OkCupid’s Unblushing Analyst of Attraction
nytimes.com
nytimes.com
This piece almost makes it sound as if Rudder has been blogging based on OkCupid results this whole time... and if you go to the OkTrends site, you see huge inline placement for Rudder's new book.
Coincided with their acquisition by match.com. Even though they said it didn't have anything to do with that (just like removing their critique about sites like match.com was voluntary).
Well, I hope they get back to writing those posts.
You people are ridiculous.
EDIT: Might as well link to it while I am at it - http://machinesentience.com/why_you_should_never_pay_for_onl...
Also, I know the explanation they gave. I don't give a damn whether anyone was pressured into taking it down and they never bothered to actually explain what was wrong with the original analysis, identify the 'public sources' used, or to do anything but try to bury the story with a cheap excuse and hope we'd all forget.
Example of such an explanation: http://www.reddit.com/r/OkCupid/comments/pr2wo/dyk_that_okcu...
> Basically the breakdown was that rather than 93%+ of the people you are shown being "dead profiles" the actual number is more like 10-20%.
> But if you take these into consideration, our conclusion would say, "Match.com users get married at about the same rate or higher as people not using Match.com!"
That's not really a ringing endorsement here, at best it shows that if we change our estimates a little we can reverse our conclusions. That to me says that they were never able to come to very reliable conclusions in the first place on this issue.
And if you read the entire comment thread, you'll see that they removed the article precisely because they were unable to come to any real conclusions.
The original blog post stated that paid dating sites made one's odds worse. Real numbers demonstrated that paid sites were, at worst, a wash.
The thing is, access to private data didn't change the fact that they were using a self-reported stat to begin with. Given that OKC's data was self-reported to a significant degree, it's hard to see why this wasn't taken into account.
And then they go on to cite how they already knew it could be 10x higher (or more) by citing OKC data. This leads to the (paraphrased) conclusion that they can't be total frauds because they're still in business.
(Source: I used to work at OkCupid Labs, one of the things that Sam went on to do under Match, where he is now CEO.)
Edit: from other comments: http://machinesentience.com/why_you_should_never_pay_for_onl...
http://static.izs.me/why-you-should-never-pay-for-online-dat...
You're not the first person to propose this question - there's a comment like this almost every time OkCupid makes the front page - and here's what I wrote the last time[0], which I think explains it well:
There were a number of factors. A bit part is that, in 2010, there were 2.5 people working full-time[1] on doing research for OkTrends, which allowed us to research, write, and publish posts much more often.
The blog posts took a lot of work. "The Real Stuff White People Like"[2] took almost two months of my time, plus some from Max and Christian as well. (Much like the product design process, since we didn't start each post off with a clear end result in mind, not all the work was visible in the final product).
I left to go back to school. Max ended up taking on more responsibility for other data/stats work, which slowed the pace a bit, and he left at the beginning of 2012 to do his own stuff. And Christian became in charge of running OkCupid after the acquisition, which meant he had even less time then he did before Max and I joined.
People asked me for the last three years whether the reason OkTrends hadn't posted since 2011 was because of the Match.com acquisition and whether Match shut them down and I had to tell everyone "No, trust me, they're still around! It's just a coincidence!". Thankfully I no longer have to. :)
[0] https://news.ycombinator.com/item?id=8097780
[1] 2.5 full-time means: Two of us full-time, as well as Christian, though he split his work time between OkTrends (the blog) and other stuff.
[2] http://blog.okcupid.com/index.php/the-real-stuff-white-peopl...
I would suggest that in the interests of the OKCupid business, updating the blog would have been a better thing to do than write a book. The blog drew many users to OK Cupid. It's possible that writing a book was in Christian's interest.
If Christian was no longer working at OK Cupid while he wrote the book, then I understand. In that case the error is on the part of the New York Times for describing him as currently in that role.
After getting acquired I think the general need for that type of PR was less (in addition to the manpower issues you mentioned).
On the whole, I also think usage of willfully submitted and gathered user data for things other than advertising is amazing. Too many companies have invested too many resources into data analytics purely for increasing revenue from advertising.
I think if companies can find ways to gain better social understanding and from that provide better value to end users (rather than to advertisers as in the advertising model) in a way that gets them revenue, it will be amazing. I am all for social media companies doing research from their user data, in the hope that we can move on from the ad-driven web.
Those people being match.com - the same people that operate the 'tinder' app.
Much like how the company that sells the eco-friendly smoothies is probably Coca Cola.
Ultimately match.com want to own online dating. Since there is always going to be a niche for free dating sites (as well as the subscription model), they have to have the best 'free'/freemium dating site going. The marketing/PR consists of 'insights' rather than TV adverts or social networking thingies. The site is just one part of their strategy to own online dating. People on the site might not pay, however those adverts probably bring in a lot of revenue - see Plenty of Fish story.
Returning to the eco-friendly smoothies - there are no saints at the helm of the company determined to make a better world, they are just doing what is needed to bring in the revenue for the parent company and making sure nobody else is in that space. Another analogy - big software company making sure their software is reasonably easy to pirate so people don't use any other software, free or paid for.
Clearly OKCupid was more valuable to match.com than anyone else hence they bought it. However the idea is no more respect worthy than the operations of the eco-friendly smoothie company owned by a fizzy drinks giant mega corporation.
I still have respect for whoever came up with the eco-friendly smoothies and was able to pose enough of a threat to Coca Cola, but instead of getting crushed by them is able to harness the giant's resources to make more eco-friendly smoothies. It's a lot better than folding under the evil.
There are no saints at the helm of any for-profit organizations, or they'd get destroyed. I'm talking within that (very realistic and tolerable) space, they do some creative things [as a past okcupid user and as someone interested in their data] that provide interesting value to the customer that are hard to find elsewhere.
Couldn't count on two hands, even a third if I had one the number of different answers I've gotten, and not a single one of them-no matter the rationality or validity behind the history of it, has yet to answer why we still use it.
That said, "Person of Color" seems to be the new trendy thing and that baffles me even more (well, to be succinctly snarky about it: other people's reactions to that phrase are what baffle me. That's an entirely different conundrum not fit for HackerNews, though.)
Here in Australia we're getting infected with this compound term, to reassert that heritage doesn't mean non-citizen. It's really annoying, because I grew up in a suburb of Greeks and Italians, who were also Australians... and now (from some corners) it's Greek-Australian and Italian-Australian. It wasn't perfect - the stereotypical Australian is still seen as a WASP - but this compound noun system isn't the answer to fixing the problem.
There's a lot of bluster that goes on among certain "progressive" crowds that ultimately sums the solutions up to policing of rhetoric; and while I have to admit they're somewhat onto something with the notions that "language informs impressions", a lot of their approaches seem deeply misguided if only because there's a vacuum of nuance.
I personally don't have the answers to how to remediate the baggage that comes with cultural nomenclature... I do however believe rather sternly, that demanding a change of vernacular is not a silver bullet.
In the context of a statistical analysis, like the OkCupid analysis, it's completely correct to segment people by ethnicity and by nationality separately, and it makes a lot more sense in that context to say "Americans of African ethnicity." "African Americans" is an equivocal phrase in an analytical context because people use it to refer to ethnicity, when it actually refers to both nationality and ethnicity.
You would not believe the amount of fire I got responding to the writings of another black writer who touched this topic: Ta-Nehisi Coates when I expressed the exact same point.
I think the troubling aspect is the disconnect in the area you highlighted: necessity in context. Sadly there is so much demagogy taking place in the larger conversation of "race" that not even natural sciences like the study of statistics are safe from mindsets that want to rule conversation with politics over pragmatism.
Nomenclature exists to easily quantify and reference, not necssiarily objectify.
It does however make assumptions about ethnicity. How about just saying what you mean and using "black"
I have a friend who is ethnically African, and he's as white as it gets (ancestors from South Africa).
I also have a friend who is black but not ethnically African (ancestors from Trinidad)
I kid, and kind of agree... I sometimes see Asian-American, but not as much as simply "Asian", and really don't see it from any other culture the same as "African-American" ... Frankly, I really don't see the point in labeling as it just persists stereotypes.
Oh man, I lost it. I laughed so hard and she just didn't get it.
We're not American... Neither was he...
>Mr. Rudder is particularly interested in the divide between the mates people claim they want and their actual online pursuits. Witness the actions of 35-year-old heterosexual men on OkCupid. These men typically search for women between the ages of 24 and 40, Mr. Rudder reports, yet in practice they rarely contact anyone over 29.
>“I see this as a statement of what men imagine they’re supposed to desire,” he writes in the book, “versus what they actually do.”
For a guy talking up all the awesome science he thinks he can do, this is rather unscientific. I'm sure he realizes there are other explanations for this trend (older women are more likely to be divorced or have kids, for example), but by promoting his own narrative he muddies the waters. For what it's worth I don't think his interpretation of the data is correct in this case.
A lot of other stuff on the OKCupid blog, e.g. claims of racism because black women receive fewer first messages than women of other races, fell into the same trap. The blog was very interesting to read, to be sure, but I couldn't take it seriously because there was so much pseudoscience and just-so stories, even if it was backed up with mountains of data and nifty charts.
I don't see how that was faulty.
It's pretty clear that Black women are involved in interracial marriage at a significantly lower rate than women of other races, or even black men. Then if you look at response rate by race (a previous OKTrends post), black women generally have a lower response rate from men of all races compared to women of other races. Data supports the idea that black women are generally not favored.
See Wikipedia.
"Racism consists of both prejudice and discrimination based in social perceptions of biological differences between peoples" ... "In sociology and psychology, some definitions only include consciously malignant forms of discrimination.[4][5] While some consider any assumption that a person's behavior is tied to their racial categorization is inherently racist, regardless of whether the action is intentionally harmful or pejorative, because stereotyping necessarily subordinates individual identity to group identity"
So if you just prefer one group, that's not racist, as far as I can tell. If you make claims about a group's behavior based on race, that's racist.
> "We're not racists," Chambers told CNN Monday. "We just want to be with white people. If that's a crime, then I don't know. It's all right to be black and Latino and proud, but you can't be white and proud. I don't understand it."
http://inamerica.blogs.cnn.com/2012/06/12/a-kinder-gentler-k...
I don't know. If someone told me "I just want to be with white people", I would think they were racist. That said, the word racism definitely has multiple and nuanced meanings at this point, so it could be more a part of my interpretation than the more established official definition.
This is like saying that because there is a KKK that all white people are racist.
No it doesn't. Many stereotypes are useful when you have very little information about a person. You're only subordinating individual identity to group identity if you refuse to update your beliefs about a person after receiving new evidence about him.
Thank you. It's so easy to brush under the rug peoples racial preferences as if they are sacrosanct
As far as slippery slopes go, Facebook crossed that line quite a while ago...
For example: > As a group, for instance, Latino men rated Latinas as 13 percent more attractive than the average for the site, while they rated African-American women 25 percent less attractive.
That is an insane generalization going on.
What metric are we using to determine this? Is it possible that people who tend to participate in the rating of the looks of potential mates are more inclined to align with race? Not purely Latino men. Which is an enormous initial generalization to be making at the outset.
> Witness the actions of 35-year-old heterosexual men on OkCupid. These men typically search for women between the ages of 24 and 40, Mr. Rudder reports, yet in practice they rarely contact anyone over 29.
Again on this one, was age the only possible metric that caused the under 30 to be contacted more by 35 year old men? There isn't anything else that might be different about an under 30 profile that causes more communication to occur?
This is the actual scary stuff to be publicly releasing as real science. Just as the general public is ill informed about the experiments going on, they are also not aware of what metrics are used to determine these results. In my experience, many of these metrics are not as concrete as the appear and full of pushing the data to fit a narrative etc.
Combine that with the fact that most of this data is proprietary and private with no way to be peer reviewed. Dangerous stuff.
yes, it can be shocking to be presented with plainly spoken, cold hard facts when you grow up in and are surrounded by a sterile bubble of non-fact-based discourse. especially in a topic as taboo as dating, sex, race, and marriage. this is called cognitive dissonance.
trust me, as an asian american male, none of the cold facts of reality were ever hidden from me behind a facade of soothing lies. when i was much younger, i couldn't get laid, and i had to face reality head-on, and seriously work on myself to get ahead of the curve during my 20s.
> In my experience, many of these metrics are not as concrete as the appear and full of pushing the data to fit a narrative etc.
in my experience the okcupid blog's conclusions fit the narrative of reality when it comes to sexual market value, race, and attraction, whereas you're espousing a more politically correct kind of conclusion that fits an artificial narrative of egalitarian bullshit that in no way, shape, or form applies to the modern sexual marketplace (online dating and the hookup scene). there are winners, and there are losers, period. if you're not getting laid, you are losing. there is no long courtship period anymore. chivalry is dead.
how is it so radical and "dangerous" to conclude that men generally prefer younger, more fertile women, or that latino men generally prefer latino women? i mean are you even being serious?
Grow up and take some responsibility for the world around you.
what you're doing is called "projection" - you're taking your own thoughts, and assuming others have the same ones.
ironically, i bet you're a white guy - because all of the black, latino, asian guys i know talk about this stuff openly. in general, my white friends find this kind of talk extremely unsettling and uncomfortable.
Maybe you should go back and read what I wrote one more time. I said that "people in the general public", not "I think that".
1. I know this is pseudo science so I don't read anything into it. 2. Plenty of people have no experience with data science and do not know the difference between causation and correlation.
Can you do me a favor, and stop projecting your issues with political correctness onto me? I'm sorry you've had your difficulties with it in the past, but my comments have nothing at all to do with the topics making me uncomfortable. It's because as a society as a whole, there are many people that will use this type of information to spread hate. If okCupid or anyone else is going to do it, they need to release all of their data publicly at the same time so that there can be a complete and two sided discussion with all the data available.
Lastly...
> because all of the black, latino, asian guys i know talk about this stuff openly. http://www.urbandictionary.com/define.php?term=I%20Have%20Bl...
/sigh