Humanities & Social ScienceIt has been known for years that interracial marriages have higher than expected divorce rates. But I did not know that the rates varied quite a bit contingent on the combination of race & sex. Gori Girl* has a post up, Interracial Divorce in the U.S. - Statistics and How Much They Matter:
- Marriages that do not cross a race barrier, but do have different ethnicities (i.e. white/Hispanic white) have a rate of divorce just a little higher than white/white marriages.- Interracial marriages that have one white person and one person of another race mostly only show higher divorce rates when the white spouse is a female (i.e. white guy + other race girl don't show particularly high divorce rates compared to same-race couples).
- Black husband/white wife marriages are twice as likely to divorce as white/white marriages, and Asian husband/white wife marriages are about 60% more likely to divorce as white/white marriages. Which, I suppose is an unfortunate statistic for Aditya and me (and one I didn't expect at all)!
- White husband/black wife were nearly 50% less likely to divorce than white/white couples, and white husband/Asian wife couples had pretty much the same divorce rate as white/white couples
- Compared to Hispanic/Hispanic couples, Hispanic white/white couples showed a higher likelihood of divorce (not surprising). Likewise, Asian/white couples were more likely to divorce than Asian/Asian couples. However, black/white couples only show a higher rate of divorce compared to black/black couples if the white person in the relationship is a woman.
- The researchers were unable to evaluate other sorts of interracial marriages, such as black/Asian, because of the low number of such couples in the sample data.
The higher divorce rates for white female/non-white male dyads somewhat surprised me, but perhaps I shouldn't have been. There seems a robust body of social science that white women in the United States are particularly race conscious. I stumbled onto a paper which emphasized this point, and its subtle nuances, recently. I didn't blog it because the sex difference in racial mating preference is such a robust finding that it doesn't seem noteworthy (though perhaps racialist message boards should focus a bit more on male rather than female "race traitors," as it is the XY who are more open to sullying the purity of the volk). In any case, the paper is Gendered racial exclusion among white internet daters. Here's the figure which stood out to me:
The big finding is this: among white women who stated a racial preference on Yahoo Personals, 73%, 64% of these selected that they wished to date white men only. In contrast, among white men who stated a racial preference, 60%, only 29% of these stated that they would like to date white women only. In other words: men who expressed racial preferences were excluding particular types of women, while women who expressed racial preferences were doing so to include only men of their own race. This seems to comport with previous work which suggests that once you control for perceived physical attractiveness men do not discriminate much in dating, while women continue to discriminate even if they assess out-group males to be physically attractive (e.g., one would conclude from this that the white male aversion to dating black women has to do with assessments of physical appearance, as much of the discriminatory effect disappears once you control for assessed attractiveness). This makes some sense, men generally weight physical appearance higher than women in the calculus which determines their preferences.
What can you infer by combining the findings of the two surveys? I'll leave it up to readers. I am struck though by the low divorce rates in the pairings of white males with black women. This is a combination against general social expectation. There are a fair number white men who prefer to date and marry black women, Robert De Niro and Roger Ebert are two prominent instances. One might guess that these men go through a "social filter" whereby their pairings have to be strongly favored for them to be continue onto marriage. But how to explain the high divorce rates of white women who marry Asian American men? The same logic should apply. I suspect that a general model may not be helpful here, and that particular cultural and social dynamics may differ across all the pairs (e.g., I assume Gori Girl expected higher divorce rates for Hispanics who marry whites than those who in-marry because Hispanic Catholic culture frowns against divorce). In contrast to this complexity, there are now a large number of studies which suggest that women have a stronger endogamous preference than men. This holds true in elite Ivy League samples, and more general population samples. Though the magnitude may vary by race (white women are invariably extremely race conscious, black women seem to be more varied according to study demographics, and in some cases Asian women are not racially discriminatory against white males, though they do not prefer them) the male-female gap seems to persist robustly. I'll withhold from offering an evolutionary psychological explanation lest "miko" unleash his wrath :-)
Note: Of course, until we get some non-American studies perhaps these are simply telling us something about American culture. Though I believe that the Columbia speed dating study including international students....
* "Gori" is a term South Asian term for a European female. More specifically, it seems a Hindi-Punjabi term which has spread in the Diaspora. I hadn't heard the term until I was an adult because I never watched Hindi movies, and my family speaks Bengali, and we do not use this term (at least in the dialect of Bengali which we speak). Though I think "Gori Girl" is married to a Bengali man, so I don't know.
Comments
I'm not at all surprised the divorce rate is higher. I'm in a similar ethnicity/multi-cultural marriage myself and all I can say is that even when we find our own common ground, the pressures due to incomprehension and intolerance in the extended family are very high. Our marriage wouldn't have lasted long without a severe limiting of contact with them on both sides. Being similar ethnicities probably means that we don't have to add problems with our wider social circle to the mix.
Posted by: Pen | March 6, 2010 10:11 AM
Razib, do you think this may have something to do with gender roles and expectations somewhat peculiar to American society? That male Asians and Blacks behave in ways which White women "wont take lying down" but women who have fewer choices or different expectations will accept? that Asian and Black women who marry White males are less willing to rock the boat?
When I was at an elite University, I saw a very large number of "White man-Asian woman" couples, mostly Chinese women, and more than once heard the comment that they are "good looking, low maintenance, hard working and very good with money"...I realize this is entirely anecdotal but do you think these notions have any place in this or do some other biological factors trump all the cultural gloss we put on our actions?
Posted by: omar | March 6, 2010 12:27 PM
omar, i don't know. but i think the data above suggests that good looking women are somewhat pickier re: race, fwiw.
Posted by: razib | March 6, 2010 2:23 PM
The dating preferences don't seem relevant to me to the divorce rates. Yes, a lot of white women don't want to date other races. More don't want to marry other races. But why should this tell us about the ones who actually do? Maybe Omar's comment explains why interracial dating doesn't lead to marriage, but by the time people do marry, shouldn't they be past that hurdle?
Posted by: Douglas Knight | March 6, 2010 2:52 PM
douglas, for what it's worth, i'm pretty sure lots of women who would race filter on a dating website would still possibly marry a colored guy if they met him at work and he was a nice guy. the same issue probably crops up for short dudes. if you're short, you're fucked on a dating site. not totally easy in real life, but not as much of an issue because you aren't automatically filtered, just manually.
Posted by: razib | March 6, 2010 2:57 PM
An additional limitation is that our data did not allow
us to capture the relative influence of other types of
assortative mating, such as class or religion.
from the study. not controlling for whatever primitive superstition people adhere to is a major limitation i would think.
Posted by: razib | March 6, 2010 5:29 PM
There's the whole "Asian" confusion thing again. I get the feeling "Asian" means East Asian here, because of the presence of "East Indian" as a separate category, but then, the blog you cite has this comment
"Which, I suppose is an unfortunate statistic for Aditya and me (and one I didn't expect at all)!".
Posted by: deadpost | March 6, 2010 6:42 PM
deadpost, don't know. the data i'm seeing suggest that perhaps east indians/east asian males are interchangeable in this context. you know, the stereotype of nerdy small dicked dudes. probably it would be diff in the UK, where the typical brown person is more likely to be punjabi (so larger), and a thug or someone who blows themselves on a bus because they're a moron. vancouver would be a good contrast, as the sikh community there is pretty thuggish (i asked a couple of far-left friends who went to UBC for a few years, and they admitted that looked at brown people different after their car was jacked by young brown males a few times who were only in it for thrill-crime it turned out).
Posted by: razib | March 6, 2010 6:52 PM
Razib, I remember some place you did mention that South Asians didn't have a sex disparity in dating, unlike East Asians.
From my experience, it does seem that the nerd stereotype is a bit less strong for South than East Asians -- perhaps South Asians are seen more like say Filipinos or other south-east Asians-- sometimes depicted as shy and extraverted, sometimes more thuggish and outgoing. I do see it's most apparent in Canada (well, at least Toronto and Vancouver) because that's where both groups of "Asians" occur side by side in large populations, not one numerically dominating the other as in the Us or UK.
Maybe the popular definition of "Asian" in North America as a whole is expanding with more "brown" people, but from what I gather, not that many South Asians want to identify as Asian -- when they do they do it's often only in the context of the shared stereotypes like "Us asians have more conservative family values/are good at math etc.".
In Toronto, where South Asians and Chinese are near equal in demographics, I'll even hear a phrase like "he's a South Asian and his friend is Asian" which sounds strange. Brown or South Asian are equally common synonyms here, of course with the former more vernacular and common with young folk. Perhaps it's easier to identify as a separate "race" and call yourself "brown" and not Asian in Toronto or Vancouver, whereas I'm under the impression that brown means or can be confused with either "mixed race black/white, Hispanic or even general dark-skinned person" when spoken in the US, so perhaps South Asian "browns" stick to the Asian label.
Hmm... its an interesting question where did the label "brown" first appear -- among the S.Asian diaspora in the US, Canada or the UK?
Posted by: deadpost | March 6, 2010 9:40 PM
yes, no sex disparity. but for east asians a lot of the disparity goes away if you match those who were born and raised here to those who were born and raised here.
as yes, "brown" is claimed by latinos, and sometimes even blacks in the USA. i think that's dumb, as they have enough accurate terms, and south asians are browner than latinos, and blacks have two really well known terms already.
Posted by: razib | March 6, 2010 9:59 PM
I think it's possible for Asians/Hispanics and a white spouse, that number of generations raised in the U.S. has some bearing on the probable success of the marriage.
There are three Asian husband/white wife marriages in my immediate (but blended) family, but all the participants' are at minimum 2nd generation U.S. There's also a strong military cultural connection in that all the Asian males have served in the U.S. military as have two of the females.
The only major cultural disagreements at family get-together's include several sets of in-laws has been over which weeds are edible and whether raw fish ever is. Unlike Pen's situation (#1) there is very little intolerance and lack of comprehension because all have a shared U.S. culture.
Posted by: Donna B. | March 6, 2010 10:17 PM
Fascinating stuff. A few thoughts:
1. White men/Black women: Although it is not PC to say it, the evidence at hand seems to indicate that White men find most Black women unattractive. A quick scan of female porn stars bears this out (although it also seems to indicate that a fairly large chunk of White men are turned on by Black man/White woman porn).Porn is, of course, an excellent diognostic tool for testing sexual attraction, as the typical consumer is looking for the most efficient way to get off, PC be dammned.
2. White/Black realationships: although such relationshps are rare (East Asian/White relationships are much more common in my social circle), I have noticed that the Black women of my aquaintence who date White men are striking "White" in social affect. I.E., their accent, music tastes, clothing style, film preferences, are indistinguishable from those of my White friends. This is much less true of the White woman/Black male couples of my aquaintence. In these relationships, the Black male is typically quite Black in affect, especially in terms of accent. This might help to explain the dichotomy in divorce rates, as it indicates that Black women are more culturally compatible.
3.White/White Hispanic: WE really need to see religious data. Are the White Anglos (I need some kind of adjective) mostly Protestant, or are they Catholic? I think that we can assume that White Anglo Protestant/White Hispanic Catholic marriages would have a higher divorce rate.
4. Browns: You have to give this one up, Razib. DEspite the phenotypical evidence being against them, the Hispanics are not going to give up calling themselves brown. I suggest that South Asians appropriate Bronze, as the Hispanics seem to have cast that one aside.
Posted by: trajan23 | March 6, 2010 11:12 PM
.Porn is, of course, an excellent diognostic tool for testing sexual attraction, as the typical consumer is looking for the most efficient way to get off, PC be dammned.
just a minor note here: i think a lot of the trends in porn are driven by "super-users." that is, 10% of guys consume 90% of the porn, and those 10% get really depraved and hardcore (i remember reading this in bill margold's book). i doubt most guys who watch DPs would ever want to be part of one, but i suspect super-users can't "get off" on anything softer anymore so they are driving demand.
Posted by: razib | March 6, 2010 11:27 PM
one error above: the yahoo personals data set showed that heavier white women were more likely say they'd date black men. those with slimmer physiques were pickier. but, the columbia data set showed that good looking males and females were the least race conscious. so conflicting results.
Posted by: razib | March 6, 2010 11:30 PM
. White/Black realationships: although such relationshps are rare (East Asian/White relationships are much more common in my social circle),
this is wrong. it was my impression too. but actually there's approximate numerical parity, just not proportional parity. that is, there are higher outmarriage/relationship rates for asians than blacks, but there are 3 X more blacks in the USA than asians. you can look at the census data.
Posted by: razib | March 6, 2010 11:39 PM
1. Porn and super-users: I think that you are right on the topic of the jaded super-user of porn, Razib. Perhaps that demographic is fuelling the more outre styles of porn (Dps, shemales, bukkake, etc). On the other hand, I do think that my point regarding the dearth of Black female porn stars holds up.
2.Rarity of White/Black relationships vs. White/East Asian: That was sloppy phrasing on my part. I meant to say that White/Black relationships are less common than White/East Asian relationships in my social circle, not overall.
Posted by: trajan23 | March 7, 2010 12:33 AM
REgarding interracial rlationships, Razib, have you read Peter Frost's recent stuff? He has a theory that the Chinese will, in the very near future, begin importing large numbers of Black African (largely Ethiopian, he argues) women to make up for the Chinese female birth dearth. I am very dubious about this and am curious as to your thoughts on the matter.
Posted by: trajan23 | March 7, 2010 12:50 AM
I am very dubious about this and am curious as to your thoughts on the matter.
LOL.
japanese and koreans get them from the philippines and vietnam.
Posted by: razib | March 7, 2010 12:59 AM
Razib, the data shows that 96.2% of white women will not date an East Indian and 82% of white men will not date an East Indian woman
Does this not contradict your expectatio that the Desi will vanish by inter-marriage
Posted by: Shan | March 7, 2010 1:51 AM
shan,
please read more closely. those percentages are *subsets*. so around 70% of women (nearly 100% of the 72% of women who selected racial preferences). second, my assertion was based on census data. ~30-40% of american-born indian americans here who marry native born people do not marry indian americans. without immigration that would lead to the dissolution of the group as indian american fertility is at or below replacement, so the endogamous component would shrink. third, there is some difference between what people will do on internet dating sites, and revealed preferences. the yahoo personal data would imply that indian american women would have notably higher outmarriage rates. they do not. this is probably because operationally the pool of potential non-browns who is date/marry browns is so large in relation to the browns that the difference in numbers is not relevant.
is that clearer for you?
Posted by: razib | March 7, 2010 1:57 AM
Who's Robert Ebert? Do you mean Roger?
Posted by: chanson | March 7, 2010 3:09 AM
my bad chanson.
Posted by: razib | March 7, 2010 3:23 AM
the same issue probably crops up for short dudes. if you're short, you're fucked on a dating site.
I'd say that is exactly what you're not likely to get. (-;
Posted by: Martin R | March 7, 2010 7:24 AM
Just noticed another Shan commenting on your blog Razib (not me, the Shan from gmail chat!).
Also, gori = white girl, gora = white guy, gorai = white people. Hear this term often in my Urdu-speaking circles.
I don't know if this is just a stereotype or what, since I have no data, but maybe the higher percentage of white females is indicative of a general greater willingness among white women to consider divorce than non-white women? I know in many non-white cultures, divorce is frowned upon and as such, they may avoid it more even if there are problems.
Posted by: Shan | March 7, 2010 10:48 AM
Surprised there was so much discrimination against East Indians.
Posted by: TGGP | March 7, 2010 2:14 PM
"Black male is typically quite Black in affect...Black women are more culturally compatible." from above:
I agree with this. I have known white women who date/marry/then divorce black men, and as a white woman I've met quite a few black men that have hit on me, and one thing that always strikes me about these situations is that as much as I want to be PC, I notice that the black men make a very big deal out of their race. I can sit down with an Asian man or a black woman and have a conversation where race as a subject does not come up, long before the topic will ever arrise; yet, I have not seen a SINGLE exception out of dozens of experiences where black men do not bring up race in talking to a woman. "Ever date a brother?" "You know what they say about black men." "your problem is you should date black men instead. They treat women better" etc. If a white man said the same thing he'd be branded a racist. Sorry, it's not PC, but highlighting differences and talking about race all the time rarely makes for a comfortable arrangement. I'm certain there are exceptions, which might be some of the marriages that 'work'.
Posted by: Carolyn | March 8, 2010 1:49 PM
I wonder how class works into this? I've been in a happy interracial marriage for a long time and come up with a predictor of "marriage success" after seeing friends and acquaintences' relationships: if the husband and wife are from similar class backgrounds, whether in the "old country" or here, it is likely to work. If they're from different class backgrounds (ie, he's an American professional, she's from the Chinese working class), it may be more problematic.
My guess is different culture + different class makes for too many moving parts to make a successful relationship.
Posted by: Foobarista | March 8, 2010 1:55 PM
@Douglas Knight - Doesn't surprise me. The dating preferences obviously speak to huge pressure to marry white for white women. Even if they break free of that and marry a non-white, there is no reason to believe that the pressure drops externally or that they embraced their choice completely internally.
All I can say is that when a non-white meets a white girl who would consider dating them, it is always a surprise. I remember pretty well from my single days that white women don't even look at non-whites as potentials... they look right past them.
Posted by: kes | March 8, 2010 2:39 PM
First, as has been mentioned, "stated willingness to consider for dating" is a very, very, different number than "actually will marry". Yes, they are related, but I would want to look at the actual marriage statistics before going very far with this data.
Having said that, I'm amazed that the "c" word has not come up. While men have a long and ingrained history of indiscriminate mating, women are the opposite. While men mate for looks, women do for status. News flash: whites have a distinct class advantage in the United States.
Lower-class women are far less likely to divorce a high-class man. The persistence of a large black underclass in this country directly explains the W-m/Black-f "anomaly" in the data.
The family pressure to divorce need not be racist at all. If I knew someone who was being beaten by her husband, pressuring her to divorce would be the most mild of my responses. If a family member were to marry into someone from a background where such behavior were normal, things would get ugly. I don't care how much money he has or what his color. I hear it's all the same on the inside, anyway.
To summarize: I don't see anything in this data which cannot be explained by 1) women's tendency to "marry up", and to tolerate more in marriage when they succeed in this goal; 2) familial insistence that those marrying a woman of the clan treat her as well as they treat their wives; 3) extant correlations between race and class in the United States.
Posted by: Right Wing-nut | March 8, 2010 2:50 PM