ok, I'm not a programmer, and really had no interest in it when I was at University (a bad experience with fortran put me off!) But I did do a science degree and then a math heavy economics masters - so there is a part of me that likes playing around with this stuff.

I'm currently more on the 'social' side of IT - applications of applications and open data in the public realm. But I constantly come up against "oh you probably don't want to be a part of that...it's very technical"...I don't hear this being said to men. I passed up a good job to be a part of a big development project - and now I'm sidelined from development - and dealing with 'culture' and support. I recognise it's not just my gender though... but it's a big part of it. I have more technical skills than others who are involved with development.

.... Without cultural norms exposing women to the fun parts of CS outside of those courses, they have no motivation to stick around...
As a woman, I have always found myself sidelined into QA despite always saying that I would like to become a programmer. Everytime I try to get into the development side, people always ask me why I don't want to be a business analyst or a project manager.

This leads me to wonder if people sub-consciously see women as having better communication skills than the guys but weaker programming skills?

I have actually done freelance work programming in .NET and have done many IT courses on top of my computer science degree that it is frightening!

I have to admit, I have always had respect from the developers as a tester because of my technical knowledge. Anyway, now testing is becoming much more developer orientated with the growth in automation testing, I am finally getting to do a job with a lot more development. Hurray!

Btw, don't let me start on what its like to tell people outside of IT that you work in the IT sector when you are a woman (I am reasonably pretty IMHO which makes it worse!). Generally, I find they can't believe for a second that you know anything about IT . This is confirmed every time you don't know how to do something straight away or do not have in-depth knowledge on anything remotely resembling a gadget.

My partner, who is a developer, is always the person people ask for help with all their IT problems. This is definitely a gender issue! Although to be fair, I am relieved I am not asked for some help that will take a big chunk of a Saturday afternoon to sort out....

Ana, I have heard hundreds of stories just like yours. Women are sidelined in programming environments. They tell their families their horror stories from work, and their little sisters and cousins realize that computer science is just not worth it because they guys will attack them and keep them in a career box.
Hi Susan,

From the perspective of becoming a programmer, there does seem to be more hurdles to overcome as a woman. It is a shame as I can't see why guys should make better programmers than women anyway. I know during my computer science degree, there were no differences between the men and the women in terms of ability to program. There were the able and the not-so-able for both genders.

On the other hand, it's not like I am being held back from having a career in IT. The work areas of QA, business analysis and project management actually pay the same or more than programming. They also have the advantages of not needing nearly as much work in your own time to keep your skills up to date too.

I agree that it is awful that women who want to be programmers are sidelined into different career boxes by a male dominated industry. Yet I don't think they are not exactly pushing women into the worst career boxes in the world either ;0)

For those women who do want to become programmers, I would say QA is now the place to be. Historically, women have been pushed into QA roles rather than programming. However, QA is now rapidly changing because with Agile projects, many tests are developed at the same time as the actual product. This means QA testers are now developing. I suspect and hope that this means more women will end up developers via this route instead.

So why are there not many women programmers? I suspect it is because overall, people (male and female) believe in men more than women when it comes to techie matters. From my experience, it doesn't matter whether this is true or not. It is the perception that is important.

Ana ,

(sorry its a long post, please read)...

I experienced the same thing in grad school for CS --> Women were equally talented as the men, and in project teams women were the ones who produced working code faster, and their portions of the team projects were less buggy and better documented. The teams with majority women members produced systems more robust and full featured and were 'deliverable'.  The teams that were primarily men presented a lot of 'vapor ware' with 'wowzer' bells & whistles on the side. But the men were offered the jobs, women weren't, and this was in the 90's. It's still the same bias today with swaggering guys who boast that men are mysteriously better, and nice guys who don't tell the swaggerers to shut up. Strange, how with each decade (1960's, 1970's, 1980's, 1990's, 2000's, 2010's) a different reason for the situation is given.

So the argument that women aren't offered jobs because they're not programming games for fun doesn't fit the problem description, because this bias existed prior to the phenomena/perception that guys program for fun and women don't.
The 'programming on the weekend' is just the latest male fairytale that some guys believe makes them different and better than women, but it actually describes nothing of importance, just that a few guys boast and this effects the entire sector of talented women who are looking for jobs as programmers, but get escorted out the door due to the latest fad in male excuses.

Being sidelined into QA is still a problem, even in Agile environments. 
The one women who has been able to transfer from QA into kernel development at Canonical had a very difficult time.  Many people had to pull strings to get her transferred.
Another woman I know of who was talented enough to be on the kernel team was turned down because she "hadn't participated in enough 'open source' projects". She's young, and had been defensive about a silly remark that was made about how her work on an open source project was always accepted because her boyfriend was involved in the acceptance process. This made her famous/infamous. At her interview she was offered a job on the QA team - yep, a talented proven successful programmer with a good track record in open source, was offered a job on the QA team. She is an interesting, quirky and intelligent person yet has a bad rep because she stepped up to protect her technical reputation. So was offered a position beneath her skillset with no future in kernel development.
I know that kernel development is difficult, but why keep moving the bar high for women, then lower it for men? Or move it over, etc. And why be concerned about the women who have had to stand up for themselves, and not be concerned about the men who are making sidewise remarks about women and their competence?  The men who make these statements are nice guys, they just don't realize that they can't make essentially racist statements in good fun anymore.  I know of both of the people involved in this story, I know the guy personally - he's lovely and I consider him one of the sweetest and nicest and most intelligent people ever. I hope he considers himself my friend after this post.  He's fabulous. But it's the woman who has paid for this situation.  It's very sad.

And why don't men realize that they use weird subjective ways for determining if a woman is good enough to be on their team? Again, I say these men are truly good guys.  Yet there are guys who are not-so-good who proudly define themselves as mysteriously inherently better at technical subjects than women, and also insist that men make life choices superior to women which therefore results in men being technically superior. This is getting too Nietsche-esque in favor of male superiority. Why isn't anyone taking the companies to task who don't have women in 16% of the jobs on their prime programming teams? Why doesn't anyone questioning the men who complain about the few women who have made it onto those teams?
I've never heard an actual provable complaint about a woman programmer.  All I've ever heard in complaints is about the woman's personality. Strangely, the personalities which are complained about are always better, nicer, and more team-oriented than the personality of the guy who does the complaining, and the complainer is usually the unreliable team member who delivers late, and boasts about his awesome talent to anyone, etc.

Nobody mentions the wierd situation when there's only one women on the team, and this woman gets along with the  women on other teams but tries to get rid of any new technical woman on her team. This is a phenomenon I've seen several times.  Bizarre.

Also, nobody mentions how women have to not let it show that they have many battles with bullshit male beliefs they have to fight everyday just to get their work done. Lot of work for very little reward.

BTW - Producing QA code to test somebody else's code is a whole different environment.  It is *not* the same as programming with the big guys. The pay is definitely different, the rewards are different, the career future is different.

Companies and contractors, big or small, need to get over their bad selves and just start hiring women into the inner sanctum of programming at the same rate as men. The same % of women CS graduates should be showing up as programmers in corporations, as the difference in % between men and women CS grads and men and women programmers is a crime. 

- Susan
Hi Susan,

It sounds like you have been working in more sexist circles than I.

I am located in the UK and have also worked in the European mainland. It is still sexist here with women being pushed more into QA, business analysis and project management. However in Europe, the salaries for QA testers are in line with developer salaries except it is easier to become a team lead or a test manager than to become a senior developer as the latter takes a number of years. Your average business analyst and project managers tend to earn more than your average developer too.

I have seen women who did break into programming but then reasonably quickly get promoted into the business analyst or project manager jobs. I have seen men be promoted in this way too but they usually return to programming whereas I have seen this less so for women. I think the reasons for this is complex.

For a woman, the biggest problem is trying to break into programming in the first place.

First of all, I think there is a general idea by some developers that only a handful of people can actually do this type of work. This only confirms to these developers that they are truly very clever and are very special. In other words, it's an ego trip. So if they see lots of other people doing it ie women, it destroys their ideas of just how clever and special they really are to both themselves and to others. Never mind that women are perfectly capable of PHds in maths, physics, computer science and **can program like men** etc....

It sounds like your female friend who was pushed into QA away from programming feel victim to this mentality.

Secondly, and this is entirely from my own perspective, it seems like if there are two trainees where one is a male and one female, they will judge the woman more harshly in terms of ability than the male. I think this is because a woman stands out a lot more and therefore everything she does is magnified. This is, of course, a nightmare when you are still training to be a decent programmer.

On the other hand, if a woman has technical knowledge and works in any other area of IT away from programming, most of the time you receive **a lot** of respect from everyone in the work environment. This is because you understand technical issues that non technical people understand and you have the advantage of being able to communicate well with technical and non-technical alike. You are truly seen as being a very bright woman.

On a different note, I have now seen two women move from being automation testers on Agile projects into actual java developers. I don't know how often this occurs but I hope this trend is reasonably widespread and will continue into the future :)

As for working with other women, I have come across some great women who I have a lot of respect for and I have come across some bloody awful women too. But then this is the same for the guys too; you always find people you get on with or not get on with wherever you go ;0)

Ana

WOMEN on AVERAGE are incapable of dealing with CS. That's a cold hard truth.

I'm a female software engineer (C/C++/Java), no scripting garbage, btw.
There's no need for BULLSHIT increase of women in CS, and stupid affirmative action.
Either you can do it or you CAN'T and you aren't competitive if you can't, and this means you can't cut it and can't hack it.

I always was among top 5% of my class in CS. Yes, there was "slight" discrimination--the way I look at it: THE MORE THEY HATE YOU FOR NOT BEING A MAN, THE HARDER YOU'LL HAVE TO TRY--THE BETTER YOU WILL GET, SURPASSING THEM, AS A RESULT.

As to that bullshit that kids are "conditioned" since early age.... well I was conditioned to hear that girls can't be good at math, hearing it many times during my childhood. This did not prevent me from being a top student in my math classses. When picking a toy as a child, I always went for a car or Lego, not stupid dolls--nobody can force a child to play dolls if they don't want to. Nature is nature. Most girls are just not inclined for CS.

As to women being "pushed" into QA: nobody can push no one into QA.
If you can't write code, you'll have to be a QA. If you can write code, actually, you will not even CONSIDER going into QA, as you know you can do much better.

By the way, I grew up in Soviet Union :] Our engineers are better than Western ones. :]

Hmmm. Consider the following question:
Is it true that if a person believes the following statement to be true --
"WOMEN on AVERAGE are incapable of dealing with CS"
then that person is an idiot?
Well, is it or isn't it true that they're an idiot?
The answer is that it can't be determined from the information given.

It doesn't take anything away from +murderous_vampire prodigious programming talents to state, based on numerous statistically valid studies, that many qualified female programmers are steered towards QA because it is erroneously believed that females can't program as well as males.