Are Marriages That Start Online Happier? The Research
A 19,131-person study found marriages that began online were slightly happier and less likely to end. Here's what it showed and who funded it.
Quick Answer
The largest study on the question found marriages that began online were slightly happier and slightly less likely to end. Cacioppo and colleagues surveyed 19,131 people married from 2005 to 2012: 5.96% of online-met marriages had ended in separation or divorce versus 7.67% of offline-met ones, and satisfaction was marginally higher. But the effect was small, the survey was commissioned by eHarmony, and independent research finds meeting online is no worse, not clearly better.
For years, the assumption was that relationships formed online were second-best: more superficial, more disposable, built on too much choice. The research doesn't support that. But the better-known study often cited for the opposite claim, that online marriages are happier, needs to be read carefully, including its funding.
The Study Behind the Claim
The paper is "Marital satisfaction and break-ups differ across on-line and off-line meeting venues" by John T. Cacioppo, Stephanie Cacioppo, Gian C. Gonzaga, Elizabeth L. Ogburn and Tyler J. VanderWeele, published in the Proceedings of the National Academy of Sciences in 2013.
The data came from an online survey run by Harris Interactive in 2012. The analysis covered 19,131 respondents who had married between 2005 and 2012. Each person reported where they had met their spouse, whether the marriage was still intact and, if so, how satisfied they were.
About 35% had met their spouse online. The paper puts it at 34.95% of respondents, and summarizes that "more than one-third of marriages in America now begin on-line." Among those who met online, 45% met on dating sites; about 21% met through online social networks.
What It Found: Breakups and Satisfaction
Breakups. Among all respondents, 5.96% of marriages that started online had ended in separation or divorce, compared with 7.67% of marriages that started offline. The difference remained statistically significant after controlling for year of marriage, sex, age, education, household income, ethnicity, religion and employment.
Satisfaction. Among those still married, people who met online scored an average of 5.64 on the satisfaction measure, compared with 5.48 for those who met offline. Satisfaction was measured with standard questionnaire items, including the four-item Couples Satisfaction Index, which asks things like how happy, all things considered, the marriage is.
| Outcome | Met online | Met offline |
|---|---|---|
| Marriage ended in separation or divorce | 5.96% | 7.67% |
| Average marital satisfaction (still married) | 5.64 | 5.48 |
Both differences are real in the statistical sense. They're also small. The authors say so directly: the effects of where people meet "are understandably quite small and do not hold for everyone."
Where the Happiest Couples Met
The study also broke results down by specific venue, which is where it gets more interesting than the online-versus-offline headline.
Among offline venues, people who grew up together or met through school, a place of worship or a social gathering reported the highest marital satisfaction. People who met through work, family, a bar or club, a blind date or other venues reported the lowest.
Among online venues, people who met through an online community or chat room reported lower satisfaction than those who met through other online venues.
Who Funded It, and Why It Matters
The paper discloses that Harris Interactive was commissioned by eHarmony to conduct the survey, and that some authors had eHarmony affiliations. To address that, two independent statisticians on the author list, Elizabeth Ogburn and Tyler VanderWeele, oversaw and verified the statistical analyses.
Funding doesn't make a result wrong. This was peer-reviewed and published in a major journal with its conflicts stated. But a dating company has an obvious interest in showing online-met marriages do well, and that's a reason to look for independent work before treating the finding as settled.
The authors also flagged other limits:
- The sample was voluntary and online. They note this "may partially limit representativeness," and an online survey may over-represent people comfortable online.
- The time window was short. The results cover roughly the first six or seven years of marriage. Whether the gap grows or shrinks later is unknown.
- Selection is a big unknown. People who met online differed demographically. The paper reports they were more often men, aged 30 to 49, Hispanic, higher socioeconomic status and employed. Controls help, but can't account for every difference in who chooses to date online.
What Independent Research Says
The most useful independent check is Stanford sociologist Michael Rosenfeld's 2017 paper in Sociological Science, "Marriage, Choice, and Couplehood in the Age of the Internet." Using longitudinal data, he found that "meeting online does not predict couple breakup," and that meeting online, particularly through dating websites, predicted faster transitions to marriage for heterosexual couples.
Rosenfeld was careful not to claim causation. He wrote that the data "are more consistent with a positive or neutral association between Internet technology and relationships than with a negative association." We covered his broader findings on how couples meet in what Stanford's data shows.
Put the two studies side by side and the defensible summary is:
| Claim | Supported? |
|---|---|
| Relationships that start online are more fragile | No |
| Online-met couples are at least as stable as others | Yes, in multiple studies |
| Online-met marriages are clearly happier | Weak evidence: one industry-funded study, small effect |
| Online-met couples move to marriage faster | Supported in Rosenfeld (2017) |
What This Means If You're Dating Now
Drop the stigma. If you've worried that meeting someone through an app or service makes a relationship less real, the research doesn't back that up. Online dating has its real problems, which Pew's survey work documents in detail (see what Pew research says about online dating), but the durability of the relationships it produces isn't one of them.
The venue matters less than who's in it. The venue findings suggest something practical: places where people share values, social circles or intentions seem to produce more satisfied couples. That's an argument for choosing where you meet people deliberately, including online, rather than defaulting to whatever has the most profiles. We've written about the limits of letting an algorithm choose a partner for the same reason.
Don't over-read small effects. A 1.7 percentage-point difference in breakups is interesting at the population level and close to irrelevant for any one couple. How two people communicate, handle conflict and agree on the big decisions will matter far more than where they met. For the bigger picture on marriage stability, see our look at U.S. divorce rate trends.
The Bottom Line
Marriages that start online aren't worse, and that's now well supported. Whether they're better rests mostly on one 2013 study: large, peer-reviewed and transparent about its eHarmony funding, with effects its own authors called small. The more useful takeaway is about context. Couples who met where people shared values and intentions tended to report happier marriages, online or off.
Sources:
- Marital satisfaction and break-ups differ across on-line and off-line meeting venues — John T. Cacioppo, Stephanie Cacioppo, Gian C. Gonzaga, Elizabeth L. Ogburn and Tyler J. VanderWeele, Proceedings of the National Academy of Sciences 110(25), 10135–10140 (2013)
- Marriage, Choice, and Couplehood in the Age of the Internet — Michael J. Rosenfeld, Sociological Science 4, 490–510 (2017)
Frequently Asked Questions
Are marriages that start online happier?
One large study says slightly. Cacioppo and colleagues (PNAS, 2013) surveyed 19,131 people married between 2005 and 2012 and found those who met online reported marginally higher marital satisfaction, 5.64 versus 5.48 on average. The authors called the effect small, and the survey was commissioned by eHarmony.
Do couples who meet online get divorced less?
In the 2013 PNAS study, 5.96% of marriages that began online had ended in separation or divorce, compared with 7.67% of those that began offline, and the gap held after demographic controls. A separate 2017 analysis by Stanford sociologist Michael Rosenfeld found meeting online did not predict breakup, which is best read as 'no worse,' not 'better.'
Was the online dating marriage study funded by eHarmony?
Yes. The paper states that Harris Interactive was commissioned by eHarmony to run the survey, and some of the authors had eHarmony affiliations. Two independent statisticians oversaw and verified the analyses. It's a peer-reviewed study, but the funding is a reason to weigh it alongside independent research.
Where do the happiest couples meet?
In the 2013 study's offline categories, people who grew up together or met through school, a place of worship or a social gathering reported the highest satisfaction. Those who met through work, family, a bar or club, or a blind date reported the lowest. These are correlations, not evidence that a venue causes a happier marriage.
Why might couples who meet online be more satisfied?
The authors suggested a larger pool of potential partners, selection effects in who uses dating sites, and differences in how people disclose information when they first meet online. They said which explanation is right remains an open question.
Related reading
Do the 36 Questions Work? What Aron's Study Found
The 36 questions to fall in love come from a 1997 closeness study. Here's what Arthur Aron's research measured, found and never claimed.
Americans Are Socializing Less: The Time-Use Data
BLS data shows Americans spent 35 minutes a day socializing in 2025, down from 41 in 2015. What the numbers say, and what they mean for meeting someone.
Is Cuffing Season Real? What the Research Actually Shows
Cuffing season is usually blamed on cold weather. A 129-country study of search data points somewhere else: the holidays. Here's what the evidence supports.
Ready to find your person?
Tenr is the invite-only 10-minute date app for high achievers.