Imagine a scale that goes from autism to whatever the opposite of autism is, with neurotypical as the middle. The hypothesis is that men are more likely to be on the autism side of dead center, even for people who are still in the neurotypical region. A higher percentage of men being diagnosed with autism is evidence of the hypothesis being true.
> So I guess I'm still firmly in the "Its a pipeline problem" camp.
They're kind of orthogonal. It clearly is the pipeline problem -- you can't hire people who don't get the relevant qualifications. But it's possible that biological differences contribute to the pipeline problem.
In reality anything is affected by everything. So it makes sense to work against the pipeline problem regardless, because some of it is caused by non-biological factors that could be addressed. But some of it may also be caused by biology, which means that even if some progress is attainable, expecting the end result to be a 50/50 split may be unreasonable.
The scale exists because it's a scale of severity in diagnostic criteria. The scale exists to distinguish between "non-verbal and incapable of ever caring for oneself" to "experiences impairment but can generally hold a job and live independently". It's not a scale of how good you are at finding boxes in colored square pictures.
Frankly it's kind of insulting to imply autism is just a grouping of logical behaviors and anyone who is good at those things has some "level" of autism. It's a disability. I got picked on, I got mocked, I struggled to understand basic things and it took me a lot longer than my peers to get this communication thing down.
Besides, if programmers had enough incidence of autism to justify this level of imbalance, would pair programming really be a thing? (this was a joke).
Some would say Williams Syndrome. http://www.npr.org/templates/story/story.php?storyId=1262248...
Sure some of the symptoms seem opposite (but others are actually kind of similar), but the genetic component of that disease is pretty specific and well understood. Not like autism which has been much harder to pin down on any single chromosome.
Maybe AnthonyMouse could've been more precise and said "personal behavior opposite of behaviors characteristic of people on the autism spectrum", but that's awfully wordy and pretty clearly what he was getting at.
If I had to guess, it'd be the extravert. The social butterfly, life of the party, comedian. The one who wins all the popularity contests and the weekly sales competitions. The person who finds math (and any math-related technical field) to be arcane and impenetrable but has all the business class profs eating out of his/her hand.
Nearly every disease has an opposite which is also bad. If something was an unmitigated evil then there would be strong selective pressure for it to not happen and nothing pressing in the other direction to cause it to actually occur in the population.
> Would you say there is a scale ranging from having the flu, to being normal, to... having the opposite of flu, I guess?
Contracting the flu is a failure of the immune system to fight off the virus, so the opposite of contracting the flu would be an autoimmune disease.
[citation needed]. Also, you can't use evolutionary arguments like that, to justify everything. What about the plague? What about an anvil to the head? What's the opposite of getting a heavy object dropped on your head and if there is none then why didn't we evolve a titanium skill? You see where I'm going? It just didn't because we don't immediately evolve defenses against every single danger and disease in existence. Evolutionary arguments aren't worth very much and certainly can't be applied in such a maximalist way.
"[citation needed]" is the claim that an argument is flawed because it lacks an appeal to authority.
An immune system that is too lax leads to infections, too stringent leads to autoimmune diseases. Pick almost any substance in the body -- glucose, sodium, serotinin, etc. Too little is one kind of bad, too much is a different kind of bad.
And as to the original point:
http://journals.plos.org/plosgenetics/article/file?type=prin...
> What about the plague?
That seems to be in the same category as the flu.
> What about an anvil to the head? What's the opposite of getting a heavy object dropped on your head and if there is none then why didn't we evolve a titanium skill?
Are you asserting that there are no trade-offs there? Flexibility, self-healing, availability of carbohydrates vs. titanium compounds in available food sources, etc.
> It just didn't because we don't immediately evolve defenses against every single danger and disease in existence. Evolutionary arguments aren't worth very much and certainly can't be applied in such a maximalist way.
Evolution is not a divine engineer creating things that are maximally perfect, it's a brute force pragmatist that destroys anything that can't out-compete its neighbors.
It isn't that purposeless flaws literally cannot exist at all, it's that they are selected against. Something that beat improbable odds due solely to dumb luck.
The assertion that something is the exception to the rule is the one that needs independent support.
Autism is a spectrum and there is no clear cut between a perfectly neurotypical person and a profoundly disabled. You can measure number of autistic traits in general population and see that there are differences between groups, look e.g. here:
- The Autism Spectrum Quotient (AQ): evidence from Asperger Syndrome / high-functioning autism, males and females, scientists and mathematicians (https://link.springer.com/article/10.1023%2FA%3A100565341147...)
There is no a single paper proving that HFA is a binary condition, and many hints that it isn't.
(BTW: I dived into this topic some time ago, with already linked here http://crastina.se/autistic-traits-science-and-the-nerd-ster...)
So the opposite of autism would be something like hallucinations and delusions.
(Idea not mine, it was somewhere in http://www.afieldguidetoearthlings.com/ book.)
Nonetheless I'd like to point out that there are broad, statistical, gender differences between those who are "thing-oriented" and "people-oriented", that would need to be addressed before we could hope to see 50-50 splits of M/F ratios in every subject today: http://slatestarcodex.com/2017/08/01/gender-imbalances-are-m...
Donald Trump
Is a biased pipeline even a problem? What are we trying to solve?
It seems there are separate issues: equal numbers, we need to tackle the pipeline and convince women STEM is great; what happens happens, we need to not be biased against a gender and admit proportional to applications.
I can see both sides; I think in different moods I ultimately agree with either.
At the moment, I think if we spend long enough with admission (universities, jobs whatever) reflecting competence in the pipeline, then the pipeline will gradually come to reflect the proportion that's interested - 50:50 or not (either way around).
The idea that women want different things than men is pernicious and it's not something we should accept without proof. For a long time we rationalized keeping women out of the work force by saying that they preferred focusing on the home. Yet, I know more stay at home moms that are bitter about having given up their career than the other way around. It's a lie we told ourselves to justify a social arrangement that disempowered women. We should know by now not to fall prey to that sort of thinking.
Besides, men and women have had distinct gender roles (and largely the same roles) in every society since before history; while (or rather, "because") I support women's rights to do whatever they dammed well please, it seems unlikely to me that we discovered in the last fifty years that that was entirely an ancient conspiracy to make women miserable, and thus we should push women into the careers we know they want (never mind that the evidence points in all different directions) even at great expense to the industry.
Anyway, the question isn't "should we force women to be programmers or housewives?". It's not a dichotomy and we shouldn't be pressuring anyone into anything.
The goal should be that everyone can freely decide which career path they want to join regardless of how the ratios in the final pipeline look.
I'm certain you know that's what he meant as well. You can disagree with him that any such obstacles exist (you'll be wrong, but at least you'll be coherent). But why instead waste our time and your own with a pointless no-op of a rebuttal?
What exactly are the barriers that exist for programming that were not cleared at the same time as those for math, economics, and medicine?
I suppose the word I'm trying to unpick is 'facilitate': is that by fair treatment of those already making it through the pipeline, so that the bad reputation is reduced, and more enter it, and so on; or is it by pushing more into the pipeline immediately?
> So if there are lots of girls who would feel the same way if society didn't bias them against the profession from an early age
I'm just not sure to what extent society has a duty to 'make people realise' that they'd enjoy a particular career path.
No it doesn't. Bias isn't always bad or oppressive. It needs to be proven that the bias is "bad".
If rich kids are underrepresented in tech because their families have connections in finance and they tend to major in finance more often, is this a problem that needs to be solved?
Frat bros are underrepresented in engineering in tech atm.
HR, recruiting and hundreds of other job families and careers have "biased pipelines". Should resources be spent to solve all of them?
Why only tech?
"Great" is necessarily a function with (at least) 2 inputs, the thing being considered, and the person considering it (or their goals).
Who is more likely to have a better job security, a STEM graduate or a med school graduate? What are the failure modes of each? Is a QA career better, or worse than that of a nurse (just spit-balling here, it might very well be the other way around)?
--
I'd argue that men are more likely to be the risk takers, for evolutionary reasons. It makes perfect sense that they go for a career with lower chance at a bigger win. Less to lose -> take more risks.
Now, if you are a woman, you are extremely likely to have a quality of life higher than the average man. Why would you take risks when you don't need to?
That is the elephant in the corner of the room, isn't it?
Maybe it's not the women are irrationally opting out of tech. Maybe it's the men are irrationally opting in?
Lets stop there. What are we trying to solve? I'd say we are trying to solve this question: What makes a good life? STEM gets a lot of press because, obviously, it is a well paid field.
However, the answer to what makes a good life is not just money. I'd wager it is (almost) universally agreed that a good life consists of: (1) Enough money (see below), (2) a career you enjoy, (3) free time to pursue things you enjoy outside work, (4) good health ((4a) mental and (4b) physical) and (5) connections with other humans. Maybe people can quibble with this to some degree, but I think that is about as uncontroversial as anyone can be on the subject of a good life.
Accepting my definition for the sake of argument, how does a gender imbalance in STEM affect this? Does a STEM career affect all 5 of those elements? Just some? Just others? Certainly, a career in STEM provides "enough money". But is it a career EVERYONE will enjoy? Do you get a lot of free time? It seems rather unhealthy as well - there aren't a lot of healthy people in offices in general. As for connections, YMMV on that one.
I think we focus too much on money because it is so easily measured. But more money beyond about $75K in the USA doesn't appear to make people more happy: http://content.time.com/time/magazine/article/0,9171,2019628... If you make above $75K, then surely that is the time to optimise the hours of your day not towards making more money, but to add more of the other 4 elements of a good life.
STEM is not a panacea, and money is not the be all and end all of life. Trading better financial security for health, a career you enjoy and more free time and personal connections may be a sensible decision. I'm not convinced more people would enjoy STEM careers, any more than I think everyone would enjoy being a special units detective, and dealing with a lot of awfulness.
> We focus on money because it's easy to measure
My inner cynic thinks it's because it's the disparity that's politically expedient. Lots of other disparities are easy to measure (flexibility, workplace injuries and fatalities, etc), but we don't post much attention to those... :/
There is no such thing that I am aware of in medical literature. Did you make that up, or can you point me to some reference(s) on this supposed scale?
No it isn't
Wouldn't that anecdote also apply to men? E.g. We take 2 men currently in bio/chem and talk to them about switching to CS which had never occurred to them?
Therefore, does the "uncounted" men cancel out the women "missing in the compsci pipeline"?
I started because I needed a statistical model that SPSS didn't provide, and quickly realised that coding is one of the most fun things in the world.
I think the author also intends to try and suggest that the 50/50 split doesn't need to be blamed on men.
However the title was: "Is gender inequality in technology a good thing?"
So is the author saying it is a good thing? Does he mean that gender inequality (as it presents today) is good? Or just that some gender equality (60/40 for example) would be as good as 50/50? I hope he means the latter, but the wording implies the former.
I'm just worried that he didn't acknowledge the potential for factors besides autism to be relevant or provide a more fair target equality ratio to strive for. That makes me think he thinks the current numbers are fine, which I disagree with. The level of imbalance we see today indicates to me we are missing out on talent, and my experience makes me think it's a pipeline problem due to underexposure of tech careers to logical, but non-tech hobbyist young females.
Playing video games and having the desire to make my own was a huge factor for me getting into tech personally. I had a subscription to "Nintendo Power" magazine, and I still remember reading an article in there about a programming school called "Digipen", and just being enthralled by the concept of going there and building my own games.
In retrospect, I'm glad I didn't have the patience to stick with game development. My career is probably better off for it. My real gateway into a tech career was when I got put into a Cisco networking class by accident. I learned there that Linux live-boot usbs were excellent for getting around permissions restrictions on school computers. I also learned that the school firewall could be bypassed by using IPs instead of hostnames. I never looked back from there.
This theory is advanced convincingly in a Planet Money episode -- the graph here is incredibly striking, and for me strongly challenges any talk about the gender gap being primarily genetic/trait-base:
http://www.npr.org/sections/money/2014/10/21/357629765/when-...
The shift from tracking progress with other STEM fields up to >35% female workforce participation in 1985, then suddenly breaking with their progress and going in the other direction is astounding. That's a clear exogenous change, nothing 'trait-based' could occur on that timescale.
The only other vaguely plausible explanation I've heard (but am unable to find anymore sadly) was a suggestion that in the late 70s, IBM changed their recruitment guidelines for programmers to rigidly select for what we'd now call autistic-spectrum / raw analytical personality types, which biased hiring towards males. According to this theory, other companies just followed IBM's lead, and this triggered the paradigm shift.
However that doesn't really explain why there was another massive dip from a fairly stable 10yr rate of ~26-27% down to ~18% in 2003; I can't recall anything particular in the news around then but wonder if there was another scandal or news cycle highlighting negative experiences of women in tech.
>, and I would be shocked if there isn't a strong correlation between interest in video games and computers in general.
Video games are marketed more to males but that didn't stop females from participating in large numbers.[1]
Why would "male marketing" dissuade females from computer programming but at the same time, it doesn't affect computer gaming?
https://www.google.com/search?q=more+women+play+video+games+...
[0] http://quanticfoundry.com/2017/01/19/female-gamers-by-genre/
So "early on", we're talking about 1960s/1970s advertisements for mainframes of IBM, DEC, VAX, Burroughs, etc and the 1980s ads for personal computers like IBM PC (original), Apple IIe, Commodore 64, Radio Shack TRS-80, etc.
If we want to blame early period "marketing" for women's lake of interest in computer _programming_, we have to also explain the counterexample of early marketing for PacMan, DonkeyKong, Missile Command not having the same decades long suppression of female computer _gaming_. The video gaming magazines and its ads in the 1980s/1990s were heavily male-dominated. (Arguably, the videogames ads were more male-themed than mainframes' ads. The videogames' ads often had abundant cleavage and skin. The mainframe ads did not.)
I don't buy that. If you include mobile games, women outnumber men but female mobile game developers don't.
I don't think there's a causation between video games and programming. Correlation, sure. But not causation (Cod is insanely popular among non-programmers)
A company that had internalized misogyny to such a degree that it couldn't function. All companies are sexist.
I think there are also probably two camps of people, even among people who go into software: those who never needed to be encouraged or asked to do it, and those who eventually started because of some outside pressure or incentive.
A company I worked for actively seeks out and trains women for web development work for free in an effort to prop up their female numbers. They bring more than the total number of employees at the company through that program in a given year. Even with this intensely focused pipeline, they still can't keep up pace with the qualified men who arrive through normal hiring routes. It's even a bit unfair, I was hired at that company rapidly with no official credentials (granted earlier on), just an interview/demonstration; whereas apparently if I were a woman, I could still be hired even if I wouldn't qualify when I walked in, and I'd at least walk out with professional training. The men there trained and represented themselves, or paid for an education.
I feel like at some point you have to stop wishing something is within your control. I don't think you can construct a fair world where women freely choose to be software developers at the same rate as men.
A workplace has no obligation to be equally attractive to every working age person, I don't think they have an obligation to be attractive to exact proportions of men and women in gen pop.
The visual test in the blog post is actually a field dependence test: https://link.springer.com/referenceworkentry/10.1007%2F978-1...
What I'm getting at is that there are broader, non-clinical variations in cognitive style that complicate the notion of neurotypicality. Anyone studying science is unlikely to be at the extreme field dependent end of the scale. So part of the imbalance may rest on real but non-clinically significant cognitive differences.
I'm referring to the DSM-V defined "autism spectrum disorder". In my comment, NT refers to "person not meeting criteria for DSM-V autism spectrum and not diagnosed as such". The field dependence test is interesting, but does not replace a actual diagnosis by an expert as far as I can tell.
While I can accept "most autistic people do well on the test " (it was pretty instant for me), I'm not convinced that a strong performance by someone who does not meet the criteria for autism spectrum disorder is proof that the person in question is actually somewhat autistic. Maybe there are other factors that make someone good at recognizing it?
And even if it was a good indicator, there is no guarantee that the 4:1 sex imbalance in diagnosed autism spectrum disorder translates to an equivalent imbalance in high-performing non-autistic-diagnosed (but hypothetically still autism spectrum by field dependence criteria) people. Would love to see some studies though.
FWIW, as a woman on the autism spectrum myself, I actually kind of agree that some level of imbalance is due to this. But people on the autism spectrum make up such a small population of humans and software engineers (we are over-represented compared to other industries, but in my experience we are still very much a minority overall) that it cannot be nearly enough to account for the sex-imbalance numbers we see today.
Anecdotally, I've convinced two NT (neurotypical) women to switch from bio/chem degrees to CS (they are both software engineers now) and the reason they hadn't considered before was b/c nobody ever talked to them about it. Neither grew up playing video games (though one said she asked for a console and the parents put it in her brother's room for them to "share") and there weren't any mandatory programming classes, so they just were never exposed to it as an option, even though they are both highly logical people (thus the chem or bio degree paths before switching).
So I guess I'm still firmly in the "Its a pipeline problem" camp. Until we account for cases like the ones above, tech is going to lose out on some talented women, and some of those women will miss out on tech. I think mandatory high school programming and networking classes are what will really help.