Following on from a heated debate online yesterday I thought I would post about why promoting careers in computer science/programming/science to under-represented groups is necessary and worthwhile. As it happens, this also comes as Chi Onwurah, MP for Newcastly-Upon-Tyne, has called for greater gender equality in the games industry (as detailed here).
Expect a graph and lots of footnotes…
I’ll focus on women in computer science/programming roles because there is a reasonable amount of data – but will note that in different societies there may be different groups under-represented for different reasons. In this regard I don’t think that the games industry is unique – my belief is that it is a microcosm of the greater IT industry, only more so in some regards.
So while female students are in a minority in UK IT and computing courses, the fact that they are in a minority in game programming courses is not surprising. That (in my experience) they are in an even smaller minority is worth noting, and not something we should be proud of.
Why is this a problem? Well, if an industry wants the best talent, it needs to recruit its talent pool from as much of the population as possible. If talented individuals are being turned away at an early stage, then the industry loses. If around 50% of the population is being discouraged from considering it as a career path, then that is pretty awful… how many brilliant people are we missing?
A common response is to ask ‘well perhaps women just aren’t interested in this kind of career’. To which we can ask in turn, why not? What is it that puts women off studying science or computing, or from maintaining careers in these areas? And I do mean it when I refer to discouragement, and putting women off of computing – because there are less women studying and working in this area as a percentage of the total than there used to be. Be it active discrimination (I don’t think it is – EDIT: but see comment), or some blend of social factors, role-models and expectations… something is happening. Here is the graph. Be sure to read the footnotes which are full of caveats.
If you read all the footnotes then you’ll see that I’m not really totally confident about all these numbers (which are based on US data), but I believe the trend is genuine. And is backed up elsewhere, e.g.:
This troubling statistic can be held up against a troubling trend: in 1991, 36% of IT jobs were filled by women; in 2008 that number slipped to 25% [2, p.14]. The decline in female participation in the IT workforce matches the decline in participation in university programs. In 1985, females represented 37% of computer science undergraduate degree recipients. In 2008, the number was 18%. Overall, from 2000 to 2008 there has been a 79% decline in female undergraduate students interested in majoring in computer science .
Overall, this is a significant trend. If women ‘just aren’t interested’ in technical areas of computing, then it seems they are even less so now than they used to be. I think that cultural expectations and role-models have a lot to do with this, but I am not an expert and I would encourage folk to read more of the extensive published literature. But anecdotally…
I was told once by a former lecturer of mine that Universities noticed a big dip in percentage of female students around the time that computing started to be taught in high schools. Perhaps this was making children choose at a younger age whether or not to identify themselves as being ‘into’ technology, perhaps something happening inside the schools was at fault. Or perhaps the increased visibility of computers in society and in the home as a predominately ‘boys toy’ had an effect on expectations. The trend has continued in a downwards direction ever since.
The discussion and looking at the figures has convinced me that this is an issue that I need to become more actively involved in – in trying to get something running at my university to encourage more women to consider careers in computing and in game technology. I don’t think women are uninterested in games, but I want to ensure that the university environment is welcoming and that we do something to raise awareness of the career opportunities. As was known in 1946, but seems to have been forgotten since, programming is women’s work.
ACM Inroads is the ACM’s education oriented magazine. Articles about outreach projects and current state of gender balance and minorities in computing appear reasonably regularly. Articles such as this one recognise that the issues arise before university. Judy Robertson wrote this recent piece on one way to put women off careers in tech. Which makes me aware of one seemingly innocent classroom behaviour I have to watch out for. The notion that cultural factors have a large impact seem reinforced by studies like this one which see different gender splits across different regions of the globe.
1. There are a lot of footnotes. If the table was extended back to start in 1842 and the inclusion of Ada Augusta, then the first entry would list women as 100% of the workforce
2. The previous footnote obviously disregards Babbage, who designed the hardware. In fact that is a common theme in this data where it is primarily the software development (programming) that is considered. The NSF data used for workforce ratio for the later years actually has a separate category for electrical engineering and computer architecture – if that was included then the percentage of women would appear even lower.
a. In 1946 I am assuming that the Harvard Mark I was still operating, as was ENIAC. There were three noted programmers on the Mark I, two male and one female (the famed Grace Hopper, who created the world’s first compiler). ENIAC had a team of six female programmers. Prior to this date, women were commonly employed as part of the computing workforce – they were employed as computers, i.e. it was their job description. When computing machines were developed a natural development was to employ women to program them. The computing culture of the time considered this a kind of clerical work, suited to women. Does this help make the point about culture and expectations?
Choosing 1946 conveniently allows me to overlook the (to me) unknown programmers of the Zuse computers and consideration of whether the female operators of Colossus were the programmers, or even from having to consider to what extent Colossus was a ‘programmable’ computer.
b. Workforce figure here is using an unverified “mid 1980s” figure from Wikipedia article on Women in Computing.
c. Graduations figure from here.
d. Workforce figure again from Wikipedia, but refers to data from NSF government data available here, but which I have not double checked.
e. Graduation figure as above.
f. See d.
g. See e. Which refers you back to c.
h. This figure was the most up to date one I could find in the NSF data. I took this value from Table 12. There is a separate entry for careers that relate to hardware/architecture that I have not used, much as I have focussed on the people who program the systems rather than build them in the other data sets.
i. Go straight to c. Or the Trends paper it links to here.