Hacker News new | past | comments | ask | show | jobs | submit login
Gini coefficient (wikipedia.org)
25 points by sbolt on Feb 4, 2020 | hide | past | favorite | 19 comments



One of the most interesting uses of gini coefficient I've heard of, was counting messages, winks and likes received on a dating site as money or wealth and measuring the gini coefficient for men and women.

Apparently the distribution is pretty fair for women and extremely skewed for men.


I think this is the article you are refering to.

https://medium.com/@worstonlinedater/tinder-experiments-ii-g...


I remember it as being done on OK cupid, but this is a really good link. Thank you!


is there a non-hetronormative analysis available as well?


It's a little odd that the author almost consistently refers to men as "men" but uses the word "femlaes" consistently to refer to women. A take on the experiment is here: https://medium.com/@ericschulman/2-big-problems-with-the-tin...

I feel as though thinking of dating as a kind of economy with transactions likely isn't the best way to find someone, especially when the "training" for such "jobs" in the economy is so one-sided - men are very rarely taught how to properly look after their appearance, but it's a core part of every girl's life, 12 and up (and usually sooner than that). It does not surprise me that those with the training (through receiving positive and negative signals on their appearance) are more picky.

You'd find that a Javascript programmer asked to choose a JS framework would probably make a better decision (and of course, be much more demanding given the options) than a documentation writer or graphic designer on the same project.


How many times each term appears on the page:

women: 13

men: 14

male: 6

female: 6

The language used is equal and every case where "male" or "female" was chosen, it was matched by its inverse in the following statement. Same goes for "men" and "women", I think you're perhaps reading something from the article that isn't there.


The context matters. Look at these sentences:

>The most important data I needed was the percent of men that these females tended to “like.

>The females who responded to my questions could have lied about the percentage of guys they“like”

>I have to assume that in general females find the same men attractive.

>Most females only “like” the most attractive guys.

>According to my last post, the most attractive men will be liked by only approximately 20% of all the females on Tinder.

>According to this analysis a man of average attractiveness can only expect to be liked by slightly less than 1% of females (0.87%).

In sentences where men and women are mentioned together, more often than not, "female" is used instead of "woman".


I agree with you that the study was probably biased, but the take on experiment have even more faults than the experiment itself:

,,Problem #1: Personality Matters''

Of course it matters, it's just far not as important as fitness indicators of the other sex. The study was clear that it's about distribution of swipes, which is a prerequisite of even getting to know the personality of eachother.

,,men are very rarely taught how to properly look after their appearance''

I had this problem in my life, but nowdays the internet has all the information easily accessible and for free, and instagram gives people the tools for imitation/experimentation, so taking care of appearence got much easier for men in the last decade.

,,part of every girl's life, 12 and up (and usually sooner than that)'' My niece is 3 years old, and I see her already thinking about her appearance all the time (as opposed to my nephews who just want to play games).

,, It does not surprise me that those with the training (through receiving positive and negative signals on their appearance) are more picky.''

Evolutionary biology shows that it's much much deeper within the sexes than just training. It's a fact that we have twice as many female ancestors than male, and this fact is true in other species as well.


>The study was clear that it's about distribution of swipes, which is a prerequisite of even getting to know the personality of eachother.

The author notes that the content of the picture itself is highly relevant to judging personality from a picture, and how we perceive people in pictures. Painters, for instance, have known to use this fact for the past two thousand years when portraying people - often, the context the subject is placed in is just as important, if not moreso, than the appearance of the person themselves.

>so taking care of appearence got much easier for men in the last decade.

Being easy does not mean that it is something that they (in general) follow through with. It is now more easy than ever to watch films, but that does not mean I will dedicate considerable time to learning to appreciate films, or learn what differentiates a good film from a bad one.

>Evolutionary biology shows that it's much much deeper within the sexes than just training.

I agree, but this just reinforces my point that it is a fallacy to expect a "market" that is so heavily inbalanced in this skill to be anything near equal. Men have to put a lot more effort into the first picture they show, since there is a large body of research showing that women are less stimulated by visual representations than men.

Many other variables were ignored in the experiment, all of which directly and importantly influence the charactersitics of the "Tinder economy"; profiles are assigned an ELO on Tinder, such that those who swipe more are pushed down the stack, using the boost/super like functionality dramatically increases the chance of getting a match, women and men both select on the contents of the bio, or whether it has an Instagram link, and how many followers are on the user's page, Tinder shows the mutual connections within social groups, etc. When I'm on Tinder, at least, I scroll through all the pictures, and I don't just look at the first one. It's unclear whether the author even used multiple pictures, forcing people to make decisions based on just the one - decisions they may not have made if there were multiple.

Not to mention the extremely small sample size, assuming that women all rate men (and indeed they rate them along the same axes as men rate women, even given just a picture), the use of deception by the author, and the bizarre comparison to real-world economies whose inequality is shaped by a variety of factors, from ownership of capital, unemployment, gender, race (the author even put South Africa on the graph!), etc. "Attractiveness" is left undefined qualitatively (and therefore meaningless to assign any value to its quantiative measurement, as if such a thing were objective, even on Tinder), and it is unclear whether this includes the user's bio or not.

A like, on its own, on Tinder doesn't get you anywhere. It opens the potential (and a low potential) to find something (usually sex, in which it is no surprise that people selected based on atractiveness and extrapolate on that atractiveness from personal biases). The currency of Tinder, if anything, sohuld be modelled as "successful" meets.


The author notes that the content of the picture itself is highly relevant to judging personality from a picture, and how we perceive people in pictures. Painters, for instance, have known to use this fact for the past two thousand years when portraying people - often, the context the subject is placed in is just as important, if not moreso, than the appearance of the person themselves.

When we are talking about the gini coefficient for liking eachother, it doesn't matter why it happens or not happens.

How many percentage of the people who you see on Tinder do you swipe right on?


>When we are talking about the gini coefficient for liking eachother, it doesn't matter why it happens or not happens.

You said that swiping is a prerequisite for judging personality. My only point was that it's not entirely true, since most people believe you can tell a lot about someone's personality from the pictures they've uploaded.


"Men would have an easier time dating if they behaved more like women" is such nonsense. If men could increase their mating prospects by obsessively cultivating their appearence, they would do it. But it doesn't, so they don't.

If you want to know "what women desire in men," look at things that men do that women don't. And, by the same token, if you want to know "what men desire in women," look at things that women do that men don't.

There's a general rule here: people, in aggregate, are better at solving these kinds of problems ("how to attract the opposite sex") than your mind is.


> If men could increase their mating prospects by obsessively cultivating their appearence, they would do it.

Tinder isn't about "mating", it's about hook-ups (primarily). I would say that the average guy assigns much less value to getting a hook-up than a "mate", and like all human optimiztions, it is a balancing act. It may be important, but not $1000 worth of beauty products and time a year level of importance.

>If you want to know "what women desire in men," look at things that men do that women don't.

This is obviously fallacious. Women and men often both desire very smiliar things in a relationship, known as the ability to reciprocate. Whether that's emotional support, working a job, or giving oral sex.


> Tinder isn't about "mating", it's about hook-ups (primarily)

You can separate sex from mating in your conscious mind but most of your mind is not conscious.

Your view of human behavior is overly simplistic/idealized. We don't choose what we find attractive and most of what motivates us are chemical systems that are beyond our understanding. You can consciously posit that sex is "something pleasurable that people do to feel good" but that's how not how you, or anyone, really feels about sex if you go even an inch below the surface.

> This is obviously fallacious. Women and men often both desire very smiliar things in a relationship, known as the ability to reciprocate. Whether that's emotional support, working a job, or giving oral sex.

"The ability to reciprocate" is a pretty broad category. What men find attractive in women is very different from what women find attractive in men. If you want to learn about what women find attractive in men, there's a billion dollar industry that caters to women's fantasies. Read Twilight, 50 Shades of Grey, or nearly any romance novel.


>You can separate sex from mating in your conscious mind but most of your mind is not conscious.

It is one thing to say that we are driven to sex because of the unconscious desire to reproduce; it is another thing to say that, even unconsciously, a woman would like her Tinder hook-up to impregnate her, and further that it's someone she'd raise a child with (and vice versa, obviously). The truly simplistic model will only look at evolutionary psychology and neglect social psychology. It also seems that asexuals who masturbate can separate sex from mating, as can homosexuals, infertile people, pedophiles, etc.

> You can consciously posit that sex is "something pleasurable that people do to feel good" but that's how not how you, or anyone, really feels about sex if you go even an inch below the surface.

People in developed countries are having more casual sex than ever, and casual sex inside marriage. A large portion of women take the pill, and a large number of men wear condoms. You may have heard of a whole app dedicated to this culture: Tinder. As it turns out, below the surface people do want to feel good. The fact that this desire or that desire is really about wanting a child is irrelevant, because that's not how we practice the act. Look under the surface all you want - it does not change the fact that the surface has a particular expression in concrete terms in the way we navigate the sexual world.

>there's a billion dollar industry that caters to women's fantasies. Read Twilight, 50 Shades of Grey, or nearly any romance novel.

All this goes to show is that both sexes, in general, find dominance sexy. Sure, you could say that a men generally do not want women to be dominant, and women want men to be dominant, but that only goes to show that it is just as much a difference in desires as it is the same desire reciprocally exercised.


Here's an online calculator[1] for Gini coefficient that I found some time ago. Tinkering around with some numbers there was what finally gave me an proper intuitive understanding of the Gini coefficient and the Lorentz curve.

[1]: http://shlegeris.com/gini


I think this is a bad way to gauge an economy.

If you want to look at why people are poor, I would look towards corruption rather than merely inequality. Or maybe people in a country are poor on paper but can afford a hundred dollars of luxuries a week.

Using gini reminds me of GIGO. You used a poor metric for information, your decisions are going to be poor.


Could you please stop creating accounts for every few comments you post? We ban accounts that do that. This is in the site guidelines: https://news.ycombinator.com/newsguidelines.html.

HN is a community and we want it to remain one. For that, users need some identity for others to relate to. Otherwise we may as well have no usernames and no community, and that would be a different kind of forum. https://hn.algolia.com/?sort=byDate&dateRange=all&type=comme...

You needn't use your real name, of course.


My mind went straight to gini impurity




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: