A. What Drives Interracial Marriage Patterns?

Most married couples do not randomly end up together but rather are the result of assortative mating--the tendency of people to date and marry individuals like themselves. We generally partner with people who are similar to us in terms of race, education, and socioeconomic status in part because we spend a lot of time with people with similar levels of education at school or at work. Our family members, friends, and neighbors also tend to be of the same race and similar socioeconomic status. Online dating studies suggest, however, that even when the pool of potential mates is not limited by whom we meet at school, work, the gym, or local bar, we still prefer to date people like ourselves. As one aptly-titled article noted, “In the End, People May Really Just Want to Date Themselves.”

*110 Dating and marriage outcomes are the result of both preferences and opportunities and thus cannot explain whether opportunity, preferences (and if so, whose preferences), or both, drive the different rates of interracial coupling. Researchers have addressed the limitations of dating and marriage outcomes by directly examining the preferences of individuals seeking a romantic partner. Studies that focus on stated preferences--what people say they want in a partner--generally ask date-seekers to identify the traits they seek in a romantic partner or examine the traits date-seekers have identified in a personal ad or online dating profile. Not surprisingly, individuals may not be completely truthful when describing the traits they seek in a partner because they fear they will be judged as superficial, elitist, or even racist. Moreover, even when we are completely honest, our stated preferences may not reflect our true preferences. As evolutionary psychologists have discovered, we often do not know what we really want in a mate.

To address the limitations of stated preferences, researchers have examined the revealed preferences of online date-seekers by observing how they respond when contacted by daters with certain traits. For example, Günter Hitsch and his colleagues examined the search behaviors of almost 22,000 heterosexual online daters. The date-seekers, who did not know that their behaviors would be observed by researchers, provided detailed profiles noting their age, gender, race, education, income, height, weight, marital status, political and religious affiliations, interest in dating someone *111 of a different ethnic background, and whether they were seeking a casual or long-term relationship. Many users also provided a photo which the researchers rated for physical attractiveness based on the opinions of objective observers. Date-seekers browsed other users' profiles and sent emails to individuals they might want to date.

Not surprisingly, online daters' search behaviors revealed a universal preference for physically attractive individuals with high incomes. However, women valued a man's income more highly than his physical appearance, and men ranked a woman's physical attractiveness above her income. Online daters' search behaviors also revealed strong racial preferences even when they did not state those preferences. For example, 55% of the women expressed no racial preferences in their profiles, but their revealed preferences--who they contacted and who they responded to when contacted--showed equally strong preferences as the women who had expressed a racial preference. In other words, 95% of female online daters in Hitsch's study had racial preferences even though only 41% stated those preferences in their profiles.

Other online dating studies have revealed racial preferences. They also reveal a racial hierarchy of preferences. For example, the majority of straight White men in an online dating study conducted by Cynthia Feliciano and her colleagues stated a racial preference. The majority also expressed*112 interest or willingness to date interracially. However, they were quite specific about which groups they were willing to date. About 50% of White men who stated a racial preference expressly excluded Asian-American women and similar numbers excluded Latina women. Yet, more than 90% refused to consider African-American women. The chart below illustrates this hierarchy.

It is no longer socially acceptable to express racial preferences in most contexts and it is illegal to act upon such preferences in settings such as education, employment, and housing. In fact, 84% of online daters in one study stated that they would not date someone “who has vocalized a strong negative bias toward a certain race of people.” Despite this strong anti-discrimination norm, studies have found a racial hierarchy in which White men rank African-American women significantly below Asian-American, Latina, or White women. This hierarchy is also reflected in straight White men's response rates when contacted by female online date-seekers. White men are most likely to respond to messages from White women and from multiracial Asian-American and Latina women who *113 are part White. They are less likely to respond to multiracial African-American/White women, and almost never respond to messages from African-American women.

White women's preferences also reveal a racial hierarchy. Almost 75% of straight White women in Feliciano's study expressed racial preferences and a majority of those (64%) preferred to date White men only. Although most White women excluded all non-White men, they were more than twice as likely to exclude African-American and Asian-American men as compared to Latino men.

Data from millions of online daters on Match (the most popular dating site in the U.S for the last 20 years), OkCupid, and Date Hookup confirm the racial hierarchy in the online dating market. Straight White women on these sites rated Asian-American and African-American men as significantly less attractive than the average man. This hierarchy is also reflected in White women's response rates when contacted by online date-seekers. Several studies conducted by Curington, Lin, and Lundquist revealed that White women respond mainly to White men and ignore messages from men of other races with one exception--multiracial men who are part White. While more than 90% of White women rejected Asian men as potential dates, they responded to messages from multiracial Asian/White men at similar rates as they did to messages from mono-racial White men. They also responded to Latino/White men and African-American/White men at higher rates than their mono-racial counterparts.

Online date-seekers have many preferences, including age, body type, education, income, and religion. But race ranks particularly high on their preferences. For example, while 59% of straight White men in Feliciano's study stated a racial preference, only 23% expressed a religious preference. For these men, a woman's race was more important than her education, religion, employment, marital status, or whether she smoked. Straight White date-seekers on the online dating site OkCupid revealed similarly strong preferences for Whites even when the system's algorithm determined that their best “match,” based on their responses to approximately 300 questions about their beliefs, needs, wants, and activities they enjoy, was a person of a different race.

College-educated minorities and Latinos/as are more likely to intermarry *114 than their less-educated counterparts, so one might assume that college-educated Americans as a group have weaker racial preferences. However, online dating studies suggest otherwise. Hitsch's study of heterosexual online daters discussed above found that the vast majority of White women, regardless of their level of education or income, have strong preferences for White men. Feliciano and her colleagues found that college-educated Whites are more likely than Whites with only a high school education to exclude African-Americans as romantic partners. And Lin and Lundquist found that racial preferences trumped educational preferences. College-educated Whites are more likely to contact and respond to messages from Whites without a college degree than to messages from African-Americans with a college degree. White men without a college degree received more messages than college-educated African-American and Asian-American men. College-educated African-American women received significantly fewer messages than women of other races with lower levels of educational attainment.

Racial minorities and Latinos are generally more willing than Whites to date interracially, yet their preferences reflect a similar racial hierarchy. For example, 70% of straight Asian-American and Latina women in an online dating study conducted by Belinda Robnett and Cynthia Feliciano expressed a racial preference and *115 overwhelmingly excluded minority men other than their co-ethics. The vast majority, however, were willing to date White men.

This racial hierarchy is also reflected in Asian-American and Latina women's response rates when contacted by online daters. They are most likely to respond to emails from White men and their multiracial co-ethnics who are part White (Asian-American/White men and Latino/White men) than to messages from their mono-racial co-ethnics. Surveys of college students' dating preferences have also found that many Latinos and Asian-Americans prefer Whites to other groups, including their own co-ethnics.

The preferences of straight Asian-American and Latino men also reflect a racial hierarchy. For example, Robnett and Feliciano found that over 60% of Asian-American and Latino men who expressed a racial preference were willing to date White women, but less than 20% were willing to date African-American women. Approximately 50% of Asian-American men were willing to date Latina women and similar numbers of Latino men were willing to date Asian-American women. The graphs below illustrate the preferences of straight Asian-American and Latino men.

While the preferences of Asian-Americans, Latinos/as, and Whites reflect a similar racial hierarchy, the stated preferences of African-American men and women do not follow this pattern. For example, African-Americans express stronger same-race preferences than Asian-Americans and Latinos and are three times as likely to expressly refuse to date individuals of other races. The majority also expressly exclude Whites as potential dates. The graph below illustrates the stated preferences of African-American online date-seekers.

Yet, in contrast to their stated preferences, the revealed preferences of straight African-American men and women suggest that they may be more willing to date Whites than indicated by their stated preferences. One online dating study conducted by Mendelsohn and his colleagues found that African-Americans were ten times more likely to contact Whites than Whites were to contact them. That study and others have also found that African-Americans are more likely to respond to messages from Whites and from multiracial African-American/White individuals than messages from African-Americans.

These studies demonstrate an undeniable racial hierarchy in the dating market. Daters' preferences for multiracial persons who are part White over mono-racial minorities further reveal a hierarchy that values light skin and European appearance. These studies also suggest that while partial Whiteness can elevate Asian-Americans and Latina women to White status, it does not have the same power for African-American women. For example, Asian-American, Latino, and White men rated online profiles and photographs of multiracial African- *118 American/White women as significantly more attractive than mono-racial African-American women but as less attractive than women of other races.

The preferences revealed by these studies are consistent across daters of different ages, incomes, education, geographic location (including urban v. rural dwellers), and self-identification as liberal or conservative. Speed dating studies and surveys of college students' preferences have found a similar racial hierarchy. For example, 381 college students at a public university in California completed an anonymous questionnaire that asked them to describe the traits they desire in a romantic partner, whether they were willing to date someone of a different race, and if so, to rank their preferred racial or ethnic groups and explain their reasons for their rankings. All of the non-Black male students who expressed racial preferences ranked African-American women last but White students were significantly less likely than Asian-American or Latino students to report any racial preferences or to expressly exclude African-Americans. However, students' explanations for their preferences reveal a racialized and gendered hierarchy fueled by Western notions of beauty, stereotypes, and family and societal disapproval. Students' most commonly stated reasons for excluding African-Americans or ranking them last included lack of physical attraction, cultural differences, perceived aggressive personality or behavior, and social disapproval. Rates of exclusion varied by gender. Heterosexual White male students were more than twice as likely as their female counterparts (67% v. 30%) to exclude African-Americans as potential dates. Asian-American males were also more likely than females to exclude African-Americans as potential dates. Men were more than twice as likely as women to cite lack of physical attraction (such as skin tone, hair texture, and body type) as reasons for excluding African-Americans as potential dates.

As noted earlier, all daters prefer physically attractive partners. Beauty may be in the eye of the beholder but throughout most of the Western world, a light complexion and phenotypically European features, such as straight hair and a *119 narrow nose, are perceived as most attractive, especially for women. These physical features are deemed desirable not only by Whites but also by African-Americans and other minority groups. As scholars have asserted, light “skin tone is also a form of social capital that grants access to ... marriage to higher status men.” One need only name a few African-American female celebrities considered universally beautiful (such as Beyonc‚ Knowles, Halle Berry, and Alicia Keys) to conclude that women with lighter skin and more Eurocentric features are perceived as most attractive.

Gendered and racialized stereotypes affect how individuals are perceived in the dating market. For example, one study found that White men who expressed a body type preference were more likely to exclude African-American women as dates, presumably because they associated African-American women with a particular body type. Another study found that the more highly a man valued femininity, the higher the likelihood that he would express interest in dating Asian-American women but not African-American woman. Several studies have found *120 that Americans perceive Asian-Americans to be more feminine than other groups, and African-Americans to be more masculine. They also associate dark skin with masculinity. Given the importance that men place on a partner's physical appearance, and the value all races place on light skin on women, it is not surprising that lighter-skinned women are higher in the racial hierarchy of the dating market.

Societal notions of masculinity and femininity are reflected in stereotypes and media portrayals of minority groups. Asian-American women are depicted as hyper-feminine, Asian-American men are portrayed as effeminate and asexual, and African-American men are depicted as hyper-masculine. Although the media is beginning to portray African-American women as desirable partners, historically, cultural depictions of Black women have generally been limited to images of matronly caregivers, sexually immoral, or emasculating, angry women.

Gender differences in the racial hierarchy are also apparent when one examines stereotypes about different groups' personalities and behaviors. Although straight women (and gay men) reject African-American men at high rates, African-American women are excluded at even higher rates. Studies show that while both men and women rely on stereotypes about African-Americans' “aggressive personality” as a reason for excluding them as dates, straight men are significantly more likely than straight women to do so. Their stated reasons reflect cultural assumptions about African-American women as emasculating, domineering, and angry and African-American men as dangerous. While the stereotype of African-American men as hyper-masculine and sexually aggressive fuels the perception *121 that they are threatening and dangerous, these are also traits that some straight women (and gay men) find appealing.

Many college students in the California public university study expressed concern that family members and society in general would not approve if they dated African-Americans. Another study of White college students' racial attitudes similarly found that they feared family and societal disapproval if they married interracially. Interestingly, Asian-Americans and Latinos/as were significantly more likely than Whites to cite social disapproval as a reason to exclude African-Americans as romantic partners. The frequency of these concerns varied by gender. Asian-American and Latina students were significantly more likely than their male counterparts to express concern that parents, friends, and strangers would disapprove and they feared they would be discriminated against if they dated African-American men. These concerns are not unfounded. Interracial couples face greater opposition and disapproval from family members and society than same-race couples.

Parents' objections to their children's interracial relationships confirm the racial hierarchy apparent in the dating market. Asian-American, Latino, and White parents all express greater objections to their children intermarrying with African-Americans as compared to other groups. They express fear that society will discriminate against their adult children and mixed-race grandchildren, and also express concern about the racial identity and psychological well-being of mixed-race grandchildren. Although not always expressly stated or acknowledged, parents also fear their own potential loss of status. One study found that Latino parents express disapproval of intimacy with African-Americans even before their *122 children start dating because they fear jeopardizing the family's status in the racial hierarchy. Other studies have found that White parents are similarly concerned about the loss of status for the family, especially when the child marries an African-American partner.

Parents' objections to children's interracial relationships reflect not only a racialized hierarchy, but also a gendered one. Their reactions to the relationship depend not only on the race of the child's partner but also the gender. Families are much more likely to express strong disapproval when daughters (as compared to sons) date or marry out. For example, White women in interracial relationships experience greater disapproval than White men dating minority women or minority men dating White women. Latino parents are similarly more likely to express opposition when daughters (as compared to sons) date African-Americans.

Societal disapproval of interracial relationships also depends on the race and gender of the minority spouse. Numerous commentators have noted greater objections from both African-Americans and Whites to relationships between African-American men and White women as compared to relationships between African-American women and White men. In fact, a 2005 Gallup poll found that while 72% of Whites approve of a White man dating an African-American woman, only 65% approve of an African-American man dating a White woman. White women married to Asian-American men also experience greater objections from *123 both the White and Asian-American communities than Asian-American women married to White men.

Racial preferences are problematic for many reasons. They undermine our commitment to anti-discrimination and perpetuate social distance between groups. They also affect marriage outcomes. The two groups least preferred by online daters--African-American women and Asian-American men--are also the groups with the lowest rates of intermarriage. For African-American women, racial preferences affect not only their rate of intermarriage, but their likelihood of marrying at all and raising a child without a co-parent.

Marriage rates and non-marital birth rates vary significantly by education. College-educated women are more likely to marry than women with lower levels of formal education. They are also significantly less likely to have children outside of marriage. However, African-American women are much less likely to marry than women of other races and are also more likely than women of other races to have non-marital children and to raise them in single-parent households. While two-parent households are not superior to single-parent households, in the United States children raised by single parents are less advantaged in myriad ways, even when the single-parent has financial resources. The next section will briefly describe these relative disadvantages.