This here study lists several such traits: e.g., birth weight, adult weight, BMI, blood parameters, physical performance on 60 meter dash, performance in undergraduate exams. They conclude:
Thanks. That was really intersting, not least on grade point averages - where we’ve been led, by statisitcal analysis, to understand girls do better than boys. In short, greater variance drags down the average for boys.
Very interesting!
I find it interesting to but for different reasons -
- It informs to the answer I proposed in post #39: the context is the first line of the abstract:
- The data they use, for example BMI among 2700 people in Norway, conflicts with the results found looking at larger datasets, as referenced above. BMI is clearly actually more variable in females across the United States population. Height is actually more variable in females across the massive datasets used by both the CDC and the WHO. Blood parameters? I don’t have access to the article but it would be nice to know which one they mean. Birth weight, looking through actual large datasets (such as included here - pdf) myself I find that both males and females have an SD of just under 12% of mean birthweight at full term. Still I’ll grant that males are a bit more likely to be born preterm and to do poorer for the gestational age. Taking “physical performance in the 60 meter dash event of 575 junior high school students” or “undergraduate university examination grades” as evidence of inherent biological variability is a bit laughable.
Here is more about where this variability hypothesis comes from:
Well, looks like this is a case of dueling data!
At any rate, whatever the relative merits of the conflicting data, regarding your last point of the origin of the variability hypothesis, can I suggest you read up on the moralistic fallacy? 
The top percentile-- like, the 10th percentile, or something, is men, but more contestants are women, and you have to pass a screening to get on the show, so more women than men make the baseline.
Summers was idiotic in the way he phrased. He all but came out and said “We should figure out why all these men are so smart, and what’s wrong with the couple of deviant women in the math department.”
No, it’s not true for everything. It’s not true for longevity, but that would play into the same “men are expendable” explanation.
The 60 meter dash and undergraduate scores would be terrible ways of comparing men and women, but when you’re comparing men with other men, and women with other women, and talking only about the ranges within those genders to compare them, then they are valid.
Why is weight “clearly” more variable among women in the US?
In any case, I think this thread was meant to be about the outliers. The outliers - the tallest, shortest, highest IQ, etc (for all things except age) do tend to be men, and a lot more men have various psychological conditions or learning disabilities like autism or schizophrenia.
Even when it comes to height, for example, when men tend to be taller than women, the shortest human ever known is male, and so’s the next shortest, and the next. ([url=List of the verified shortest people - Wikipedia]Wiki - seems well cited, and I haven’t counted the people they consider dubious).
Age is the only criterion where women are consistently the outliers. Well, also pregnancy, but let’s discount that. 
I don’t think blood pressure is what is on people’s minds when they think about whether men tend less towards the norm. It probably is true that blood pressure is more variable among women, though I wonder if it would still be true for women who never get pregnant.
Correct. I think I phrased my response in a poor way if it is to be taken literally. By ‘mental’ and ‘physical’, I was thinking about behavioral traits and not strictly physiological ones. Blood pressure may be more variable among women but I wasn’t centered around that type of answer and I do not believe most people that ask it are.
What I was referring to was any type of performance based score whether it is verbal, spatial cognition or math. By ‘physical performance’, I meant skills like golf putting, billiards, ping-pong, video games or car racing where physical attributes should not matter much but all the top performers are still male even though some females have made it near the top.
There is only one unisex performance sport that I am aware of in which a female competitor has consistently beaten most male competitors. That is aerobatic flying. Patty Wagstaff is getting older now but she has been among the best in the world at her sport if not the best for decades and I respect her greatly.
I realize that this breaks down into separate statistical questions. One is whether the male variance is higher than females for any individual skill and the other looks at if a male is the absolute outlier. I haven’t looked at the former question that much personally except for a few traits centered around intelligence measures but I think both questions are interesting.
Scores on a 60 meter dash among 575 junior high school students is not a valid dataset to draw a conclusion about the comparitive genetic ability of genders. Did that school have males encouraged to join sports teams and females less so?
Weight is clearly more variable among females by the very large CDC and WHO datasets referenced upthread. Many more both underweight and obese females than males, the 3 to 97% a larger percent of body weight for females than for males, same as for height.
The op was not asking about the extremely rare single most extreme. It asked if “various standard deviations are usually much lower for women than men” and the statement was made such is true for “almost every physical and mental trait that can be measured.”
That statement is clearly false. There are a handful of traits that it is true for, another handful that the opposite is true for, some with questionable evidence and reasoned debate, and many more that have no evidence of any difference whatsoever.
Psychological conditions? More depression in females. More schizophrenia and autism in males. Bipolar earlier in males but more common in females. Dyslexia more in males. Overall? No difference:
On preview I see that Shagnaasty has modified his assertion but his current claim does not at all answer the question asked.
Top performers in a skill or an activity that is impacted (even marginally) by education or practice does not tell us much about SD even though wider SD can be a possible explanation of it if the mean of genetic potential and both the opportunity to train and the encouragement to train are completely equal across the populations.
If the mean is off by even a smidgen, if one group gets even slightly less opportunity or encouragement or for cultural reasons chooses to or not to pursue the skill in different numbers, then the far end will be impacted with SDs of genetic potential that are exactly the same.
Do males on average have a smidgen faster reflexes, enough to impact the far end where that tiny smidgen is the difference between best in the world and number 163? It seems that it might be so.
Would that level of difference matter to most? Nah. But at the very top level? Oh yeah.
Are males exposed to different environments to large degrees? Of course.
Do more males attempt to succeed in physical skills based activities such that the n is much larger? Of course. The outlier female golf putter may be less likely to have tried golf than the outlier male. Can sociologic factors impact putting performance? Of course.
And again math was mentioned when the cite that Shagnasty provided documented that gender variability and performance in math seems to be highly cuture dependent, more commonly more male variable but also sometimes the opposite, going exactly against the case he is attempting to make and instead more consistent with a cultural explanation.
I have to say, you are changing my mind a bit about this subject. I did think of it in the way the question was posed - about the outliers - but that does tend to mean focusing on traits that affect men more. There definitely are more males who have autism, but more females have depression. Autoimmune disorders is another one.
Still, I’m thinking of the y-hypothesis (Google is being very unhelpful here, but I’m sure I haven’t imagined it) where a lot of the possibly helpful, possibly harmful stuff was dumped onto the y cell, much more so than on the x cell.
I think what you might be thinking about is covered here. It is not though exactly how you are recalling it.
Unlike other chromosomes, including the X, the Y has no partner. Thus when a mutation error occurs in the Y it cannot use the information in its partner as a template for repair and eventually the only other option was removal of the material resulting in the tiny Y of today.
OTOH the genes that are left have very little variation across male humanity:
Still there are other good just-so stories to explain greater male variation if it did exist, which is why I at first stated that I could believe it was. But having a good story in hand for why something might happen is not the same as haing evidence that something does.
The bottom-line is that the hypothesis is put forth as an alternate to sociological ones (be it bias or culture that pushes one gender into one set of direction and the other into others) to explain why the are more males at the highest level of achievement in academe, especially in maths.
It is almost always stated in the formulation that Shagnasty began with: the statement of the “fact” that males have more variability on almost all mental and physical traits (despite his modification later to “performance” on skills based tasks). Yet when critically examined that “fact” apparently has no basis.
Restricting the analysis to cognitive skill sets the hypothesis rests on two legs: frequency of mental retardation also being greater in males; and past IQ studies showing slightly greater SD in males tested than in females tested. And indeed greater cognitive capability SD could be consistent with those findings, but there are other explanations as well.
-
Mental retardation. The most common identified cause of inherited mental retardation is Fragile X. This is passed on the X chromosome and is therefore much more common in males. It impacts the far low end and in no way informs to the high end. Males (as noted upthread) are also more likely to be born premature and to do more poorly when born premature than females, which also contributes to greater mental retardation rates than females. Male fetuses may also be more at risk to maternal factors (for example mental retardation as the result of low maternal thyroid hormone, historically referred to as “cretinism” is more common in males than females). This also informs naught about SD to the high end.
-
IQ. The argument based on IQ test results rests both on the assumption that IQ test results reflect genetic potential that it is not impacted by experience and culture and the asumption that the IQ test is an accurate measure of “intelligence” - both of these assumptions are highly debatable. See all the Bell Curve debates we’ve had here (and I have no interest in participating in another one of them).
I do not state that it is impossible that there is a broad difference SD for cognitive ability in males than in females. I do however dispute any claim that it is clearly established that there is one or that such is part of a pattern of greater male variabilty in general. The latter has no evidence to support it at all and the former is a hypothesis supported by a small amount of evidence and much speculation.