Long time ago I did some first-hand research on speed dating results and read a lot of papers. I am not familiar by the recent research, but I find it unlikely that the behavior of people would have changed dramatically.
I found that when we look at the behavior, such as replying to messages on OkCupid or Tinder, selecting a partner in speed dating events, or who people choose to dance with in couples dancing events, women behave differently from men.
The behavior of women generally follows the power law distribution, and the behavior of men follows the normal distribution.
This means that women rate the attractiveness of men based on something that is based on preferential attachment, for example how women believe others rate the attractiveness of the man, whereas men rate the attractiveness of the women on something that is independent of the evaluations of others.
When the pool of potential mates grows, the inequality experienced by most men grows, whereas the inequality experienced by most women stays the same. And the pool s growing due to internet dating.
Interesting. I would like start off on an updated literature adventure. Do you have a link to a review article, especially one that talks about the different distributions?
No. I was planning to write up about my findings but never got to it. I may have some graphs laying out somewhere. To my knowledge, the normal/power law distribution idea is my original idea and have never read anybody refer to it. I have written few times about it on this site, and few people appear to have picked the idea up. But you don't have to take my word on it being a real phenomena, just look at the data.
Most people I’ve seen would agree that female rating of males doesn’t follow a typical normal distribution like men’s does. This is why everyone talks about “the binary.” It’s definitely not a typical normal distribution - maybe more like a power law like you’re saying.
I haven't analyzed the data, so I'm going purely off of the article's numbers, but the article states that 1) the median male is a 5.9 out of 10, and 2) each individual's ratings follow a normal distribution.
Considering our median male, that gives us the following quantiles:
I played around some with the data, which they linked. This is a quick and dirty analysis but it seems like the standard deviation in the data is a lot higher than 0.4 and their thesis has more support:
And, since it's a little sloppy to just take the medians of the means and standard deviations and assume that that's meaningful, here's the direct answer to the question: "What percentage of males would be thought to be above an X by the top 20% and top 1% female?"
So, the data suggests that, for 71% of males, at least 1 in every 100 women will consider them at least a 9, given that the ratings of the male's attractiveness are uniformly distributed. (but, other comments seem to suggest that those ratings are probably not a uniform distribution)
Interesting data but the write up is poor. Graphs would be very helpful. There are two important variations here: how much does attractiveness scores for _the same person_ vary by different raters, and how does that compare to the variation for _different people_ by the same rater.
Saying what proportion finds the median person in the top percentile requires some more complicated math than simply looking up the width of a bell curve.
I can totally believe the beauty-in-the-eye-of-beholder thesis, and it’s encouraging to think that I just need to find the 3% of people that will think I’m gorgeous. But it would be interesting to run the numbers and test the following thesis: are these people all the ones that are on average way uglier than me? I’m betting they are … and now the Tubes song “She’s a Beauty” is playing in my head…
The concepts of beauty and attraction should not have been mixed up by the OP. The underpinning article speaks only of attraction (and intelligence, ambition and other parameters measured separately) and never mentions beauty - which is also, and arguably, linked to aesthetics rather than to pure personal choice - at least from a scientific paper standpoint.
Something or somebody is beautiful if it is pleasant to see it.
Somebody is attractive if you enjoy their company and you have the desire to touch them and engage in activities that would cause mutual pleasure.
These are distinct feelings, even if there exists some correlation between them, mostly because when you consider somebody ugly, you are unlikely to be attracted to them. The reverse is not true, you can consider someone as very beautiful without feeling the slightest attraction toward them.
The connection between beauty and pleasantness seems to be much weaker than suggested.
There are things that are pleasent and lack any form of particular beauty. To mind come, interior decoration, roman statues, tiktok-videos..
And there are things that have inherent beauty partly because of their wretchedness and being unpleasant. Prime example would be greek tragedy. You and the protagonist know that it will end horribly and it is almost torture to see it unfold. But still, it is beautiful.
To go a bit deeper on attration and beauty. My personal experience showed me that beauty tends to be a (surprisingly low) threshold requirement for attraction.
I have referred to "beautiful" only with its strict concrete meaning, as applicable to the sensation caused by something you are looking at, like a beautiful flower, a beautiful waterfall or a beautiful human.
You have also used "beautiful" in its generalized abstract sense, when it becomes applicable to things like a beautiful mathematical theorem or a beautiful computer algorithm.
In the second sense of the word, there is no relationship with the attractiveness of a human. In the first sense of the word, I recognize anything beautiful when I see it, without thinking about a reason. In the second sense, I realize that something is beautiful only after an intellectual analysis of it.
Also in the first sense, the word "beautiful" is frequently reused for other sensations than vision, e.g. for a beautiful song or a beautiful fragrance, though it would have been better to have distinct words for these cases.
There's a review and meta-analysis somewhere of these kinds of studies (I should be working so am not going to look it up) estimating that about 1/3 of the variance in attractiveness ratings can be attributed to the person being rated, 1/3 due to rater, and 1/3 due to the interaction.
The most interesting part of that to me is on selectivity. Male selectivity is invariant to changes in group size, while doubling the group size (speed dating context, something like doubling from 10 to 20 IIRC) makes women increase their selectivity by 13%.
In the contemporary world of OLD, imagine you scaled that group size from 20 to 2000. You would see a massive increase in selectivity (around 10 doublings), which accounts for a lot of the issues with OLD, while at the same time offering an alternative: focus on small groups instead of the digital meat market.
And yet this is directly in opposition to the old OKCupid data in which men rating women was a bell curve while women rating men was almost a power law (or highly skewed curve).
And the breeeding age opinions are most critical to the people who most care about attractiveness.
It's nice when the 60 year old grandparent thinks you're attractive. It matters whether the 22 year old clerk thinks you're attractive.
The issue with the OkC data is that self-reports aren't particularly useful. The most meaningful metric is actual conversions (in the OLD context, you could call that either a match, the existence of a reply-response pair, or something else along those lines).
That data is much harder to come by, but it'd show a significant correlation in who people choose to convert with.
IIRC, OkC also had one classic post which was something between self-reporting and conversion (how you appear to use the term) -- how often men respond to messages by women and vice versa.
What the OKCupid data showed that everyone forgot is that it didn't matter that the women rated men lower, because it didn't affect their actual behavior.
I am not sure what OkCupid data you talk about, but some of their posts also showed that women respond to messages by men much less than vice versa. It was based on actual behavior, not reporting.
The content of mens messages had a lot to do with rate of responses. I don't remember if that was quantized in any way to try to control for it and remove it as a variable, or if it was just some article that showed what the average messages were, but it was bad, and I don't think rate of response is meaningful without factoring that in somehow.
There was a famous OKCupid post showing that men rated women on a bell curve (so the average was 5/10), but women rated men on average much lower.
Everyone took this to mean that women only want to date the top 20% of men. But it actually just meant women give out lower ratings and that's the only thing it meant.
the data comes from speed dating, which is a really good environment to collect this kind of data from. (attractiveness is the biggest determinant of getting a date followed by conversation quality)
People are always attracted to the essence and personal in the first place. What is beauty from the inside is beauty from the is beauty from the outside. But primitive people have approaches to each other as in stores when buying goods.
I found that when we look at the behavior, such as replying to messages on OkCupid or Tinder, selecting a partner in speed dating events, or who people choose to dance with in couples dancing events, women behave differently from men.
The behavior of women generally follows the power law distribution, and the behavior of men follows the normal distribution.
This means that women rate the attractiveness of men based on something that is based on preferential attachment, for example how women believe others rate the attractiveness of the man, whereas men rate the attractiveness of the women on something that is independent of the evaluations of others.
When the pool of potential mates grows, the inequality experienced by most men grows, whereas the inequality experienced by most women stays the same. And the pool s growing due to internet dating.