4 ms·

Here’s the problem. Let’s say women are on average 15% less aggressive than men on average. However, we know this type of thing is always a normal distribution,

Here’s the problem. Let’s say women are on average 15% less aggressive than men on average. However, we know this type of thing is always a normal distribution, so what we have are two overlapping distributions. Classic stats 101 stuff.^ What are you going to meaningfully do with this information? If a woman walks in your store, she could be on the high end of the distribution and actually be more aggressive than the average man. The point I’m getting at is you can’t usually make meaningful assumptions about individuals based on statistical data gathered about populations. 70% of businesses fail but your business doesn’t have a 70% chance of failure, it depends on the specifics. So unless you work on projects like public health or prison design or something where you really are looking at populations in aggregate, this kind of data just isn’t that useful. At best it’s a curiosity and at worst it lets the small minded back up their biases with (irrelevant) data. Even if a doctor knows that men are 2x as likely to get heart disease, he still needs to check his female patients for it if they exhibit the symptoms. If he looks at the data and stops checking women for heart disease then that is an ethical failure and just bad medicine. ^ https://www.researchgate.net/profile/Shahab-Boumi/publication/340134676/figure/fig11/AS:961190725435392@1606177135041/Overlap-between-two-normal-distributions.ppm https://www.researchgate.net/profile/Shahab-Boumi/publicatio...

OJFord 2y ago
I don't know what you're trying to do with it in your hypothetical scenario, but you can judge the expected value of aggressiveness based on distribution for the sex and be right on average. Still stats 101. > Even if a doctor knows that men are 2x as likely to get heart disease, he still needs to check his female patients for it if they exhibit the symptoms. Interesting example, because it's actually one that shows these things are useful. Women have worse outcomes despite lower incidence, because of underreported/played-down chest pain from familiarity with monthly cramp pain. So where we men might be saying 'oh god, my chest, it's the worst thing imaginable', in triage it pays to know that the woman saying 'ooh, my chest feels a little tight' is actually experiencing exactly the same thing.
lapcat 2y ago
I think the author addresses these points: > the multivariate findings may help answer a question people have been puzzling about in psychology for quite awhile: Why do we have all these studies showing that male and female behaviors are so similar, yet people in everyday life continue to think as if males and females were very separable? It is possible that people in everyday life are actually closer to the truth because when we reason about personality, we rarely reason about one trait at a time. > rarely do we consider the harm that could be caused by ignoring sex differences! One can think of many ways in which pretending something doesn't exist may actually cause greater harm psychologically than accepting the facts of the matter. > I am a strong believer that individual differences are more important than sex differences. Nevertheless, sex differences are also part of the picture, and may be particularly detrimental to a relationship if all partners go into the marriage thinking that they "should not exist", instead of coming to a healthy acceptance of sex differences

OJFord 2y ago

I don't know what you're trying to do with it in your hypothetical scenario, but you can judge the expected value of aggressiveness based on distribution for the sex and be right on average. Still stats 101. > Even if a doctor knows that men are 2x as likely to get heart disease, he still needs to check his female patients for it if they exhibit the symptoms. Interesting example, because it's actually one that shows these things are useful. Women have worse outcomes despite lower incidence, because of underreported/played-down chest pain from familiarity with monthly cramp pain. So where we men might be saying 'oh god, my chest, it's the worst thing imaginable', in triage it pays to know that the woman saying 'ooh, my chest feels a little tight' is actually experiencing exactly the same thing.

lapcat 2y ago

I think the author addresses these points: > the multivariate findings may help answer a question people have been puzzling about in psychology for quite awhile: Why do we have all these studies showing that male and female behaviors are so similar, yet people in everyday life continue to think as if males and females were very separable? It is possible that people in everyday life are actually closer to the truth because when we reason about personality, we rarely reason about one trait at a time. > rarely do we consider the harm that could be caused by ignoring sex differences! One can think of many ways in which pretending something doesn't exist may actually cause greater harm psychologically than accepting the facts of the matter. > I am a strong believer that individual differences are more important than sex differences. Nevertheless, sex differences are also part of the picture, and may be particularly detrimental to a relationship if all partners go into the marriage thinking that they "should not exist", instead of coming to a healthy acceptance of sex differences