Hacker News Reader

Comment by gshulegaard | Hacker News Reader

I don't think that the parent comment is assuming one way or another, just offering a plausible explanation given the very limited perspective of the metrics being quoted.

For example, how does this compare to other trends for female jobs in other sectors? How do the figures compare when you separate the population between those with children and those without?

When someone wants to make a claim like: "These numbers reflect two problems: 1) the percentage of women who work in tech is low, and 2) women are leaving tech jobs in droves."

I would expect more analysis than simply: "In 2014, 26% of computing jobs were held by women – and that’s down from 36% in 1991"

I think there are a lot of bogus statistics/conclusions being drawn when it comes to many social issues, female equality being one of them (https://www.youtube.com/watch?v=1oqyrflOQFc).

I am a libertarian so ultimately I believe that everyone should be free to make their own decisions even if there are gender preferences at play. Not saying there aren't issues, but simply saying, "X is anti-women, because only Y% are female," is academically dishonest.

View on HN

> I am a libertarian so ultimately I believe that everyone should be free to make their own decisions even if there are gender preferences at play. Not saying there aren't issues, but simply saying, "X is anti-women, because only Y% are female," is academically dishonest.

It's not academically dishonest, it's the application of a reasonable epistomological presumption: we presume that observations are not caused by biological differences without concrete evidence of those differences. The corollary to that is that disparities in outcome are prima facie evidence of disparities in opportunity.

It's not that people are unwilling to acknowledge biological differences between genders. Nobody argues that women have as much upper body strength as men. Rather, people are aware of the longstanding practice of chaulking things up to biological differences, and adopt a presumption against such explanations. You can see that clearly in this thread, where people are arguing that women might be biologically predisposed to being primary caregivers without any concrete evidence backing that up.

> "we presume that observations are not caused by biological differences without concrete evidence of those differences. The corollary to that is that disparities in outcome are prima facie evidence of disparities in opportunity."

Ummm...the first part is fine, the corollary reasoning is what is academically dishonest. Prima facie is not academically acceptable when determining causal relationships in statistics. The illustration of this is why the power of Hypothesis testing is centered around rejecting the Null hypothesis. Failure to reject the null hypothesis doesn't confirm/prove the null hypothesis as true.

The corollary follows from the first part because of the nature/nurture dichotomy. It's either biological, or sociological. If you presume it's not the former, unless we have evidence thereof, we must presume it's the latter. What else could it be?

Also, what is acceptable for proving a causal relationship in a statistical sense is not coextensive with what's academically honest. In the real world people have to make decisions and set policies based on imperfect information. Requiring premises underlying those policies to be proven with scientific certainty before condoning political action just establishes a default policy of preserving the status quo.

>>The corollary follows from the first part because of the nature/nurture dichotomy. It's either biological, or sociological.

Nature/nurture argument is not a dichotomy. It isn't either/or. It can be both to varying degrees, i.e. biological dispositions can be nurtured/reinforced by environment and upbringing.

You're right, dichotomy implies sets are mutually exclusive. "Jointly exhaustive" is the phrase I was looking for.

> If you presume it's not the former, unless we have evidence thereof, we must presume it's the latter.

Your presumption seems to favour the sociological explanation. Why do you not say, "If you presume it's not the latter, unless we have evidence thereof, we must presume it's the former"?

It's not really reasonable to take as a strong prior (or null hypothesis if you are that sort) that two distributions are equal. Why would they be?

Further, lets think of all the distributions where we can easily measure - upper body strength and height are two examples. The male/female distributions are, in fact, not equal. Same for the distributions of mathematical ability.

http://science.sciencemag.org/content/321/5888/494.summary https://www.jstor.org/stable/2889145?seq=1#page_scan_tab_con... https://www.ncbi.nlm.nih.gov/pubmed/21038941 http://www.ams.org/notices/201201/rtx120100010p.pdf

It seems like you want to simply assume your pet hypothesis to be true - with no real evidence for it - while demanding a high level of rigor from those you disagree with.

See Scott Alexander's excellent takedown of this fallacious rhetorical strategy here: http://slatestarcodex.com/2014/08/14/beware-isolated-demands...

We know that there are sex specific physical characteristics that differ between men and women. Nobody is arguing that's not true. Extending these to mental differences however is a stretch and not easily observed.

The people with the pet hypothesis here are those who assume without causal evidence that there is a difference in cognition between men and women. Neuroscience is converging on the fact that sex specific differences in the brain are minimal. [0] Differences in the brain have been observed between taxi drivers and the general population. We clearly wouldn't assume that they were born to be taxi drivers, we would place the causation on neuroplasticity. Even then, neuroscience is a very immature field. Neurobollocks is an entertaining read.

[0] http://blogs.scientificamerican.com/talking-back/is-the-brai...

> We know that there are sex specific physical characteristics that differ between men and women. Nobody is arguing that's not true. Extending these to mental differences however is a stretch and not easily observed.

Yes, it is easily observed and has been extensively studied. Wikipedia has several articles about "Sex differences in ..." on different topics. For example, see "Sex differences in intelligence" [1] or "Sex differences in human psychology". Lots of studies are linked. One that I recall is "Sex differences in human neonatal social perception" [2]:

> Sexual dimorphism in sociability has been documented in humans. [...] 102 human neonates, who by definition have not yet been influenced by social and cultural factors, were tested to see if there was a difference in looking time at a face (social object) and a mobile (physical-mechanical object). Results showed that the male infants showed a stronger interest in the physical-mechanical mobile while the female infants showed a stronger interest in the face. The results of this research clearly demonstrate that sex differences are in part biological in origin.

Another example is "Sex differences in parking are affected by biological and social factors" [3], which found that men are better at parking cars:

> The stereotype of women’s limited parking skills is deeply anchored in modern culture. Although laboratory tests prove men’s average superiority in visuospatial tasks and parking requires complex, spatial skills, underlying mechanisms remain unexplored. Here, we investigated performance of beginners (nine women, eight men) and more experienced drivers (21 women, 27 men) at different parking manoeuvres. Furthermore, subjects conducted the mental rotation test and self-assessed their parking skills. We show that men park more accurately and especially faster than women.

Another known result is that male IQ is higher on average has higher variance than female IQ. The intelligence study "Sex differences on the progressive matrices: a meta-analysis" [4] concluded:

> Results showed that there is no difference among children aged 6–14 years, but that males obtain higher [mean IQ] from the age of 15 through to old age. Among adults, the male advantage is 0.33d equivalent to 5 IQ points. These results disconfirm the frequent assertion than there are no sex differences on the progressive matrices and support a developmental theory that a male advantage appears from the age of 15 years.

"Brother–sister differences in the g factor in intelligence" [5] claims that male IQ is not higher on average, but has "substantially greater variance". If male IQ had greater variance, this means we would expect that there are more men than women with very high IQs (as well as more men with very low IQs). The section about current research in [1] has more.

These are just a small sample of peer-reviewed research articles examining mental differences between men and women. There may not be consensus on all topics, but there is on some, such as the higher variance of male IQs, and it's being actively studied.

[1] https://en.wikipedia.org/wiki/Sex_differences_in_intelligenc...

[2] http://www.sciencedirect.com/science/article/pii/S0163638300...

[3] http://www.ruhr-uni-bochum.de/biopsy/papers/Wolf_Ocklenburg2...

[4] http://www.uam.es/personal_pdi/psicologia/pei/download/Lynn2...

[5] http://www.sciencedirect.com/science/article/pii/S0160289606...

The mobile/face study has not been reproduced. Researchers who tried again, making the study gender-blind (i.e. dressing the babies in gender neutral outfits and not looking at the names) found that there was no significant differences. [4]

[3] is thoroughly conditioned into women [5]. Not only do people say that males are more visually intelligent, boys take part in more activities that develop spatial reasoning and coordination, namely sports and video games. This difference starts at a young age, with boys receiving more functional toys that develop logic and hand eye coordination, while girls receive more dolls, dollhouses, stoves, strollers, etc.

[4] http://link.springer.com/article/10.1023%2FB%3ASERS.00000110...

[5] http://www.ncbi.nlm.nih.gov/pubmed/21876159

Spatial reasoning does seem to have strong male/female correlation. Give a hundred men and a hundred woman a map and an address to reach, and guess which cohort gets there more easily? Or give the same groups a Lego set and the instruction booklet (especially the tediously piece-by-piece modern ones), and see which population completes the set more quickly and more accurately.

Do you have a source, or is this a thought experiment?

> Or give the same groups a Lego set and the instruction booklet (especially the tediously piece-by-piece modern ones), and see which population completes the set more quickly and more accurately.

fwiw, Lego are overwhelmingly marketed to, and given to, boys. I'm not sure it's safe to draw a conclusion from this.

The abstract of link [4] says the following:

> Gender differences were found [] due to an increase in girls' gaze behavior. Girls also made more eye contact in female–female dyads and in the second interaction over the first. Boys' behavior remained unchanged over time. The data provide evidence for gender differences in mutual gaze in a younger sample and wider context than previously demonstrated.

This seems to reinforce the claim that there are gender differences, doesn't it? (I don't have access to the full paper and can't see the result you're referring to)

> This difference starts at a young age, with boys receiving more functional toys that develop logic and hand eye coordination, while girls receive more dolls, dollhouses, stoves, strollers, etc.

Did you know that a monkey research study examined toy and play preferences of male and female monkeys? Like humans, male adolescent monkeys prefer to play with wheeled vehicles, while the females prefer dolls. "Sex differences in rhesus monkey toy preferences parallel those of children" [7]:

> Sex differences in juvenile activities, such as rough and tumble play, peer preferences, and infant interest, share similarities in humans and monkeys. Thus if activity preferences shape toy preferences, male and female monkeys may show toy preferences similar to those seen in boys and girls. We compared the interactions of 34 rhesus monkeys, living within a 135 monkey troop, with human wheeled toys and plush toys. Male monkeys, like boys, showed consistent and strong preferences for wheeled toys, while female monkeys, like girls, showed greater variability in preferences. Thus, the magnitude of preference for wheeled over plush toys differed significantly between males and females.

[4] http://link.springer.com/article/10.1023%2FB%3ASERS.00000110...

[7] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2583786/

The abstract is phrased in the way that it's most sensationalist to the neuroscience community, unfortunately. The authors said that they found effects in a follow up study, but in the initial study that replicated [2] they found no significant results. Leeb and Rejskind's analysis suggests that the expectations of the experimenters influenced the study in the previous example.

The thing is, talk to girls -- they don't actually like to play with dolls. Many of them envied the lego sets and toy cars of their male peers.

I'm losing steam. I might flesh this out tomorrow. I've been arguing with people far less reasonable than you all day.

I agree it's not easily observed. What I'm disagreeing about is the idea that if two numbers are not easily observed, they must be assumed to be equal.

Similarly, it's unreasonable to assume that the number of dollars in my wallet is equal to the number of dollars in your wallet unless someone can come up with p<0.001 proof otherwise.

Also, on this topic, I don't trust any pop-science writing (such as Scientific American) or journalism about science papers. I don't even trust the abstracts of science papers. Take a look at my citations about mathematical ability - unless you look at the data tables, you won't actually realize that those citations provide strong evidence of significant differences in math ability between men and women.

For example, consider the article with the title "Gender Similarities Characterize Math Performance", while the first data table shows a variance ratio of 1.11-1.21 and the second table shows the result of this in the right tail.

Unfortunately scientists need to tread very carefully and obscure the "wrong" results - loudly proclaiming these results is a career limiting move.

> we presume that observations are not caused by biological differences without concrete evidence of those differences. The corollary to that is that disparities in outcome are prima facie evidence of disparities in opportunity.

Wouldn't it be better to cite studies comparing women who do and do not care for children as proof that there is some other cause, instead of asking everyone to accept that assumption?

Casual studies of that scale are not common. You'd literally have to document a person's entire life.

we presume that observations are not caused by biological differences without concrete evidence of those differences. The corollary to that is that disparities in outcome are prima facie evidence of disparities in opportunity.

This is a textbook circular argument.

Is it? I think it just means you need evidence of biological differences (or lack thereof) from things other than disparities in outcome - from outside the circle.

If I say "Women aren't as fast at sprinting as men, and it's biological, as evidenced by the fact women aren't as fast as sprinting as men" I agree that's a circular argument.

But if I say "Women aren't as fast at sprinting as men, and it's biological, as evidenced by these studies of testosterone levels and their effects" that's a non-circular argument.

I agree with the last part. But here OP said "we presume X is not a factor", and then goes on to conclude X is not a factor.

OP requires evidence for biological differences, but requires no such thing for the discrimination theory.

Throwing around "epistomological presumption", "corollary"' and "prima facie" does not make this a scientific argument.

In reality there is plenty of evidence for biological gender differences (and for discrimination). Pinker's "Blank Slate" is a good start if you have time to read 600 pages.

And to say that people are making their decisions in a vacuum is intellectually dishonest. Unless of course you just assume that women have different brains, that make them too intellectually feeble to have interests in computing, engineering, machinery.

I don't have an interest in forwarding the notion that one gender is superior to the other in anyway shape or form. But neither are both genders identical.

For example, the video I linked above, there is a brief observation made about how more female doctors drifted towards pediatrics than men. I don't think that says anything about the potential of either gender, but I would possibly describe it as a gender preference.

And ultimately, I think we should aim for freedom of choice. And if we do that, I don't think we can expect to split our population up by any characteristic, examine flat participation percentages and think that anything other than 50-50 is caused by prejudice.

>"But neither are both genders are identical."

I'm going to take a step back and substitute gender for sex. Saying that the two sexes are not identical is a very interesting, and strong stance. Especially seeing as it is impossible to prove with socialization and neuroplasticity.

I don't doubt that there are preferences based on gender. Gender is the word that we use to describe a set of sex-stereotypes that we project onto members of our society.

Aiming for freedom of choice makes sense sans socialization. I overuse this example, but imagine Douglas Adams' cow that was raised to be eaten -- if we raise girls with the notion that they should love children, be self sacrificing, value helping others over money, not be selfish, and raise boys with a completely separate set of traits, is it reasonable to not apply a critical lens to these choices when they cause a detrimental effect on the quality of life of one group, and the set of available experiences to both groups?

Is raising a cow that chooses to be eaten any better than raising one that doesn't? Is it even worse?

I'm not saying it's true that biological sex creates certain predispositions, but it's also not unreasonable. The people who dogmatically deny it outright are as bad as the people who strongly affirm such differences despite lack of evidence.

Would you say the same about any other causal relationship? Are there any implications of it that don't perpetuate the idea that women are intellectually inferior? That don't shift the burdens of emotional labor and child rearing onto children? The idea is so obviously sexist...

How do you mean? A common stereotype is that men are aggressive while woman are kinder and gentler (the former true to some extent because of testosterone). Surely that's a good trait, so not all stereotypes disfavour women? Point being, recognizing differences between the sexes (which it exists in areas such as the aforementioned aggressiveness, or in kids where you can see boys are more restless and prone to suffer ADHD, but almost certainly not in intellect like you suggest) is not "so obviously sexist". It's the reality.

Actually testosterone studies have shown that it increases aggression in men who have anger problems. Estrogen has been shown to have similar increases in aggression in women. Too lazy to find on mobile but I'll edit later.

Boys are held to a lower standard of behavior than girls. "Boys will be boys" is a societal mantra.

It's worth noting that these "inherent" traits vary across societies. For instance Italian culture believes that girls are more violent and boys are kinder and more emotional.

Women being kinder is a sex trope not a neurological reality.

Also both of your examples are perfect fits for what I was talking about. Women being kinder means that kindness is expected of them. Also that they are available to be emotionally dumped on.

"Men being aggressive" excuses male violence. "Boys being hyperactive" means we have low standards for their behavior.

>Women being kinder is a sex trope not a neurological reality.

But what did I freaking say? -.- I did specifically say it was a stereotype, to counter your point that differences between sex necessarily imply an idea of inferiority of women, and here you go lecturing me on how what I said is a false stereotype D:, well no shit. It's an absurd notion. Negative stereotypes go both ways. But that's not the point.

The point is that you clearly have preconceived notions about how the world should be. And damn the facts. I find that an intellectually dishonest way of looking at things. The point is, for instance: in all likelihood there is no difference of, say, intelligence between the male and female population, but you _cannot_ discard that possibility a priori. You can't just say that, because men being inferior to women in that regard, or the reverse, would be undesirable, that you can close your eyes and go la-la-la. You can't. There are physiological differences, hormonal differences, would it be such a stretch that there would be neurological differences too? Or that the hormones would generate different patterns of behaviour in males and females?

I mean, just look at your last sentence. You don't actually care to know if men are naturally more predisposed to aggression, or if boys are more hyperactive and thus have more difficulty at school (if that were the case maybe methods of teaching could be more tailored to that reality, but that is beside the point). You just say that "Boys being hyperactive" would mean we have low standards for their behavior, and as that is a conclusion you do not desire, you reject the premise, without caring to know if it is true or not!

When you close it with we have to be aware of the realities of behavioral difference between the sexes, your beliefs become ambiguous. A thorough restating of positions doesn't make the arguments less clear.

I have the preconceived notion that we cannot assume that there is a biological difference in cognition. You're right. I'm wary of it because people have been doing it for hundreds of years for equally inane reasons. My wife isn't missing 5 ounces of her brain, her brain isn't held together by more delicate strands, the shape of her brain relative to the skull doesn't make her more subservient. What you seem to miss is that this assumption is not rooted in any scientific reasoning -- in fact more and more neuroscience is coming to the conclusion that there are very few differences in brain structure. Catherine Dulac's research from Harvard has been getting close to this, and although I think her conclusions overstep the data that she has, it's very interesting to me. And you've missed my point. I do think that there is a difference in cognition between the sexes, but I do not think that there is a biological basis for this. Just like I don't think that there is a biological basis for the reason that black children score lower on standardized tests (and there is an easier study setup for this, which is excellent!). The thing is, the studies that they've done show pretty similar results between increasing either sexes primary sex hormone and the effects on cognition. There are very similar results when either sex has a hormone imbalance.

I do actually care to know if men are naturally predisposed to aggression. Here are the links to the studies refuting that testosterone causes increased aggression [0][1][2]. And again, I do care that boys are more hyperactive. I don't deny that it's happening, I think it's a problem, and I think that what we can do to fix it is increase discipline and develop healthy coping strategies.

I'm not rejecting the premise, I'm rejecting that there is an inherent flaw in boys that makes them violent, hyperactive rapists. I'm rejecting that there is an inherent flaw in women that makes them docile, logically incapable therapists.

The fact that this is controversial just blows my mind.

How do you feel about the cross cultural differences in expected behavior between the sexes?

[0] http://dx.doi.org/10.1038/nature08711 [1] http://jcem.endojournals.org/cgi/content/abstract/81/10/3578 [2] http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6W...

Equating liking children with the suicidal urge to be killed and eaten is on the overdramatic side.

What's so "detrimental" about women's choices? Is becoming doctors and lawyers so much worse than becoming programmers?

Think about this: How do you know socialization is responsible for these differences, and not inherited personality and interest (average) differences?

What would be observably different in a world where our species worked that way?

TFA isn't a position piece intended to argue the point. It was saying, "hey, this is a problem, here are two numbers indicating the scope of the problem, and here are some people who are working on it." Not "This is a problem, and here are statistics to prove it to skeptics." The data on why women leave tech is out there if you're genuinely interested.

In general, complete data on population trends is very lacking. So if you have a source I would genuinely be interested in examining it.

But I guess where you and I disagree is whether or not the problem identified actually is the problem. Which I think makes creating essentially sexist sub-networks/organizations questionable. If the crux of the issue isn't some sort of gender discrimination (nefarious or not, structural or cultural, etc.) then the proposed solution doesn't actually address the underlying issue leading to the observed macro trend.

Reply on news.ycombinator.com