Studies

Kristen Bell is universally thought of as good-looking.

Megan Fox is frequently considered super hot, but is also regarded as unattractive by many.

And how do regular OkCupid users stack up against such famous competition?

To explore how mathematics can determine a man's reaction to a woman's appearance, the OkCupid research team found users willing to submit their photographs to scrutiny under the site's 5 star rating system. Woman A had an attractiveness rating of 3.4/5 stars, and the number of messages she receives per month is 0.8 times the site's average. Woman B had an attractiveness rating of 3.3, and the number of messages she receives per month is 2.3 times the site's average. Though their star rating is very similar, the number of responses these women receive is very different. The pattern of votes that determined their attractiveness rating is also very different: for Woman A, there was a clear consensus, while for Woman B, there was a split decision. In simple terms:

• Girls who are described as "cute" tend to be ignored by men.
• The more men as a group disagree about a woman's level of attractiveness, the more they end up liking her.
• Having some men think she is "ugly" can actually work in a woman's favor.

These are three of the surprising conclusions drawn from the latest OkTrend's study, "The Mathematics of Beauty." This time, the researchers at OkCupid focused on female attractiveness by looking beyond a woman's profile photo and "into the reaction she creates in the reptile mind of the human male." As always, the data used was gathered from the activity of actual OkCupid users, in this case 1.54 million votes, 596,000 messages, and 64,000 profiles.

It should come as no surprise that the hottest OkCupid members receive more page views and messages than less good-looking users, and it probably also isn't a surprise that men dedicate more time to the pursuit of attractiveness than women do. A beautiful woman on OkCupid receives approximately 4 times more messages than an average woman, and 25 times more messages than an unattractive woman. The OkCupid team examined a sample of 5,000 women, and sorted them based on attractiveness and the number of messages they received during the last month.

The graph, which can be found here in the original article, was adjusted to account for differences in factors like "race, location, age, profile completeness, login activity and so on," so that the only meaningful difference between the subjects plotted on the chart was their physical attractiveness. The graph revealed an extremely wide range of results, showing women with the same level of attractiveness receiving a vastly different number of messages per month.

Armed with the data, the OkCupid team set out to determine what caused the disparities they had found...and the key, it turns out, is in mathematics.

If we rate attractiveness using the classic 10-point scale, a woman who is rated a 7 could be a 7 because everyone who sees her considers her to be one. On the other hand, she could be a 7 because the majority of people consider her a 10, and a few think she is a 0. If all we know about a woman is that she's a 7, we don't know how that number was determined. Writer Christian Rudder uses actress Kristen Bell to illustrate the point: while she is certainly good-looking, she is not out-of-this-world, supermodel hot. Most people would probably rate her as "very attractive," while a smaller number of people will likely consider her "super hot" and few would say she is "unattractive." In the case of Megan Fox, however, many people are likely to rate her attractiveness as extremely high - higher than Kristen Bell's - but a significant number of people will also probably rate her attractiveness as low (lower than Kristen Bell's).

What happens when this theory is applied to real OkCupid users?

We'll find out next time.

If asked to describe your personality, what would you say? Do you consider yourself adventurous, dorky, and a little bit kinky? Are you introverted, artsy, and into politics?

Perhaps you're compassionate but also aggressive, or ambitious but also polite.

These characteristics are several of the 20+ personality traits examined by Christian Rudder in the OkTrends blog's latest study: "Gay Sex vs. Straight Sex." Using the match questions answered by OkCupid members, the site collected more than 669 million responses and created a straight/gay comparison chart for personality categories like "Confident," "Literary," Religious," and "Romantic."

Straight men, as it turns out, live up to the clichéd image of the "bullying jock." Based on their answers, they were rated more aggressive, more confident, more violent, more horny, more competitive, and more into sports than gay men. Their hyper-masculinity was tempered, however, by being rated more dorky, more religious, and more romantic than their gay counterparts.

Stereotypes appeared on the opposite side of the t-chart as well: gay men were ranked higher on personality traits that are often considered feminine, like "Compassionate," "Trusting," "Literary," "Artsy," and "Generous," though they are apparently also more ambitious, spontaneous, and political.

Gay women scored higher then straight women on almost every characteristic, with particularly strong leads in the categories of "Adventurous," "Artsy," "Into Drugs," and "Kinky." The only personality traits that were stronger in straight women were "Ambitious," "Into Sports" (at least the women managed to defy a stereotype!), "Optimistic," "Polite," and "Religious."

Rudder's next move is a return to OkCupid's "The REAL 'Stuff White People Like'" study (my reviews of the study can be found here, here, and here), this time using members' descriptions of themselves to measure traits based on gender and sexuality. The words and phrases that appeared most frequently in gay men's profiles were: "The Devil Wears Prada," "Britney Spears," "Mean Girls," "Kelly Clarkson," and "The Color Purple." With only a few exceptions, the terms were all related to films, books, theatre, and music. Gay women overwhelmingly placed "The L Word" in the top spot (so much so that Rudder had to shrink the size of the text to fit the site's template), followed by "Tegan and Sara," "Ani DiFranco," "Piercings," and, again in the 5^th spot, "The Color Purple."

Straight men define themselves by traditionally masculine interests, like cars, war movies, sports, beer, and fixing things. Straight women consider their girlfriends very important and are searching for Mr. Right, but are also determined to prove that they are "Independent" and "Self-Sufficient." They also enjoy make up, romantic movies and books, and careers in the medical field.

Rudder's final analysis of the data measures "the frequencies of all one-, two-, and three-word phrases against the site-wide rates." According to his chart, gay users are significantly less "mainstream" than straight users, with bi users predictably falling in between. "In fact," Rudder notes, "what these numbers are saying is that the average gay person has only about 30% in common with the average American." Before anyone writes an angry letter to OkCupid about discrimination, however, consider Rudder's closing comments: "Adding an ethnic group to the plot helps put it in perspective." When Indian users are added to the graph, they are ranked even less "mainstream" than gay members.

So what can we take away from this?

Defining "normality" is an impossible task, and "no one's saying that Indians shouldn't get married or shouldn't be allowed to adopt children....We hope gay people can expect the same treatment very soon."

Have you ever had a sexual encounter with someone of the same sex?

The 252,900 straight people OkCupid polled in their newest study, "Gay Sex vs. Straight Sex," provided surprising results: nearly a quarter responded that they had participated in a same-sex experience, and an additional 11% reported that they had not but would like to.

Writer Christian Rudder then broke down the replies to the inquiry geographically, creating a color-coded map of the "straight people who either have had or would like to have a same-sex experience in the continental U.S. and lower Canada." Most of the results are predictable - the West Coast and the North East are more gay curious, while the Southern states are significantly less so - but some are unexpected. Even in the states that reported being less interested in exploring same-sex experiences, pockets of gay curiosity, in cities like Austin and Madison, can be found. And perhaps most interestingly of all, Canada is entirely orange and red, the colors that indicate the highest level of gay curiosity, from coast to coast.

In a brief aside, Rudder discusses the "many awful things our elected officials have said about gay people" that he encountered while writing his post, mentioning in particular a quote from Jim Demint: "If a person is a practicing homosexual, they should not be teaching in our schools." Rudder facetiously notes that "a fun game to play with stuff like this is to replace the words 'homosexual' and 'gay' with 'politician,'" because "then you have something that's actually true," and then proceeds to attempt to discredit Demint's claim by searching through the answers to match questions. What he finds is "one question with a surprising disparity, not between orientations, but between genders."

The question is straightforward - "Which is bigger: the earth or the sun?" You'd think that anyone who passed the 4^th grade would be able to answer this simple astronomy query with ease, but it appears that was not the case - over 10% of straight women incorrectly answered that the earth is larger than the sun, and just under 10% of gay women answered similarly. The men fared much better: only 5% of gay men, and fewer than 5% of straight men, got the question wrong.

Rudder's results can be summed up in one word: Yikes. Perhaps we'd all be better off abandoning our academic pursuits in favor of moving to Canada and exploring our same-sex inclinations.

Stick around for a review of the final findings of OkCupid's "Gay Sex vs. Straight Sex" study, in which Rudder evaluates the personality traits that are most associated with each orientation, revisits the fun of the "The REAL 'Stuff White People Like'" study, and attempts to define "the norm."

The results of OkCupid's latest study are in, and the findings are bound to amuse some, offend others, and leave everyone else wondering why Canada is so intrigued by same-sex sex.

Blogger Christian Rudder explains the reasoning for collecting the latest research, in a study called "Gay Sex vs. Straight Sex," with this intro: "Gay issues have been in the news a lot lately, from the debate over same-sex marriage in Congress to a sickening rash of gay-bashing here in New York City. We see a lot of emotion out there, instead of information, and we wanted to provide some data-based context on sexuality so that people might make better choices about what they say, think, and do."

It's certainly a noble goal, but did they achieve it?

The initial conclusion the researchers draw is this: "First of all, gay sexuality is not a threat." Wording - which is bound to ruffle a few feathers - aside, the study makes a valid point: "The subtext to a lot of homophobic thinking is the idea that gays will try to get straight people into bed at the first opportunity, or that gays are looking to 'convert' straights." After reviewing more than 4 million match searches, the OkCupid research team found almost no evidence to back up the stereotype. Only 0.6% of gay men had ever searched for straight matches, only 0.1% of lesbians had ever searched for straight matches, and only 0.13% of straight people's profile visitors were gay. There were no gay users, male or female, who predominantly searched for straight members.

The researchers' next task was to debunk another virulent, but persistent, stereotype: that gay people are more promiscuous than straight people. "The statistical reality," Rudder writes, "is that gay people as a group aren't any more slutty than straights." The median reported sex partners for straight men, gay men, straight women, and gay women was exactly the same: just 6 people. A little under half of gay people (45%) and straight people (44%) had had 5 or fewer partners, while 98% of gay people and 99% of straight people had had 20 or fewer partners. The myth of gay promiscuity turned out to be just that - a particularly stubborn myth that was not supported by the research. In fact, the OkCupid team found that "just 2% of gay people have had 23% of the total reported gay sex," a finding that Rudder eloquently describes as "pretty crazy."

The study's next conclusion is certain to come as a surprise to many: "Straight people have gay sex, too." Almost a quarter of the 252,900 straight people asked "Have you ever had a sexual encounter with someone of the same sex?" answered "Yes." Rudder breaks the responses down into a pie chart: 66% responded "No, and I would never," 11% responded "No, but I would like to," 6% responded "Yes, and I didn't enjoy myself," and 17% responded "Yes, and I enjoyed myself."

Rudder further breaks the results down into charts for men and for women. Predictably, women expressed more interest in exploring their same-sex desires than men did. 1 in 3 straight women reported intimacy with another woman, and of those who hadn't hooked up with women, more than 1 in 4 stated that they would like to. As for the men, 13% responded that they had had a same-sex experience, and an additional 5% said that they hadn't yet, but would like to.

Stay tuned for Part II, in which Rudder takes on politicians, teaches an important astronomy lesson, and examines Canada's homosexual curiosity.

David over at the Online Dating Insider found a study of online dating called "Scripts for Online Dating: A Model and Theory of Online Romantic Relationship Initiation". This 298 page paper was authored by Bridget L. Long. She is a student at Bowling Green State University and the paper was one of her requirements for the degree of Doctor of Philosophy. I have only skimmed the paper at the moment but the study examined the online dating processes from both the standpoints of the services themselves, and members. Bridget used the dating sites Match.com, eHarmony and Plenty of Fish in her research.

To download the study, visit the OhioLink ETD page.