Does Online Dating Increase Racial Intermarriages?

Today about a third of all new marriages are between couples who met online. Online dating has an interesting property–you are likely to be matched with a total stranger. Other matching methods, like meeting through friends, at church or even in a local bar are more likely to match people who are already tied in a network. Thus, the rise of online dating is likely to significantly change how people connect and are connected to one another in networks. Ortega and Hergovich consider a simple model:

We consider a Gale-Shapley marriage problem, in which agents may belong
to different races or communities. All agents from all races are randomly
located on the same unit square. Agents want to marry the person who is
closest to them, but they can only marry people who they know, i.e. to whom
they are connected. As in real life, agents are highly connected with agents
of their own race, but only poorly so with people from other races.

Using theory and random simulations they find that online dating rapidly increases interracial marriage. The result happens not simply because a person of one race might be matched online to a person of another race but also because once this first match occurs the friends of each of the matched couples are now more likely to meet and marry one another through traditional methods. The strength of weak ties is such that it doesn’t take too many weak ties to better connect formerly disparate networks.

Interracial marriage, defined to include those between between White, Black, Hispanic, Asian, American Indian or multiracial persons, has been increasing since at least the 1960s but using the graph at right the authors argue that the rate of growth increased with the introduction and popularization of online dating. Note the big increase in interracial marriage shortly after the introduction of Tinder in 2009!

(The authors convincingly argue that this not due to a composition effect.)

Since online dating increases the number of potential marriage partners it leads to marriages which are on average “closer” in preference space to those in a model without online dating. Thus, the model predicts that online dating should reduce the divorce rate and there is some evidence for this hypothesis:

Cacioppo et al. (2013) find that marriages created online were less likely to
break up and reveal a higher marital satisfaction, using a sample of 19,131
Americans who married between 2005 and 2012. They write: “Meeting a
spouse on-line is on average associated with slightly higher marital satisfaction
and lower rates of marital break-up than meeting a spouse through
traditional (off-line) venues”

The model also applies to many other potential networks.

Hat tip: MIT Technology Review.


What happened at 1980?

A random fluctuation?

Reagan was elected. Pope John Paul II visited Brazil, including my hometown, the Shah died, Mugabe was elected president.

M*A*S*H made it OK to date Asian chicks

Mariel boatlift, 1980. They classify Cubans marrying whites as "interracial".

Bonus trivia: it's theoretical but to my knowledge not yet actual grounds for deportation if you fail to mention in your immigration application that you met your husband online, TC would know about this rule. I'm sure Trump's racist minions will try and enforce this stupid law, passed a few years ago in response to Green Card marriages, at some point. Also, immigration officers are allowed to, and at times ask, "CIA interview" questions (if you are familiar with this term, and yes, years ago the CIA did ask these questions) such as 'how often do you have sex' and (CIA question) 'in what positions, what duration, what frequency'?

Perhaps 1980 was a Census count using the full enumeration while the other years were estimates based on samples?

The attribution in the chart above is wrong. It is from the Pew Research Center but the underlying data are from the ACS 2008-2015 and the 1980 Census only. These are "look-back" statistics based on year of marriage collected in 2008-2015 from the 2008-2015 data for 1980-2015 marriages. The data for 1979 and earlier come from year of marriage data collected in the 1980 Census (5% sample). The blips in 1980 and 1985 are probably artifacts of a tendency to "round-off" years in reporting distant events in the past. I wouldn't read too much into them

Mt. Saint Helens eruption

In 1998, Posner "defended" sex discrimination (harassment) in the workplace because, with so many women working, the workplace serves as an important site for courtship. See Cowen's blog post before this one by Tabarrok. Now that online dating is so popular (did it even exist in 1998?) and an alternative to the workplace as a site for courtship, Posner may have changed his view about sex discrimination (harassment). It's probably just a matter of time before courts recognize a new class of sex discrimination (harassment) in online dating, perhaps borrowing from Posner's distinction between "low status women" and "high status women" in the workplace, providing the "low status woman" a claim for rejection by someone who only will accept a "high status woman" in online dating.

Forms of online dating have existed for a very long time - like singles ads and also back in the mainframe days, matching organisations where they basically matched people using probably similar algorithms to the existing ones used by Tinder. And of course mail order brides also was a big thing in the 1990s. What has changed seems to be attitudes towards these things, people used to hide the fact that they had to find partners in non traditional ways, it was probably a matter for shame that they could attract a partner via their existing networks. But now people seem much more OK with it, so these trends shown by Alex are probably a reflection both of new technologies but also changing attitudes.

Did they control for the share of the relevant age group which is non-white?

Will this tell us if Stevie Wonder is actually blind?

I think online dating probably definitely helps break sub-bubbles between bigger bubbles but I'm with Art Deco on this.

I think a large part of this trend is due to the fact that the share of non-whites among millenials is higher than previous generations and you have increased college attendance where you're more free of familial/community prejudices (for both ways of the racial equation)/can meet more non co-ethnics from other cities, regions and countries.

Interrmarriage seems both as a good metric and good tool for cohesion in diverse societies so nice to hear.

They really should have controlled for the share of the relevant age group that's nonwhite and nonblack, and there's no way they did that. My guess is that the online dating effect is hitting at the same time that the assimilated children of the Hispanic/Asian immigration boom (who are much, much more likely to marry outside their race than whites or blacks) reach marriageable age.

In addition, we know that online dating isn't resulting in more sex partners. I'm skeptical that it results in increasing the number of potential marriage partners.

+1 This. Absolutely. Correlation is not causation. More interesting is several dating sites (OKC etc.) have run there own metrics using their own data and show distinctly the desirability ratios by certain age and race populations have gone down, not up in several assertive populations online.

I am very skeptical of their conclusions.

Thanks for your comments. We do not claim causation and this is explicitly written in the paper and on this blog. Your concerns are tackled in the paper but we are happy to hear any further comments you have.

I assume your "explicit" non-causative claim is this:
"We do not claim that the increase in the share of new marriages that are
interracial in the last 20 years is caused by the emergence of online dating
alone, but this nding is in line with Hypothesis 1 in our model."

But it is followed shortly afterwards in the conclusion by:
"With a basic structure, it can generate very strong predictions. It suggests that the diversity of societies, measured by the number of interracial marriages in it, should increase drastically after the introduction of online dating."

It makes me wonder what a claim about causation actually looks like.

They are less desirable, but the bleakness of the situation goes up. So after a while, you are like OK I guess I'll just agree to go out with that black woman with four PhDs and $1MM in annual income, because no white woman is going to go out with me.

Why do Blacks think racism is still keeping them down? These are just the facts. The Civil War ended over 150 years ago, there's no more racism now for pete's sake.

Wait, are there really women like that on dating sites going begging for dates? My strategy for becoming a millionaire is all wrong.

"we know that online dating isn’t resulting in more sex partners."

Really? I thought that was what Tinder was primarily used for. Cite for this? I guess I could believe this for the median male, and maybe the 33rd percentile woman.

Number of sex partners isn't up, therefore online dating isn't increasing the number for its users.

Of course, that's bad logic.

Sorry, "Number of sex partners for the total population isn’t up"

Links to a couple different sources here:

Dan Kopf at Pricenomics has attempted to divide the rise in the intermarriage rate into two factors: rising propensity in intermarry versus rising propinquity (is that the right word?) due to demographic change. He finds most of the recent increase in intermarriage to be a mechanical result of greater diversity in the population. Taboos against interracial marriage had largely fallen by, say, 1980.

Dontrolling for demographics, the propensity toward interracial marriage (counting Hispanics as a race) was 72% as great in 1980 as it is today.

Regarding the online dating hypothesis, the propensity to marry interracially was pretty flat from 1980 to 2000, then rose modestly in this century, which might have something to do with the hypothesis.

When in this analysis were hispanics marrying whites counted as well interracial? As you know in CA for miscegenation laws Mexican Americans were counted as white. My Mexican grandmother married my white European grandfather and the state of CA saw this as a marriage of two white people.

You can always tell an Alex post when reading it, but can you tell one just by the title of the post?

Sometimes. Not in this case for me.

Note the big increase in interracial marriage shortly after the introduction of Tinder in 2009!

It sure looks to me that there is a noticeable increase in _2008_, a drop in _2009_, a continuation of trend at the lower level in 2010-11, and then a bump up (followed by a drop again) in 2012. That doesn't seem like a pattern that's easily fit with a big increase coming from Tinder in 2009.

Agreed! It seems that 2009 marks the slowing of a trend that started in 2006 (or 2000).

Starting the chart in 1967 is telling, also. You might as well start it in the paleozoic era in terms of marriage attitudes.

If you stretched it back that far, you'd probably see a big bump in the last couple centuries. Then you could claim it was caused by the invention of the steam engine.

Starting the chart in 1967 is telling, also. You might as well start it in the paleozoic era in terms of marriage attitudes.

There was an explosion in divorce between 1967 and 1979. The attrition rate of extant marriages trebled in that 12 year period. The net change in attrition rates between 1948 and 1967 was nil. That particular year is a sensible place to start if you're tracking matrimonial behavior, though you could make a case for other dates. (BTW, Dean Rusk's daughter married a black man in 1967).

Because people who do online dating are a selected group. They couldn't find a partner the normal way. They stand in a lower position on the dating market somehow.

Not necessarily. In some cases they are high value targets whose wealth or prominence makes them vulnerable to gold diggers and social climbers in their own circle.

I don't think so. The traditional ways to meet spouses -- neighborhood, church, friends of friends, college -- aren't working so much for millennials who aren't religious, move often and delay marriage.

I know two young professionals who live in Manhattan and who are put off by the bar scene, Tinder and, in one case, prospects trolling for potential partners with presumed high incomes. One of the two doesn't have an official office, and the other works in an office where most of the workers are of the same sex. Neither they nor their friends have homes large enough to throw parties where they might meet their friends' friends. I absolutely understand why people seek out online datings sites.

Circa 2003 a friend mentioned that she and her Indian husband had corresponded back and forth online for quite awhile before she came to America (her first time leaving India) to meet in person. "Oh, so you met on a dating website?" I asked.

"Oh no, not a dating site - it was a matrimonial site," she replied.

There it's replacing arranged marriages. I'm told there are matrimonial sites for ethnic Koreans in the U.S. for those who wish to marry endogamously. Know of a doctor who found his wife that way.

Yes, they were well-matched as to - whatever; with the added bonus of that pen pal friendship. And also seemed, if not giddy with newlywed happiness, very content and considerate of one another. She had not been here long before she researched what sort of professional certification was doable for her with the credentials she had, that would have future flexible hours, and settled on speech pathologist. Childless, she could make no headway with the office women at the neighborhood elementary so asked me to get her in the door, that she might volunteer there to buff up her qualifications, or as part of a project for her classwork? - can't recall. I remember we emerged successful and giggling after finally gaining the formidable principal's office, with her muttering - somehow still sounding polite - under her breath, "Damn her! Damn that Mrs. [...]." I detected no interest in speech pathology - who could have, really? - but she had that admirable Indian thoroughness in committing to a project (I'm reminded of other Indian friends who took an online quiz to determine which dog was most suited to them; through no fault of theirs, he was in neighborhood opinion, including some rabid dog-lovers, the most worthless dog that ever lived, though they loved him dearly). I was sad when she moved. She and I liked one another, though we were mostly opposites (in ways that reflected well on her). And her husband was approximately five times more interesting to talk to than the husbands of American friends.

What kind of a person loves rabid dogs?

Since there's only one race, the human race, what exactly does this all mean? What determines the so-called "race" of either half of the couple? Was the marriage of Mr. and Mrs. Obama interracial? How many of these "races" are there? If a Korean marries a Filipino is it an interracial marriage? How about a Turk and an Iranian? The obsession with race on this site is weird.

French thinker Auguste Comte showed their are three races - White, Black, Red (Native Americans) and Yellow (Asian). Famous Haitian leader François-Dominique Toussaint Louverture is, according to Comte's Brazilian followers, the greatest Black of all time.

Truth Seeker

There are only three numbers: 1,2,3, and 4.

And four, see below.

French thinker Auguste Comte showed there are FOUR races - White, Black, Red (Native Americans) and Yellow (Asian). Famous Haitian leader François-Dominique Toussaint Louverture is, according to Comte's Brazilian followers, the greatest Black of all time.

I take it you've never heard of the census.

Not sure it's anatomically possible, in my case, but I wouldn't have minded. I don't care who fills invented professions for women.

Well, why did you delete that? It's not every day a girl gets called a cuck! I can take it as well as the rest of you!

Ta-Nehisi Coates will not be pleased with this trend.

He can see the chipotle-mayonaise trucks arriving.

Wins. Thanks.

Too many "black bodies" are now beige!

Shouldn't that chart be on a log scale if it was trying to make such an assertion?

The GOP blames the first black President, Clinton.
And they also blame Obama, too, 'cause, like, he wasn't born here anyway...

Comments for this post are closed