More than a third of recent marriages in the USA started online, according to a study out Monday that presents more evidence of just how much technology has taken hold of our lives.
"Societally, we are going to increasingly meet more of our romantic partners online as we establish more of an online presence in terms of social media," says Caitlin Moldvay, a dating industry senior analyst for market research firm IBISWorld in Santa Monica, Calif. "I do think mobile dating is going to be the main driver of this growth."
The research, based on a survey of more than 19,000 individuals who married between 2005 and 2012, also found relationships that began online are slightly happier and less likely to split than those that started offline.
Findings, published in the journal Proceedings of the National Academy of Sciences, put the percentage of married couples that now meet online at almost 35% -- which gives what may be the first broad look at the overall percentage of new marriages that result from meeting online. About 45% of couples met on dating sites; the rest met on online social networks, chat rooms, instant messaging or other online forums.
Lead author John Cacioppo, a psychologist and director of the Center for Cognitive and Social Neuroscience at the University of Chicago, says dating sites may "attract people who are serious about getting married."
While Cacioppo is a noted researcher and the study is in a prestigious scientific journal, it is not without controversy. It was commissioned by the dating website eHarmony, according to the study's conflict of interest statement. Company officials say eHarmony paid Harris Interactive $130,000 to field the research. Cacioppo has been a member of eHarmony's Scientific Advisory Board since it was created in 2007. In addition, former eHarmony researcher Gian Gonzaga is one of the five co-authors.
"It's a very impressive study," says social psychologist Eli Finkel of Northwestern University in Evanston, Ill. "But it was paid for by somebody with a horse in the race and conducted by an organization that might have an incentive to tell this story.
"Does this study suggest that meeting online is a compelling way to meet a partner who is a good marriage prospect for you? The answer is 'absolutely,'" he says. But it's "premature to conclude that online dating is better than offline dating."
The findings about greater happiness in online couples "are tiny effects," says Finkel,whose research published last year found "no compelling evidence" to support dating website claims that their algorithms work better than other ways of pairing romantic partners.
Finkel says the overall percentage of marriages in the survey is "on the high end of what I would have anticipated."
Sociologist Michael Rosenfeld of Stanford University in Stanford, Calif., says the numbers seem "reasonable."
He says his own research, published last year in the American Sociological Review, found 22% of newly formed couples had met online, "but couples who meet online are more likely to progress to marriage than couples who meet in other ways." He says his new analysis of nationally representative data found that of 926 unmarried couples followed from 2009 to 2011, those who met online were twice as likely to marry as those who met offline.
Although Rosenfeld says the paper is a "serious and interesting paper" and "Cacioppo is a serious scholar with a big reputation," he is concerned that "the use of an Internet survey which leaves non-Internet households out might bias the results."
Harris Interactive says the results have been weighted to correct for potential bias in its online surveys. Other new data released last month from a Pew Research Center survey found that just 15% of Americans report not using the Internet.
Cacioppo defends the results, and says that before he agreed to analyze the data, "I set stipulations that it would be about science and not about eHarmony." He adds that two independent statisticians from Harvard University were among co-authors.
"I had an agreement with eHarmony that I had complete control and we would publish no matter what we found and the data would be available to everyone," he says.
Originally published by