So if prenatal testosterone has such a deleterious effect on females, what effect does it have on males? How do all those metrics they measured for the girl half of the twin pairs compare to their sibling?
Since the mother already have a lot of estrogen in her blood, it might simply be that the effect of a female twin does not impact the amount of circulating hormones. The effect of increased estrogen in the male twin similar to the increase in testosterone in female twins would be interesting, but it might just not be something that can occur naturally.
Thus if I speculate a bit, the reason for the non-symetric effect should be that the mother and father does not have symmetric roles during the pregnancy. If half the time it was the father that carried the child we would likely have seen a similar effect on males with female twins.
What would be interesting is if they looked at mothers with different levels of estrogen and see if that effects male offspring, in particular those with exceptional high and low levels.
That's not what I asked. I asked what effect in utero testosterone had on boys. Did they have lower metrics than their sisters? Did boys with higher levels of in utero testosterone have lower metrics than boys with lower levels of in utero testosterone?
I'm not sure how they would go about experiment design for this.
In the case of twins they know the mechanism for increased testosterone, and they have a big control population. Short of everyone having a prenatal hormone measurement (which presumably is not done typically) how would you identify the subjects?
This phrasing is really not accurate. It's a small difference, trivial, dwarfed by things like socioeconomic status, educational opportunities, rearing environment quality, presence of an alcoholic parent, nationality etc.
Critically, however, freemartins are a form of microchimerism. They literally have cells with Y chromosomes in their bodies, received from the male twin through the shared chorion. You're correct that there is a shared mechanism - they receive testosterone and anti-Müllerian hormone through this link, which are primarily resonsible for masculinizing the freemartin.
In contrast, this mechanism is thought to be only testosterone exposure in utero.
A better comparison, then, might be intra-uterine position in mice [1]. This is similarly a pure testosterone-exposure mechanism (no microchimerism). Briefly, mouse pups are arranged in linear fashion within the uterine horn, and so a given (non-end) female pup has 0, 1, or 2 males adjacent to her. There are clear physiological and behavioral differences in these 0M, 1M and 2M mice, corresponding to the level of testosterone exposure.
> In rural areas folklore often claimed this condition was not just peculiar to cattle, but extended also to human twins; this belief perpetuated for generations, as was mentioned in the writings of Bede. (de Albuquerque, Martim Notes and Queries Volume 2. 1857 by Oxford University Press, p. 149)
Actual observed effect that has now been confirmed by science? Or, a superstition that got lucky?
Notably, they didn't control for social effects by looking at female and male sibling pairs with an age difference of 10-18 months. How much of this is parents favoring the male sibling?
"To separate the effects of fetal testosterone from postnatal socialization, the research team repeated their analyses focusing only on female twins whose twin sibling—either twin sister or twin brother—died shortly after birth, and thus they were raised as singletons."
My girlfriend has a twin brother. Since they were born in Haiti and we live in the US I'm not sure how relevant this study would be towards her, meaning the socioeconomic changes between living in those two countries is pretty wide. How does prenatal test affect someone after they've gone through puberty? I just don't see how that can be attributed towards college graduation or marriage rates. Is that not a far reaching statement to associate the two?
> How does prenatal test affect someone after they've gone through puberty
The test doesn't affect them. Hormone levels during development are the primary way the body controls development, so even small alterations in hormone levels in utero would be expected to have large physiological consequences.
> Is that not a far reaching statement to associate the two
Luckily, the field of statistics gives us ways to reason about whether or not to believe an observed effect.
In this case[1], the researchers used data on 100% of Norwegian births from 1967-1978 (n=728,842) including 13,800 twins. From these large n's the researchers are able to get the statistical power to measure these effects at the level of precision they did. The p-values for all the reported effects are presented in the appendix [2] and ranged from <.001 to .054 for the outcomes reported to be different.
It's worth noting that many of these differences had already been observed in multiple previous studies. The difference in marriage rates, for example, already appeared in a study analyzing records of 18th and 19th century (1734–1888) Finns [3].
So, in summary, while these statements are interesting, they're in no way an overreach, given the amount of data analyzed.
Being statistically illiterate, could I get an explanation of of "The p-values for all the reported effects are presented in the appendix [2] and ranged from <.001 to .054 for the outcomes reported to be different."
I looked at the link, but I'm still not sure what the table means, or how I could explain what this means to someone else who doesn't know stats.
I'd like to share this with her, but neither of us know stats so the significance is kinda lost.
P-value is the universal* way of expressing statistical likelihood. It corresponds to a percentage: p=.05 just means 5%, and p=.001 means 0.1%, etc.
It's often inaccurately explained as the likelihood of getting our results through chance alone. That's wrong for reasons that are technically important, but not in a way that really inhibits understanding of the strength of results that have small p-values.
* It has flaws, and a growing number of researchers believe it should not have the prominent importance currently placed on it.
---- Stop reading here if you're already satisfied. ----
We want to measure if there's a difference between two groups. So, we take measurements of a portion of group A, and measurements of a portion of group B.
Mathematically, we assume that what we've actually done is sampled from the same population both times. If that's true, our data sets should be quite similar to one another, but of course there will be some difference just due to noise.
So, we compute mathematically the chance of seeing a difference at least as large as the one we see between our data sets, if that assumption is true.
We decide beforehand on a small error rate. If not otherwise stated, 5% (p=.05) is the universal standard.
If we find that the likelihood of observing a same-or-larger difference between the populations is smaller than that already-small 5% chance, we REJECT our assumption that the samples came from the same population, and conclude that the populations must actually be different.
In other words, we conclude that we measured an actual difference that contrasts two mathematically-separate populations.
In this case* the direction of the correlation is 100% knowable: economic and education outcomes in the 1990s cannot possibly have altered whether or not you had a male twin in the 1970s, unless you can reverse time's arrow.
*The OP was specifically asking about marriage rate and education outcomes. The paper also examined fertility. It's at least remotely plausible that there's some 3rd/confounding biological mechanism that actually the cause of both the presence of a male twin and the increased fertility of the female twin. In that case I would agree that 'correlation' is the more appropriate choice of term, but note that it's still correct in that case to assign 'effect' to both of the observed properties as they'd be downstream of the confounder.
It is so nice to come on HN and get some relief from the monotonous humorlessness that is life in USA today. For a moment, I imagine that I can laugh at funny statements no matter which way they point politically, rather than restricting my entertainment to grumbling in approved ways about whomever we're supposed to despise now.
It’s a WASP thing. They were killjoys when they got kicked out of Europe four centuries ago. They’re kill joys today.
Today, shocked at their ancestors’ crimes, they dilute their sense of responsibility by dropping the ASP, and making broad generalizations about whites (really, who did the Balkans ever hurt other than themselves? What harm have the Amish done to ethnic minorities?)
With smaller data sets I find the summary to be less valuable than the actual paper. In this case we are talking about 4,533 twin pairs where one is a male and the other female. The female twin had a 2.8-percentage point higher probability of dropping out of high school compared to other female students, and a similar 1.9-percentage point lower probability of graduating from college.
Those two numbers seems connected, ie that individuals that have higher risk of high school drop out have also a higher risk of not graduating. The study do not cross compare this with male students which also have higher drop out rate and lower graduation rate compared to female students, but some relation seems implied.
The study also notes a lower probability of ever having been married by age 32, and lower probability of employment. The article does not explore if the effect could be exclusively a result of lower education.
As a final thought, the sample size is relatively small and the effect relatively minor in absolute numbers. The paper does not give out any P values or confidence intervals for their primary findings. The relative effect is large, but I interpret that more like a warning flag. In absolute terms we are talking about 136 additional students born between 1967 and 1978 that dropped out compared to a null hypothesis. The same number for college drop out was 49 students that did not graduate. With so few individuals having that large effect on the conclusion I personally will wait a bit for replication studies.
> The paper do not give out any P values [...] for their primary findings. The relative effect is large, but I interpret that more like a warning flag.
Incorrect. The p-values -- all 85 of them -- are presented in the appendix[1], linked to in the first paragraph under the 'Results' heading, along with means, standard deviations, and number of observations. The article text[2] repeatedly references these tables when making each claim.
> The paper do not give out any [...] confidence intervals for their primary findings.
Incorrect. Confidence intervals are presented in 100% of the figures.
> The article does not explore if the effect [on marriage or employment] could be exclusively a result of lower education.
Incorrect. They explicitly consider the effect of educational attainment on future economic earnings.
"Behavioral changes downstream of prenatal testosterone exposure, such as any effect of disruptive behavior on schooling or educational attainment, are likely candidate pathways linking male co-twin exposure to later earnings. Consistent with this interpretation, conditioning on educational attainment in the earnings equation reduced the coefficient on opposite-sex from −0.086 (SE, 0.021) to −0.042 (SE, 0.014), suggesting that roughly 50% of intensive margin labor market effects could be explained by educational outcomes that are upstream of earnings.
> Those two numbers [H.s. dropout rate / college graduation rate] seems connected ... some relation seems implied.
These measures are already known to be strongly correlated, along with the correlations to the probability of inclusion within the labor force and lifetime earning potential. The paper makes no claims that these four measures are independent from one another.
The paper does explicitly make clear when findings of independence are relevant, particularly in usually-correlated variables. In this case I'm referring to their findings of a fertility rate effect independent from marriage rate.
Thank you, missed the appendix. Also see that in the figure page list CI as 95%.
> These measures are already known to be strongly correlated... The paper does explicitly make clear when findings of independence are relevant
The first few paragraphs in the article here and the Significance section in the actually paper imply a strong independence. My point above is basically that the independence looks to be much weaker than a casual read might imply. It also boast with the largest study ever done on the subject, which might be true, but the absolute number are actually quite low with only a few individuals being responsible for a large effect in the data. That is one of the biggest warning signs that exist in this kind of research. I thus get skeptical, at which point I start to look why there seems to be a disconnect.
Sort of. It compares female-female to male-female twin pairs in analyzing subsequent maternal fertility, to ensure the relevance of the deceased co-twin approach.
Search for 'Table S5' in the paper and the appendix.
The researchers never once use the term 'life success'. [1]
That a behavioral difference (marriage rate) seems to be influenced, in females, by in-utero exposure to testosterone (independent of any effect on fertility rate, at that) is certainly a scientifically relevant finding. Links between psychology and biology are still infrequent.
Agree. But it was listed clustered with “college/school graduation likelihood, wage earnings, fertility (again hmm for this one)” and other indicators which are arguably termable as life success metrics (or any other suitable verbiage).. my point is that “one of these things is not like the others”.
Marriage rate differences have already been shown to be correlated with hormone levels.[1][2] Measuring whether such a difference exists for the level of exposure in-utero simply by having a male twin is interesting. If anything, the natural interpretation of this data would be to expand the range of behavior (marriage rate) considered 'typical' for females.
Similarly, fertility differences have been observed already for e.g. mice exposed to different levels of testosterone in utero. The 'freemartin' cows mentioned by another poster here are infertile for a similar reason. Hormones play a huge role developmentally and are almost 100% responsible for morphological sexual anatomy.
There's no agenda present in measuring these outcomes. These are exactly the outcomes that would be expected naturally to be different. If anything, the unusual inclusion is the economic and educational factors.
Has nobody recognized that not a single biologist, chemist, neurosurgeon, hormone specialist, neo-natal specialist, or just general medical doctor participated or helped produce this study?
Here are the authors:
Aline Bütikofer - Department of Economics, Norwegian School of Economics, 5045 Bergen, Norway
David N. Figlio - School of Education and Social Policy, Northwestern University, Evanston, IL 60208
Krzysztof Karbownik - Institute for Policy Research, Northwestern University, Evanston, IL 60208
Christopher W. Kuzawa - Department of Economics, Emory University, Atlanta, GA 30322
Kjell G. Salvanes - Department of Anthropology, Northwestern University, Evanston, IL 60208
So we have, two economists, one anthropologist, and two people who specialize in "social policy."
They did not work with these twins, they did not do their own medical research and assessment. They literally "[used] data on all births in Norway (n = 728,842, including 13,800 twins) between 1967 and 1978 to show that females exposed in utero to a male co-twin have a decreased probability of..."
This is not meaningful research.
This is playing with statistics and cherry picking existing scientific literature to produce a headline that propels an agenda.
Why do you suggest fields other than anthropology or economics should be the qualification for analyzing population-level economic effects?
Are you aware that the vast majority of scientific research is done by PhDs, not MDs? And that both are doctorates?
> ... propels an agenda. This is propaganda of the highest order.
What's the agenda? And did that agenda apply to the 2011 paper calling for more research in this area - "Evaluating the twin testosterone transfer hypothesis: a review of the empirical evidence"[1]?
Was that agenda behind this 2012 Japanese study[2] which examined the ratio of middle finger to pinky finger in twins? It's the same hormone transfer hypothesis being examined. What propaganda could there possibly be about the lengths of one's fingers?
Was that agenda present in 1993? [3]
Does that agenda include propoganda about females becoming better at 3d-spatial mental manipulation tasks? [4] If so, why?
Is propaganda about mouse behaviors part of that agenda? [5] If not, why not? Still that same hormone transfer hypothesis being discussed.
Perhaps this 1988 paper on intrauterine position of gerbil fetuses [6] and resulting variance in hormonal exposure from siblings was the birth of this propaganda agenda?
Or is it more sinister than that, and this agenda extends all the way back to 1959 [7], causing these doctors to 'play with statistics' instead of doing what they said - applying testosterone in-utero and watching for behavioral changes? And then I suppose Dr Phoenix would do more 'playing' in 1967 [8] instead of actually measuring testosterone uptake rates of the various tissues he avoided actually manipulating eight years prior...
Because the presupposition is that testosterone exposure in-utero is to blame.
Without actually addressing this claim by medical professionals, everything else is manipulation of data to prove a point to address 'social policy' or an economic agenda.
Moreover, the sample size, location, and dates of birth, are small and narrow, comparatively speaking. You cannot make sweeping claims like this using such small data set. But they do, and the "Material and Methods" section of this paper goes to great length to explain how they've made these extrapolations.
This is statistical sausage.
I have no doubt you could write a similar paper that claims the opposite; that females with male twins do much better. Just like this team did, you can find, extrapolate and present data to support the idea.
additional testosterone means you read as (and possibly are) queer more often, which in turn means you gotta pay the butch tax. This result, correctly interpreted, will be unsurprising to most queer and trans people.
I have a twin sister, twin brother talking, and unfortunately this study does not show credence for me. She's smart as a whip, in a lawyer way, as I'm more math/science. We both make six figures and both are very career focused. I had to laugh at this study...she used her words like a dagger to defend against my constant barrage of physical torture growing up. It was a very contentious relationship.
56 comments
[ 4.7 ms ] story [ 119 ms ] threadThus if I speculate a bit, the reason for the non-symetric effect should be that the mother and father does not have symmetric roles during the pregnancy. If half the time it was the father that carried the child we would likely have seen a similar effect on males with female twins.
What would be interesting is if they looked at mothers with different levels of estrogen and see if that effects male offspring, in particular those with exceptional high and low levels.
In the case of twins they know the mechanism for increased testosterone, and they have a big control population. Short of everyone having a prenatal hormone measurement (which presumably is not done typically) how would you identify the subjects?
This phrasing is really not accurate. It's a small difference, trivial, dwarfed by things like socioeconomic status, educational opportunities, rearing environment quality, presence of an alcoholic parent, nationality etc.
https://en.wikipedia.org/wiki/Freemartin
In contrast, this mechanism is thought to be only testosterone exposure in utero.
A better comparison, then, might be intra-uterine position in mice [1]. This is similarly a pure testosterone-exposure mechanism (no microchimerism). Briefly, mouse pups are arranged in linear fashion within the uterine horn, and so a given (non-end) female pup has 0, 1, or 2 males adjacent to her. There are clear physiological and behavioral differences in these 0M, 1M and 2M mice, corresponding to the level of testosterone exposure.
[1]: https://en.wikipedia.org/wiki/Prenatal_testosterone_transfer...
> In rural areas folklore often claimed this condition was not just peculiar to cattle, but extended also to human twins; this belief perpetuated for generations, as was mentioned in the writings of Bede. (de Albuquerque, Martim Notes and Queries Volume 2. 1857 by Oxford University Press, p. 149)
Actual observed effect that has now been confirmed by science? Or, a superstition that got lucky?
This effect exists in twins raised as singletons and twins raised with their brother.
The test doesn't affect them. Hormone levels during development are the primary way the body controls development, so even small alterations in hormone levels in utero would be expected to have large physiological consequences.
> Is that not a far reaching statement to associate the two
Luckily, the field of statistics gives us ways to reason about whether or not to believe an observed effect.
In this case[1], the researchers used data on 100% of Norwegian births from 1967-1978 (n=728,842) including 13,800 twins. From these large n's the researchers are able to get the statistical power to measure these effects at the level of precision they did. The p-values for all the reported effects are presented in the appendix [2] and ranged from <.001 to .054 for the outcomes reported to be different.
It's worth noting that many of these differences had already been observed in multiple previous studies. The difference in marriage rates, for example, already appeared in a study analyzing records of 18th and 19th century (1734–1888) Finns [3].
So, in summary, while these statements are interesting, they're in no way an overreach, given the amount of data analyzed.
1: https://www.pnas.org/content/early/2019/03/14/1812786116
2: https://www.pnas.org/content/pnas/suppl/2019/03/14/181278611...
3: https://www.pnas.org/content/104/26/10915?ijkey=2c55ba770f14...
I'd like to share this with her, but neither of us know stats so the significance is kinda lost.
P-value is the universal* way of expressing statistical likelihood. It corresponds to a percentage: p=.05 just means 5%, and p=.001 means 0.1%, etc.
It's often inaccurately explained as the likelihood of getting our results through chance alone. That's wrong for reasons that are technically important, but not in a way that really inhibits understanding of the strength of results that have small p-values.
* It has flaws, and a growing number of researchers believe it should not have the prominent importance currently placed on it.
---- Stop reading here if you're already satisfied. ----
We want to measure if there's a difference between two groups. So, we take measurements of a portion of group A, and measurements of a portion of group B.
Mathematically, we assume that what we've actually done is sampled from the same population both times. If that's true, our data sets should be quite similar to one another, but of course there will be some difference just due to noise.
So, we compute mathematically the chance of seeing a difference at least as large as the one we see between our data sets, if that assumption is true.
We decide beforehand on a small error rate. If not otherwise stated, 5% (p=.05) is the universal standard.
If we find that the likelihood of observing a same-or-larger difference between the populations is smaller than that already-small 5% chance, we REJECT our assumption that the samples came from the same population, and conclude that the populations must actually be different.
In other words, we conclude that we measured an actual difference that contrasts two mathematically-separate populations.
What you mean is correlation, right? Effect would imply causation.
In this case* the direction of the correlation is 100% knowable: economic and education outcomes in the 1990s cannot possibly have altered whether or not you had a male twin in the 1970s, unless you can reverse time's arrow.
*The OP was specifically asking about marriage rate and education outcomes. The paper also examined fertility. It's at least remotely plausible that there's some 3rd/confounding biological mechanism that actually the cause of both the presence of a male twin and the increased fertility of the female twin. In that case I would agree that 'correlation' is the more appropriate choice of term, but note that it's still correct in that case to assign 'effect' to both of the observed properties as they'd be downstream of the confounder.
Either that or it may sound like you’re insulting all males.
Either way, funny!
It’s a WASP thing. They were killjoys when they got kicked out of Europe four centuries ago. They’re kill joys today.
Today, shocked at their ancestors’ crimes, they dilute their sense of responsibility by dropping the ASP, and making broad generalizations about whites (really, who did the Balkans ever hurt other than themselves? What harm have the Amish done to ethnic minorities?)
[1] (also Latin@ ?? - can’t keep track.)
[1]https://en.wikipedia.org/wiki/Digit_ratio
Those two numbers seems connected, ie that individuals that have higher risk of high school drop out have also a higher risk of not graduating. The study do not cross compare this with male students which also have higher drop out rate and lower graduation rate compared to female students, but some relation seems implied.
The study also notes a lower probability of ever having been married by age 32, and lower probability of employment. The article does not explore if the effect could be exclusively a result of lower education.
As a final thought, the sample size is relatively small and the effect relatively minor in absolute numbers. The paper does not give out any P values or confidence intervals for their primary findings. The relative effect is large, but I interpret that more like a warning flag. In absolute terms we are talking about 136 additional students born between 1967 and 1978 that dropped out compared to a null hypothesis. The same number for college drop out was 49 students that did not graduate. With so few individuals having that large effect on the conclusion I personally will wait a bit for replication studies.
Incorrect. The p-values -- all 85 of them -- are presented in the appendix[1], linked to in the first paragraph under the 'Results' heading, along with means, standard deviations, and number of observations. The article text[2] repeatedly references these tables when making each claim.
> The paper do not give out any [...] confidence intervals for their primary findings.
Incorrect. Confidence intervals are presented in 100% of the figures.
> The article does not explore if the effect [on marriage or employment] could be exclusively a result of lower education.
Incorrect. They explicitly consider the effect of educational attainment on future economic earnings.
"Behavioral changes downstream of prenatal testosterone exposure, such as any effect of disruptive behavior on schooling or educational attainment, are likely candidate pathways linking male co-twin exposure to later earnings. Consistent with this interpretation, conditioning on educational attainment in the earnings equation reduced the coefficient on opposite-sex from −0.086 (SE, 0.021) to −0.042 (SE, 0.014), suggesting that roughly 50% of intensive margin labor market effects could be explained by educational outcomes that are upstream of earnings.
> Those two numbers [H.s. dropout rate / college graduation rate] seems connected ... some relation seems implied.
These measures are already known to be strongly correlated, along with the correlations to the probability of inclusion within the labor force and lifetime earning potential. The paper makes no claims that these four measures are independent from one another.
The paper does explicitly make clear when findings of independence are relevant, particularly in usually-correlated variables. In this case I'm referring to their findings of a fertility rate effect independent from marriage rate.
1: https://www.pnas.org/content/pnas/suppl/2019/03/14/181278611...
2: https://www.pnas.org/content/early/2019/03/14/1812786116
Thank you, missed the appendix. Also see that in the figure page list CI as 95%.
> These measures are already known to be strongly correlated... The paper does explicitly make clear when findings of independence are relevant
The first few paragraphs in the article here and the Significance section in the actually paper imply a strong independence. My point above is basically that the independence looks to be much weaker than a casual read might imply. It also boast with the largest study ever done on the subject, which might be true, but the absolute number are actually quite low with only a few individuals being responsible for a large effect in the data. That is one of the biggest warning signs that exist in this kind of research. I thus get skeptical, at which point I start to look why there seems to be a disconnect.
Search for 'Table S5' in the paper and the appendix.
That a behavioral difference (marriage rate) seems to be influenced, in females, by in-utero exposure to testosterone (independent of any effect on fertility rate, at that) is certainly a scientifically relevant finding. Links between psychology and biology are still infrequent.
1: https://www.pnas.org/content/early/2019/03/14/1812786116
Similarly, fertility differences have been observed already for e.g. mice exposed to different levels of testosterone in utero. The 'freemartin' cows mentioned by another poster here are infertile for a similar reason. Hormones play a huge role developmentally and are almost 100% responsible for morphological sexual anatomy.
There's no agenda present in measuring these outcomes. These are exactly the outcomes that would be expected naturally to be different. If anything, the unusual inclusion is the economic and educational factors.
1: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5501487/
2: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3540120/
Here are the authors:
Aline Bütikofer - Department of Economics, Norwegian School of Economics, 5045 Bergen, Norway David N. Figlio - School of Education and Social Policy, Northwestern University, Evanston, IL 60208 Krzysztof Karbownik - Institute for Policy Research, Northwestern University, Evanston, IL 60208 Christopher W. Kuzawa - Department of Economics, Emory University, Atlanta, GA 30322 Kjell G. Salvanes - Department of Anthropology, Northwestern University, Evanston, IL 60208
So we have, two economists, one anthropologist, and two people who specialize in "social policy."
They did not work with these twins, they did not do their own medical research and assessment. They literally "[used] data on all births in Norway (n = 728,842, including 13,800 twins) between 1967 and 1978 to show that females exposed in utero to a male co-twin have a decreased probability of..."
This is not meaningful research.
This is playing with statistics and cherry picking existing scientific literature to produce a headline that propels an agenda.
This is propaganda of the highest order.
Are you aware that the vast majority of scientific research is done by PhDs, not MDs? And that both are doctorates?
> ... propels an agenda. This is propaganda of the highest order.
What's the agenda? And did that agenda apply to the 2011 paper calling for more research in this area - "Evaluating the twin testosterone transfer hypothesis: a review of the empirical evidence"[1]?
Was that agenda behind this 2012 Japanese study[2] which examined the ratio of middle finger to pinky finger in twins? It's the same hormone transfer hypothesis being examined. What propaganda could there possibly be about the lengths of one's fingers?
Was that agenda present in 1993? [3]
Does that agenda include propoganda about females becoming better at 3d-spatial mental manipulation tasks? [4] If so, why?
Is propaganda about mouse behaviors part of that agenda? [5] If not, why not? Still that same hormone transfer hypothesis being discussed.
Perhaps this 1988 paper on intrauterine position of gerbil fetuses [6] and resulting variance in hormonal exposure from siblings was the birth of this propaganda agenda?
Or is it more sinister than that, and this agenda extends all the way back to 1959 [7], causing these doctors to 'play with statistics' instead of doing what they said - applying testosterone in-utero and watching for behavioral changes? And then I suppose Dr Phoenix would do more 'playing' in 1967 [8] instead of actually measuring testosterone uptake rates of the various tissues he avoided actually manipulating eight years prior...
1: https://www.ncbi.nlm.nih.gov/pubmed/21893061
2: https://www.ncbi.nlm.nih.gov/pubmed/22270254
3: https://www.ncbi.nlm.nih.gov/pubmed/8240211
4: https://www.ncbi.nlm.nih.gov/pubmed/21094200
5: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3212613/
6: https://www.ncbi.nlm.nih.gov/pubmed/3387473
7: https://www.ncbi.nlm.nih.gov/pubmed/14432658
8: https://www.ncbi.nlm.nih.gov/pubmed/6020216
Without actually addressing this claim by medical professionals, everything else is manipulation of data to prove a point to address 'social policy' or an economic agenda.
Moreover, the sample size, location, and dates of birth, are small and narrow, comparatively speaking. You cannot make sweeping claims like this using such small data set. But they do, and the "Material and Methods" section of this paper goes to great length to explain how they've made these extrapolations.
This is statistical sausage.
I have no doubt you could write a similar paper that claims the opposite; that females with male twins do much better. Just like this team did, you can find, extrapolate and present data to support the idea.
That's a little much.
I don't see any reason to believe anything other than the simple (and minor) outcome here: a small deviation in outcomes due to prenatal hormones.
That's not propaganda it's just bad math, worst case.
As that paper explains in its introduction, this is "probably the most robust cognitive sex difference", emphasis mine, which is why it's interesting.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4438761/