January 29, 2008
Hitting It Off, Thanks to Algorithms of Love
By JOHN TIERNEY
PASADENA, Calif. — The two students in Southern California had just been introduced during an experiment to test their “interpersonal chemistry.” The man, a graduate student, dutifully asked the undergraduate woman what her major was.
“Spanish and sociology,” she said.
“Interesting,” he said. ‘‘I was a sociology major. What are you going to do with that?”
“You are just full of questions.”
“My passion has always been Spanish, the language, the culture. I love traveling and knowing new cultures and places.”
Bogart and Bacall it was not. But Gian Gonzaga, a social psychologist, could see possibilities for this couple as he watched their recorded chat on a television screen.
They were nodding and smiling in unison, and the woman stroked her hair and briefly licked her lips — positive signs of chemistry that would be duly recorded in this experiment at the new eHarmony Labs here. By comparing these results with the couple’s answers to hundreds of other questions, the researchers hoped to draw closer to a new and extremely lucrative grail — making the right match.
Once upon a time, finding a mate was considered too important to be entrusted to people under the influence of raging hormones. Their parents, sometimes assisted by astrologers and matchmakers, supervised courtship until customs changed in the West because of what was called the Romeo and Juliet revolution. Grown-ups, leave the kids alone.
But now some social scientists have rediscovered the appeal of adult supervision — provided the adults have doctorates and vast caches of psychometric data. Online matchmaking has become a boom industry as rival scientists test their algorithms for finding love.
The leading yenta is eHarmony, which pioneered the don’t-try-this-yourself approach eight years ago by refusing to let its online customers browse for their own dates. It requires them to answer a 258-question personality test and then picks potential partners. The company estimates, based on a national Harris survey it commissioned, that its matchmaking was responsible for about 2 percent of the marriages in America last year, nearly 120 weddings a day.
Another company, Perfectmatch.com, is using an algorithm designed by Pepper Schwartz, a sociologist at the University of Washington at Seattle. Match.com, which became the largest online dating service by letting people find their own partners, set up a new matchmaking service, Chemistry.com, using an algorithm created by Helen E. Fisher, an anthropologist at Rutgers who has studied the neural chemistry of people in love.
As the matchmakers compete for customers — and denigrate each other’s methodology — the battle has intrigued academic researchers who study the mating game. On the one hand, they are skeptical, because the algorithms and the results have not been published for peer review. But they also realize that these online companies give scientists a remarkable opportunity to gather enormous amounts of data and test their theories in the field. EHarmony says more than 19 million people have filled out its questionnaire.
Its algorithm was developed a decade ago by Galen Buckwalter, a psychologist who had previously been a research professor at the University of Southern California. Drawing on previous evidence that personality similarities predict happiness in a relationship, he administered hundreds of personality questions to 5,000 married couples and correlated the answers with the couples’ marital happiness, as measured by an existing instrument called the dyadic adjustment scale.
The result was an algorithm that is supposed to match people on 29 “core traits,” like social style or emotional temperament, and “vital attributes” like relationship skills. (For details: nytimes.com/tierneylab.)
“We’re not looking for clones, but our models emphasize similarities in personality and in values,” Dr. Buckwalter said. “It’s fairly common that differences can initially be appealing, but they’re not so cute after two years. If you have someone who’s Type A and real hard charging, put them with someone else like that. It’s just much easier for people to relate if they don’t have to negotiate all these differences.”
Does this method actually work? In theory, thanks to its millions of customers and their fees (up to $60 a month), eHarmony has the data and resources to conduct cutting-edge research. It has an advisory board of prominent social scientists and a new laboratory with researchers lured from academia like Dr. Gonzaga, who previously worked at a marriage-research lab at U.C.L.A.
So far, except for a presentation at a psychologists’ conference, the company has not produced much scientific evidence that its system works. It has started a longitudinal study comparing eHarmony couples with a control group, and Dr. Buckwalter says it is committed to publishing peer-reviewed research, but not the details of its algorithm. That secrecy may be a smart business move, but it makes eHarmony a target for scientific critics, not to mention its rivals.
In the battle of the matchmakers, Chemistry.com has been running commercials faulting eHarmony for refusing to match gay couples (eHarmony says it can’t because its algorithm is based on data from heterosexuals), and eHarmony asked the Better Business Bureau to stop Chemistry.com from claiming its algorithm had been scientifically validated. The bureau concurred that there was not enough evidence, and Chemistry.com agreed to stop advertising that Dr. Fisher’s method was based on “the latest science of attraction.”
Dr. Fisher now says the ruling against her last year made sense because her algorithm at that time was still a work in progress as she correlated sociological and psychological measures, as well as indicators linked to chemical systems in the brain. But now, she said, she has the evidence from Chemistry.com users to validate the method, and she plans to publish it along with the details of the algorithm.
“I believe in transparency,” she said, taking a dig at eHarmony. “I want to share my data so that I will get peer review.”
Until outside scientists have a good look at the numbers, no one can know how effective any of these algorithms are, but one thing is already clear. People aren’t so good at picking their own mates online. Researchers who studied online dating found that the customers typically ended up going out with fewer than 1 percent of the people whose profiles they studied, and that those dates often ended up being huge letdowns. The people make up impossible shopping lists for what they want in a partner, says Eli Finkel, a psychologist who studies dating at Northwestern University’s Relationships Lab.
“They think they know what they want,” Dr. Finkel said. “But meeting somebody who possesses the characteristics they claim are so important is much less inspiring than they would have predicted.”
The new matchmakers may or may not have the right formula. But their computers at least know better than to give you what you want.