How Races and Religions Match in Online Dating
blog.okcupid.com
blog.okcupid.com
-Hacking data to reveal the answers to important difficult questions.
-Exploring contentious issues in a neutral & intelligent way.
-Great example of how to attract potential customers by providing valuable information that also gently demonstrates competence related to your service & differentiates your offering from competitors.
I enjoyed reading this, I learnt something, and I thought favourably of the website based on this. No idea how good they are, but that post was a good marketing lesson.
For other religions, such sites aren't nearly as prevalent.
The matches are based on answers users give to questions (both site designed and user designed).
When the blogger says they don't match, it means:
- Person A: I like X
- Person B: I don't like people who like X
So it's not totally unrelated.
Some Hindu men will say they don't.
Of course, ice cream wasn't the point and other people gave better examples.
We need error bars.
Or maybe they are specifically looking for a non-Hindu date. I mean that they might be looking for a uncompromising relation, maybe easier with a person with different roots.
I just copy and pasted the image onto the axis. It shows that white and native American males are more likely to get responses than the other races.
http://www.codexon.com/temp/mrr.png
And I wrote a small analysis here if you can be bothered to take a look.
http://www.codexon.com/posts/okcupid-dating-statistics-shows...
Your results look depressingly plausible, though, so I'm guessing not.
EDIT: Also, look at the response line from black females; they're all green except for the yellow square in response to black males.
Brilliant strategy if so.
However, I think one critical component is missing (from what OKCupid posted, not from your analysis). Who sends the first message? For example, we are seeing a lot of green in the black female column. Does this mean that:
a) Black females reply back a lot?
b) Lots of men reply to black females?
These two results would mean entirely different things. The first means that black females aren't picky, the second that men like black females.
> ". . . I point this out now so that, below, when we claim that Jewish women are easier to get along with than Christians, you don’t blame us, you blame Jesus."
And that's the first really solid evidence that I've seen shared that zodiac signs are utter BS, which makes me doubly grateful to them. Something to point to when someone says it can't be proved otherwise...
http://www.livescience.com/strangenews/your-astronomical-sig...
The article was not quite as statistically sound as it could be, but at least they aren't trying to pass it off as statistical data.
(Personally, I have a feeling that "pretty profile picture" is more influential than "high match %". But of course, it's very hard to measure what one considers physically attractive.)
I wonder, though, (a) if most people search for similar people or for different people, (b) how well does the matching percentage really reflect a "liking probability" (ie, there is always some wishful thinking when designing date profiles, and some of that might be counter-productive, ruling out good matches).
http://en.wikipedia.org/wiki/Ashkenazi_Jews#Specific_disease...
Because this data makes sense, I hesitate to protest. Nevertheless, I'd be interested to see if race is statistically significant versus religion.
So little statistical significance testing.
This tells me much about this service.
(Although saying that, I wonder if this series of theirs is to attract press attention? If so, then it would be nice to see some significance data.)
Free publicity. Great lesson in PR. Thank you.
1. People do not change their minds.
By algorithmically matching based on answers to questions, you must assume that those answers are meaningful. And meaningful means CONSISTENT. And people are NOT consistent. Especially on complex issues; in fact there are whole professions (sales (insurance, real estate, car, etc) comes immediately to mind) where success is based on one's ability to get people to change their minds. (That example is offered as evidence, not proof - but proof probably isnt a complicated exercise)
2. Like Thinkers make Great Daters
Assuming you are even GETTING like thinkers, (which is tenuous at best) where is it written that like thinkers make great companions? You don't want to have absolutely nothing in common, (at least a common language is necessary) but it isn't linear and i would be surprised if there were any correlation let alone one that was non-asymptotic. Simply - more like thoughts does not entail more compatibility.
3. Looks aren't EVERYTHING
Just based on sheer observation - this is not the case. It is unfortunate, and it speaks badly about human nature, but most human beings make judgments based on appearances. And this "appearance crutch" is only exacerbated when it comes to looking for someone to lay with. How many single, male, bleeding heart liberals would take a swing at Sarah Palin if given an iota of an opportunity? Again - offered as evidence, not proof.
In practice, the user base for OkCupid will change. And old users will reanswer questions sometimes too.
"Assuming you are even GETTING like thinkers, (which is tenuous at best) where is it written that like thinkers make great companions?"
That's not how it works. It's not as simple as "answer match questions and we hook you up with people who answered the same way". You can specify which answers your ideal match would give, even if those answers conflict with the answers you yourself give.
"most human beings make judgments based on appearances"
Which is why you upload photos to OkCupid instead of just using match percentages.
Your other point has been addressed above; people don't know what they want. Asking them what they want their "ideal match" to answer is as foolhardy as asking them to describe their "ideal mate" neuron by neuron.
People do not, in fact, know what they want. But they can make educated guesses.
How does one go about making an EDUCATED guess with NO knowledge?
To put it plainly - statistical information assimilated from meaningless data is also meaningless. I don't know what my perfect mate looks like - and no one does. If you accept that premise, you are then forced to accept that any data gathered from some person A about what their perfect mate P is like is MEANINGLESS. So statistical algorithms from any such information yield meaningless data as well; ergo okcupid cannot do what it hopes to do. QED
While I agree that one's guess at a perfect partner might not be 100% accurate, are you sure you have no information on this at all? Here are some OkCupid questions. Would you be equally willing to date someone regardless of how they answered these questions?
-Is homosexuality a sin?
-How important is religion/God in your life?
-Are you happy with your life?
-Is interracial marriage a bad idea?
And I am certain I am not unique. Indeed, half of the enjoyment in getting to know someone is in the mutual effort expended in trying to understand how the other person was formed: thoughts, feelings, desires, etc - and this enjoyment is independent of the actual thoughts, feelings, etc themselves.
As relationships grow, both parties discover the thoughts, feelings, desires, etc. that were perhaps, formed irrationally - and then adjust accordingly. This is called personal discovery - and is evidence enough that most people are too inconsistent and too nearsighted for an approach such as the one okcupid employs.
That said - I wish them the best of luck.
OKCupid does not judge match percentage based on similar answers. For every question on the system it will ask for your answer and your ideal mate's answer. The system knows that some things are important to some people - but maybe not in reverse.
Clearly they don't - otherwise they wouldn't be on okcupid in the first place.
I'm curious why asian females scored worst of all when asian males scored descent.
If you want a more interesting result, take a look at my other comment for a diagram that shows response rates between different races. It illustrates a greater schism that one would come to expect.
It suggests that certain races have good matches with other races, but horrible reply rates. Reply rate may be the best metric okcupid has as to "success rate". It also suggests strong racial biases on the part of its users.
Many people claim they aren't racist when they refuse to date other races. They always say they haven't found the right <insert minority race here> guy/girl.
What I've found from personal experience, is that the bar is set much higher for those minority races.
While it's possible for people to not date anyone of a certain race because they are bigoted against that race, I think that it's more common for people to just not feel a strong physical attraction towards certain races. My theory on this is that people that haven't grown up (at least during their childhood and/or teenage years) with many people of a certain race won't end up finding most of the 'feature' characteristics of that race to be attractive (i.e. if you grow up in a mostly white neighborhood you might not find black women attractive).
I could be completely off-base here. I have no idea how the development of attraction in the brain works (and I'm sure there's probably much research into this area of psychology). But I tend to find that a lot of people (whether they admit to it or not) have a preference for certain races and not others. Sometimes people don't have a preference for their own race. I know Asian women that are not attracted at all to Asian men, but that doesn't stop them from having Asian men as friends and interacting with them. I would hardly think that they are bigoted against Asians (seeing as they are Asian themselves).
> What I've found from personal experience, is that the bar is set much higher for those minority races.
It's not necessarily minority races. There are plenty of people that are not attracted to others of their own race. 'Yellow-fever,' 'jungle-fever,' etc weren't created for 'one-off' instances (i.e. the person accused of having 'the fever' only dated a person of that race once, and doesn't have an affinity to dating persons of that race) of inter-racial dating/mating/marriage.
Depends what you mean by that. Certain physical traits are correlated to race. To the extent that these traits matter to someone, one would need to set the bar higher on other traits to offset this. I hesitate to give examples since that could anger some people...
Differences on the race graph want to tell us something. Not necessarily something relevant, but something, anyway.
On the other hand I am glad I'm not Aquarius because statistically speaking they are insignificantly less compatible with an Aquarius. His noodelieness has blessed me well!
RAmen!
Combine that with Jewish men who ARE atheists, which is a good number of em, and you've got a super-match. Which all goes to prove what my mother always told me; "I love you."
This kind of statistical kung-fu is well worth a look even if your own interests do not lie in dating sites.
And it's not like we haven't been here before:
http://news.ycombinator.com/item?id=822782
It's quite interesting.
There is a massive pre-selection problem here (people that frequent dating sites, rather than people in general) and the link to 'hacking' is a very tenuous one.
I've seen Michael Jackson referred to as a hacker here, so I probably shouldn't be too surprised.
This is not a hack, this is some statistical analysis with input data of dubious value, pretty pictures and all, drawing conclusions that are completely off the wall.
"Jews and Agnostics get along better with people"
"Muslims of both sexes and Hindu men get along worse"
"Catholics are more universally liked than Protestants"
Please, if that isn't flamebait I really don't know what is.
Where are the control groups, where are the standard deviations and so on.
Effectively this is a dating advert for programmers.
This is infotainment at it's best, statistical noise at its worst.
That said, "is it hacking or not" isn't really a terribly interesting meta-discussion to have, imho, so I'll try and go in a different direction.
Please, if that isn't flamebait I really don't know what is.
The data appears to support the assertions you are calling flamebait. Provided their algorithm for calculating likeability actually works, then the data does support the assertions[1]. (This is all, of course, within the context of the pre-selection you noted in your post).
[1] Note that the assertions are of the form "Members of Set X are more likeable to members of Set Y than members of Set Z". If the assertions were "Being a member of Set X makes you more likeable to members of Set Y", they wouldn't necessarily be supported by the data.
The way it is presented right now is as if simply the size of OkCupid and their data gathering methods give them the license to make this kind of claim.
It's 'interesting' but not 'rigorous'.
The three I listed above are particularly galling, I really don't think any computing algorithm can give someone license to make the statements listed above (and as an atheist I have no dog in that fight), but without proper methods it's even worse.
Of course it's not rigorous. To some degree people interacting with OKCupid's site are a self-selecting bunch. You really need to take this at face value. This is just a blog post showing some number crunching on their site. There are no sensationalist headings like "Muslim Males the Most Hated Group of People." Keep in mind that this is also not a scientific journal, not peer-reviewed research... nor is it claiming to be.
If you are going to be making sweeping statements about people, even including race and religion then you really should do your homework, or if you're doing it out of curiosity, keep the results to yourself. By presenting the data in a format that looks as though very hard work went in to its creation and by hammering home the reliability of that data you are creating the illusion of something that is scientifically solid when in fact it isn't.
Note that the end of the article is leading directly into the objections that people keep making along the lines of "this doesn't mean that these people really get along in real life", and the conflict between what people say they are looking for and the choices they actually make about who to contact and respond to.
The plural of data isn't evidence, even for large amounts of data.
Sure you can do interesting statistical analysis, but the real action is after two people have found each other, and that's where the huge flaw is in all this analysis, the statements are about interactions in real life, the data is gathered online.
Any kind of statement about people being more or less likable would have to be taken out of the context of the website and into real life, without that statements as listed above are unsupportable.
I still find interesting to know how they interact online.