Search results for okcupid

Flashback Friday.

Emma M.H., Rebecca A., Natalee B., Josh L., Anna M., Jordan G., and an anonymous reader all sent in a link to a new analysis released by OkTrends, this time of members’ profile essays and the likes/interests/hobbies the essays mention, broken down by race/ethnicity and gender. They list items that were statistically unevenly distributed by race/ethnicity, showing up much more in some groups’ profiles than others’; these aren’t necessarily the most common items listed by each group.

White men:

White women:

Christian Rudder, the author of the OkTrends post, points out an interesting trend: rural identifying/mythologizing. White men mention “I’m a country boy,” while for White women, being a “country girl” features prominently, meaning both groups are more likely to use this term than other racial/ethnic groups. The men also mention liking hunting/fishing, while White women include horses/horseback riding, bonfires, and the “midwest,” as well as country music/musicians. Most OkCupid users, according to Rudder, are in large metro areas. Of course, you can live in a city and still go riding or fishing, or these can be things you did before you moved to the city that you still really wish you could do and so remain an important part of your identity; and given current demographics, it’s more likely that a former rural resident would be White than non-White, thus showing up more in Whites’ profiles. But I also suspect that references to the “midwest,” or things associated with romanticized rural life (you know, running around in a beautiful wheat field during a thunderstorm and stuff) are a code for a certain type of masculinity and femininity. Among Whites, hunting/fishing indicates you’re a particular type of “guy’s guy,” while being “a country girl” who likes horses and thunderstorms is, I think, a stand-in for implying you’re down-to-earth, nice, not superficial. Being “country” is thus, a lot of the time, shorthand for being authentic.

Moving on, here’s the image for Black men:

We see more self-description than in White men’s profiles — “I am cool,” “tall, dark, and handsome,” “god-fearing,” “calm,” “laid-back guy.” White men (and to a lesser extent women) seem to focus on what they like, not really what they are like, with only “I’m a country boy” and “I can fix anything” showing up in the analysis.

Black women:

If you combined general references to religion, they would stand out even more. In fact, African American men and women are quite a bit more likely than other groups to mention religion:

Data for Latino men:

Like Black men, they more frequently than White men  mention personality characteristics — “I’m a funny guy,” “respectful,” “I’m a simple guy,” “outgoing and funny,” etc.

Latinas, like Latino men, mention specific dances, not just a love of music or musicians:

Rudder notes that Asian men are the most likely of any group to highlight a specific ethnic/national identity in addition to the more general “Asian” label:

I see that above with Latino men, too — references to being Peruvian, Colombian, Dominican, etc. If I had to take a stab at explaining this, I’d guess it was related to differences in how racial/ethnic categories have been applied to different groups. In the U.S. over time, White ethnic categories (say, being Dutch-American vs. Polish-American) have largely faded into the background, all subsumed under the powerful racial label “White.” Distinctions within that grouping have become largely optional, a neat thing to mention, perhaps, but not very socially meaningful. African Americans have often found themselves in the same situation, but due to much more negative forces. The generally shared experience of slavery, racism, and discrimination, as well as negative stereotypes of anyone perceived as Black, mostly erased ethnic identifications among African Americans. Being Black became a master status, such a socially important racial categorization that even those who wanted to be recognized as from a specific location (South Africa, Jamaica, etc.) often found themselves unable to get others to recognize their ethnic distinction.

The broader “Asian” and “Hispanic” labels emerged more recently in U.S. history, and members of both groups often actively fought to preserve distinctions within them. It wasn’t until the ’60s that a pan-Asian identity really began to emerge, such that being called “Asian” really meant anything to people, as opposed to thinking of themselves as Chinese, Vietnamese, etc. And “Hispanic” refers to ethnicity, not race (most Hispanics identify as White); ethnic identities are generally more flexible than racial categories. Aside from personal attachments, many groups thrown into the labels Asian and Hispanic have seen clear advantages to preserving distinctions based on nationality, believing that, say, being Japanese American would be less negatively stereotyped than being simply “Asian.”

So I wasn’t extremely surprised to see that Latino and Asian men specified identifies within those categories…but look back at the Latina image, and then this one for Asian women:

Nothing. Not one specific identifier for either group stood out. I don’t know what to make of that, and would love to hear your suggestions.

There are also specific breakdowns for Asian Indians and Pacific Islanders on the OkTrends, if you’re interested.

Middle Eastern men (a sort of odd category, but whatever) also specify nationalities, which is to be expected as this is another group that has engaged in active contests about their racial categorization in the U.S. (in particularly, fighting to be considered White, not Asian or Black) and also focus on technical/financial careers or expertise:

Middle Eastern women are the only group who prominently mention something about their physical appearance (“petite”), for whatever that’s worth, and again, no nationalities listed:

Of course, as Anna pointed out when she sent in the link, this data isn’t necessarily about people’s actual likes/interests, it’s about what they present as their likes/interests in the dating marketplace. On a dating website, you’re trying to present a profile of yourself…but one tailored to be attractive to others. She wonders to what degree social stereotypes of your racial group, as well as the group you’re interested in dating (if you have any preference) affects how you would describe your interests. That is, it’s possible that in some cases people highlight interests or hobbies that seem to fit social expectations of what they’ll like doing…or what they think the individuals they want to date will want to do, or want their date to want to do. To interpret these results, as OkTrends does, as straightforward evidence of differences in preferences by race/ethnicity, ignores the important fact that these are interests presented as part of an intentional performance for strangers, and may or may not reflect what we actually spend time doing, learning about, or paying attention to in our daily lives.

Originally posted in September, 2010.

Gwen Sharp is an associate professor of sociology at Nevada State College. You can follow her on Twitter at @gwensharpnv.

In general, married couples are homogamous.  That is, they are more likely than not to match on a whole host of characteristics: age, income, education level, race, religion, immigration history, attractiveness level, and more.

But, does homogamy really translate into compatibility?  Or, do we just think it does?

OKCupid set about to find out.  This is the second of two posts about recent revelations that they, like Facebook, have been doing experiments on users. The last one was a depressing look at the role of attractiveness on the site. This one is about the impact of match ratings.  Yep, they lied to see what would happen.

OkCupid users answer a series of questions and the site then offers a “match rating” between any two users.  The idea is that people with a higher match rating are more homogamous — by some measure not identical to those that sociologists typically use, to be clear — and, therefore, more likely to get along.

The first thing they did was artificially alter the match rating for couples whose true match was only 30%.  Users could read the profile, look at the pictures, reviews answers to questions, and see a match rating.  In other words, they had a lot of information and one summary statistic that might be true or false.

People were slightly more likely to send a message and continue a conversation  if they thought they were a 60% match or better.  This is interesting since all these couples were poorly matched and it shouldn’t have been too difficult to discover that this was so.

1

Rudder’s interpretation of the data is that you can make two people like each other by just telling them that they should.

Or maybe, he considered, their algorithm was just terrible. So, they took couples who matched at the 30, 60, and 90% rating and displayed a random match rating that was wrong two-thirds of the time.  Then, they waited to see how many couples got to exchanging four messages (their measure of a “conversation”).

2

The lower right corner suggests that the ideal situation is to be a good match and know it.  Likewise, if you’re a bad match and you know it things probably won’t get very far. But the difference between actually being a good match and just thinking you are isn’t as big as we might think it would be.  At least, not in the space of four messages.

So, does homogamy really translate into compatibility?  Or, do we just think it does?  Maybe a little of both.

Cross-posted at Pacific Standard.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

In the aftermath of the revelation that Facebook has been manipulating our emotions – the one that prompted Jenny Davis to write a post titled Newsflash: Facebook Has Always Been Manipulating Your Emotions – the folks at OkCupid admitted that they been doing it, too.

I’ll let you debate the ethics. Here’s what Christian Rudder and his team found out about attractiveness. Let me warn you, it’s not pretty.

OkCupid originally gave users the opportunity to rate each other twice: once for personality and once for score.  The two were strikingly correlated.

5

Do better looking people have more fabulous personalities?  No. Here’s a hint: a woman with a personality rating in the 99th percentile whose profile contained no text at all.

6

Perhaps people were judging both looks and personality by looks alone.  They ran a test. Some people got to see a user’s profile picture and the text and others just saw the picture. Their ratings correlated which means, as Rudder put it: “Your picture is worth that fabled thousand words, but your actual words are worth… almost nothing.”

Their second “experiment” involved removing all of the pictures from the site for one full workday.  In response, users said something to the effect of hell no.  Here’s a graph showing the traffic on that day (in red) compared to a normal Tuesday (the dotted line):

1

When they put the pictures back up, the conversations that had started petered out much more aggressively than usual. As Rudder put it:  “It was like we’d turned on the bright lights at the bar at midnight.”  This graph shows that conversations started during the blackout had a shorter life expectancy than conversations normally did.

2

It’s too bad the people are putting such an emphasis on looks, because other data that OkCupid collected suggests that they aren’t as important as we think they are.  This figure shows the odds that a woman reported having a good time with someone she was set up with blind.  The odds are pretty even whether she and the guy are equally good looking, he’s much better looking, or she is.  Rudder says that the results for men are similar.

3

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

Kristie, Dmitriy T.M., and Tiffany L. sent in this post at OkCupid comparing 3.2 million profiles of straight people to those of gays and lesbians. Undermining the persistent stereotype that gays are more sexually promiscuous than straight individuals, OkCupid users, gay and straight, reported the same median number of sex partners (6), and the overall pattern is nearly identical regardless of sexual orientation:

And sexual encounters with someone of the same sex aren’t limited to people who identify as gay. Here are the results from a survey of 252,900 users who identify themselves as straight; about a third have either had at least one same-sex sexual encounter, or would like to:

Straight-identified women were significantly more likely to report a same-sex experience (and that it was pleasurable) or interest than were straight men. Here’s the pie chart for women:

And this is for men:

My guess is a lot of people will attribute that to women “playing” at being bisexual or going through a “stage,” but it seems likely to me that part of what is going on is that men’s gender performance is policed so much more harshly and constantly that men suffer greater consequences for same-sex encounters and have more reason to avoid them and to avoid even thinking of them as a possibility.

Reports of same-sex encounters or interest varied significantly by region. In the map, orange = higher rates, blue = lower (OkCupid doesn’t give any percents to go with the different colors, sorry):

There’s other data on personality profiles and, uh, the number of people who think the earth is larger than the sun (!) at the original post.

Also see our previous post on race, gender, and preferences on OkCupid.

Following up on our most recent re-cap of data analysis from OkCupid, sent in by Sara P. and an Anonymous Reader, in this post I summarize their findings on reported sexual orientation and recorded messaging.

It turns out that a whopping 80% of all users who identify as bisexual message men or women, but not both.

The reasons for this are likely complex, diverse, and not immediately obvious.

Blogger Christian Rudder’s hypothesis:

This suggests that bisexuality is often either a hedge for gay people or a label adopted by straights to appear more sexually adventurous to their (straight) matches. You can actually see these trends in action…

The figure below plots age against the percent of self-identified bisexual men who message both men and women, only women, or only men.  The percent that are bi in practice as well as theory message both men and women drops by about half between the ages of 18 and 54 (from about 20% to about 10%), but men in their 30s and early 40s are much more likely to message only women.  Ticking biological clocks and hopes for a wife and kids perhaps?

The narrowing blue swatch may reflect the possibility that men who once identified as bisexual have come to terms with being plain ol’ gay (but the data isn’t longitudinal, so it may be a cohort thing instead of a life stage thing).

Or perhaps the distribution is the result of an interaction between age and who it’s easy to meet.  Maybe young bisexual guys have an easy time meeting women and turn to the internet to meet men; whereas men in their 30s and beyond find it easy to meet men and so turn to the internet to meet women?

Other ideas?

For women who identify as bisexual, the percentages messaging both men and women, just women, and just men show less of a trend across age.

Overall, however, 75% of women who identify as bisexual are not messaging both men and women.  Rudder suggests that there may be a social desirability factor here; that is, that straight women know that men are into bisexual chicks and, so, they claim to be bi in order to appeal to the dudes.

UPDATE: I recommend reading the comments thread for a great discussion of sexual fluidity, the meaningulness of labels like “bisexuality,” and lots more good ideas for why this data looks like it does.

Also from OK Cupid: the racial politics of dating, what women want, how attractiveness matters, age, gender, and the shape of the dating pool, older women want more sex, and the lies love-seekers tell.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

Flashback Friday.

The number crunchers at OK Cupid recently looked at how age preferences disadvantage older women on the site.  First, the post’s author, Christian Rudder, points out, the distribution of singles is pretty matched by sex at most ages:

But that doesn’t necessarily mean that women and men of the same age are reaching out to one another.

Women at most ages state a preference to date men who are up to eight years older or eight years younger:

But men show a decided preference for younger women, especially as the men get older:And, Rudder notes, men target their messages to women even younger than their stated preference.  In this figure, the greenest areas represent where men are sending more messages and the red areas are where they are sending less:

So, even though men and women are more-or-less proportionately represented on the site, men’s decided preference for younger women makes for many fewer potential dates for older women.

Here’s what the dating pool looks like for 21-year-olds (the blue = men seeking women who are 21; the pink = women seeking men who are 21):

For 25 year olds:

For 30 year olds:

Rudder offers this summary measuring how a person’s desirability changes over time:

He writes:

…we can see that women have more pursuers than men until age 26, but thereafter a man can expect many more potential dates than a woman of the same age. At the graph’s outer edge, at age 48, men are nearly twice as sought-after as women.

Thus opportunities for dating are shaped by the intersection of gender and age to the detriment of women over 26 and men under 26.

Originally posted in 2010.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

We’ve highlighted the really interesting research coming out of the dating site OK Cupid before. It’s great stuff and worth exploring:

All of those posts offer neat lessons about research methods, too. And so does the video below of co-founder Christian Rudder talking about how they’ve collected and used the data. It might be fun to show in research methods classes because it raises some interesting questions like: What are different kinds of social science data? How can/should we manipulate respondents to get it? What does it look like? How can it be used to answer questions? Or, how can we understand the important difference between having the data and doing an interpretation of it? That is, the data-don’t-speak-for-themselves issue.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

Flashback Friday.

Dating site OKCupid did an analysis of 500,000 inquiry messages to determine what keywords correlate most strongly with getting a reply.  It has some great lessons about dating and some counter-stereotypical news about what heterosexual women want from men.

This first graph shows that mentioning someone’s level of attractiveness decreased the likelihood of getting a response (for both men and women), though men were more likely to mention looks.  But general compliments about one’s profile increased the likelihood of getting a response (the middle line is the average number of responses, the green bars signify an increase in the number of responses, and the red bars a decrease):

compliments-chart

A good lesson in operationalization: “pretty” is used in two ways in our culture, so when they made sure to differentiate between pretty (meaning “sort of”) and pretty (meaning “attractive”), you can see clearly the way that commenting on looks decreases the recipients’ interest:
pretty-chart

So, in contrast to stereotypes, many women cannot be flattered into a date (though the figure above includes men and women, I’m assuming most people being called “pretty” are female).

Further, the site found that when men sent messages, female recipients preferred humility to bold self-confidence.  The words below all increased the chances of a woman responding to a man’s inquiry:
efface-chart

Instead of bravado and flattery, women appear to actually like men who take an interest in them.  They respond positively to phrases that indicate that a guy actually read their profile and is interested in the content of their person:
engage-chart

The lesson: Treat a woman (on the OK Cupid dating site) like a human being and she will respond positively.

And to answer the question, “What do women want?”  As my dear friend David Landsberg would say: “Everything!

This post originally appeared in 2009.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.