15

It's well-known that Stack Overflow has a pretty extreme gender imbalance (annual surveys show less than 7% of respondents identifier as female). Of course, this is probably partly influenced by the gender imbalance of the related industries.

Is there any information on what the demographics of this site are? On one hand I imagine the imbalance if SO might play a factor, but on the other hand this isn't a site about programming.

I'm mostly curious because after reading some recent content, I'm growing suspicious of a strong bias towards male demographics. This would seem especially problematic for a site about politics if the views of large demographics are poorly represented, given that users often post only their own views on an issue.

  • 4
    Not sure, but I think the imbalance is also transferred from SO, since many might contribute here after contributing to SO (e.g. following some link from Hot Network Questions and falling in love also with this community). – Alexei Oct 11 '18 at 5:25
  • @Alexei I think so too, but if that's the case one would hope that this site might inherit at-least some of that 7%? It seems nobody knows if that % here is even non-zero. – AquaticFire Oct 11 '18 at 5:27
  • the activity on Politics is way lower than on SO (still beta after so many years), so 7% might be so small that it would be hard to notice. – Alexei Oct 11 '18 at 5:31
  • @Alexei Perhaps. In the last year only 440 have gained more than 101 rep points, so at 7% only ~30 would be female, about 1 in 14. Definitely not a lot, but a lot more than 0. I feel like other low-traffic sites might do better. – AquaticFire Oct 11 '18 at 5:39
  • No, at present there are probably not many women on Politics.SE, or at least not many that I know of, or rather not so far as it's possible to infer by style and message content. – agc Oct 11 '18 at 9:45
  • 2
    This may be helpful - this answer cites what is, to my knowledge, the only honest to god 3rd-party study that has attempted to suss out the actual numbers on Stack Overflow and meaning behind those numbers: meta.stackoverflow.com/questions/290727/… – OpenSorceress Oct 11 '18 at 11:08
  • 5
    "This would seem especially problematic for a site about politics if the views of large demographics are poorly represented, given that users often post only their own views on an issue." Assuming that there is no significant overlap between women and men on political issues is a fallacy. Thus you'd expect the views to be represented if it is purely such a imbalance – user19831 Oct 12 '18 at 11:27
  • 2
    @Orangesandlemons - I'm afraid I don't understand what you're trying to say. Are you suggesting that there will be proportional representation of opinions relative to the general population as long as there is non-zero overlap in the probability that men and women hold a give opinion? – Obie 2.0 Oct 13 '18 at 3:23
  • @Obie 2.0 the the answers by definition arent proportional, or else they'd be the some answers. I'm saying that if there's an overlap, then one of those will post their views. I conceed that voting may be skewed, but I didn't read that as the problem – user19831 Oct 13 '18 at 19:09
  • 2
    The likelihood of the right answer being postef will also be skewed. – Obie 2.0 Oct 13 '18 at 19:12
  • I appreciate the discussion this question has generated. When thinking about an answer it would be helpful to have an example of the kind of content that you are asking about. Can you link to something? – indigochild Oct 18 '18 at 22:12
11

Is there a gender imbalance?

No one knows for sure. There is no source of data that can tell us. The common reasoning seems to be that it's likely there are significantly more men on the SO network; the magnitude of this difference is similarly unknown.

Does it matter?

In a general sense, StackExchange is supposed to be a friendly, helpful environment. As some of the questions and comments around this question show, we still have some more work to do in that arena. If our culture here is pushing women away from participating, then yes it matters very much. It is a symptom of a problem in our community and we should take it seriously.

More particular to our site, it matters because viewpoints matter when discussing politics. I'm a political scientist - my training predisposes me to thinking there are objective, evidence-based answers for questions about politics. Similarly, this site exists on the premise that we can provide factual answers. I recall reading that while we discuss politics here, our answers themselves are not political. Some have said that in this environment gender is unimportant.

I disagree. The experience of political science has been that gender matters a lot, even when discussing factual things. For one thing, gender influences the kind of questions that we ask. In my field (political theory) it's easy to see how gender has influenced past philosophies. By far the largest growth area in political theory has been critical theory, including feminism, which addresses the role of gender in our theory. There has been substantial interest in the political science community for increasing our focus on the role of gender, both in our work and in our own community.

Gender also influences our answers. Typically not in overt ways, but covertly in the kinds of sources we think our reasonable, the kind of facts we deem pertinent to an answer, and the kinds of statements that seem acceptable. In all of these personal judgments, gender is important.

What can we do?

I don't know. We currently don't do any routine active promotion of this site. However, I would encourage anyone to actively promote Politics.SE in their own corner of the universe. If you feel strongly about this issue, perhaps promoting Politics.SE at your local college's department of women studies would be helpful.

The most important thing we can do is tend our own community. There are a lot of issues beyond our control that impact users' experience and participation; let's not create more problems by pushing people away.

  • 1
    "Gender also influences our answers. Typically not in overt ways, but covertly in the kinds of sources we think our reasonable, the kind of facts we deem pertinent to an answer, and the kinds of statements that seem acceptable. In all of these personal judgments, gender is important." I would like to read more about that. As a political scientist, do you happen to know about some seminal works about it? If I get a few starting points, I can continue on my own. – Trilarion Oct 22 '18 at 14:12
10

I don't have access to the Stack Exchange market research data which could tell us the gender composition of our community. But my intuition would tell me the same. Among the top contributors over the past 12 month, most have male names or self-identified themselves as male in the past. None of these users unambiguously communicates a female identity in their name or avatar. But a lot of them do not clearly communicate any gender identity, so there could be a few women among them.

But I do not agree with the argument that it is especially problematic if "the views of large demographics are poorly represented" because Politics Stack Exchange is not a site about exchanging our political views. It is about exchanging objective information about politics and political processes. We do not need a specific female view to explain, say, the technicalities of brexit, just like we don't need a specific male view. We want a neutral and informative answer which provides all the information the reader needs in order to form their own views. Whether that answer was written by a man or a woman seems quite irrelevant to me.

What would be a problem is if some demographic would post one-sided views and not get called out for it. But this is, as far as I can tell, something which already works quite well in our community. When someone posts something which contains blatant sexism (or is discriminating against any other demographic), then it usually receives plenty of downvotes and does not stay online for long.

Nevertheless, the lack of female users might hint that our community has subconsciously built some barriers which discourage women from participating in it. It would be really sad if we would unintentionally exclude a large pool of potential contributors without even realizing it. Unfortunately I, as a cis-gendered man, do not really have a good perspective to recognize such barriers if they exist. So if anyone does see something which hinders the inclusivity of our community: It would be really helpful if you would speak up.

  • Comments are not for extended discussion; this conversation has been moved to chat. – yannis♦ Dec 24 '18 at 14:13
4

A few things that are irrelevant to the question are being brought up. Just to be clear, the overwhelming majority of people on StackExchange are male. This implies that most contributions on the site are from male people.

A lot of fragile voices feel that this question brings up issues along the lines of representation, views and opinions which are too airy-fairy to figure on a site based on hard core facts. I believe that this is unwarranted for. In fact, it's beyond facts that we truly understand a problem and try to propose a solution. We need detail, and a great way to provide detail is via context, which is inevitably a product of representation, views and opinions. For example, take a look at "How to concatenate the results of two Linux commands" where the OP gives context to his question as so

I pull the local outdoor temperature from a nearby weather station. The result today is:

5.2

I simply want to add the units, so the result should look like this:

5.2°C

We see that the question is representative of weather datasets, where the OP probably views temperature in SI units (Celsius). The answer is relevant to both representation (weather) and view (Celsius), even though the question is generic. We must acknowledge how these two elements brought us the answer. Had it not been for this tidbit of context, the problem would have been flooded with generic, irrelevant solutions that could be quite easily found with a little bit of googling.

If we had more people in diverse contexts participating on this site, we would have more diverse problems with great detail and interesting solutions. Given that the majority of users on this site are male, it is thus not very hard to imagine that most of the detail and/or context on StackExchange is overwhelmingly male. Take these examples:

enter image description here

Everything we post is highly influenced by socio-cultural, ideological and demographic factors. We do not exist in a vaccuum where anything of what we do is based purely on a single factor or decision. It is thus extremely important to have diverse perspectives in order to have more detailed, more broad scoped content. Take a look at the same question being answered on a different site:

enter image description here

As it is obvious, there is a lack of perspective, which leads to a poor quality answer. These samples are taken from StackOverflow, which tries it's very best to be context-free, objective and generic. If we can see the influence of context in StackOverflow, imagine the influence on a site which deals with subjective matters, such as politics.

  • 1
    Downvoters, care to comment? – Fabulous Oct 19 '18 at 6:14
  • Haven't downvoted, but I think this answer lacks a bit of focus and at the same time is overly ambitious "As it is obvious, there is a lack of perspective, which leads to a poor quality answer." From the few examples given, it did not become obvious to me. A possibility for sure, but obvious? – Trilarion Oct 22 '18 at 12:27

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .