Race, sex, religion, color, national origin, age, disability, and veteran status are all what are called protected classes under federal law — characteristics that cannot be used as the basis for discrimination in hiring, housing, or other arenas. There are loopholes, however; one is that it is acceptable to discriminate based on a protected characteristic if you can show that it is “bona fide occupational qualification” (BFOQ). So, for instance, if you can show that being female is a legitimate requirement for being able to perform a particular job, you can refuse to hire men. Hooters used the BFOQ argument when they were sued for sex discrimination because they would not hire men as servers.

The exceptions are race and color, which are not legally seen as ever being legitimate qualifications for doing a job. As the Equal Employment Opportunity Commission website states, “Nor may race or color ever be a bona fide occupational qualification under Title VII.” That is, there is absolutely no good reason that being of one race or another would ever be a legal basis for hiring.

And yet, there’s still at least one arena where race is blatantly and openly used as a basis for hiring: Hollywood casting. Back in 2006, Russell Robinson, a faculty member at the UCLA School of Law, looked at the sex and race/ethnicity characteristics specified in “breakdowns” — the summaries of characteristics presented in casting announcements. As Robinson explains in the article “Casting and Caste-ing: Reconciling Artistic Freedom and Antidiscrimination Norms,” his sample certainly doesn’t include all roles in the process of being cast during that period. Roles aimed at big stars who don’t go through the typical audition process may never be released as a breakdown, since there’s no intent to recruit for the role. But

Robinson’s team looked at all breakdowns for feature films released between June 1 and August 31, 2006, excluding calls for extras and stunt people. As they reported in the research brief “Hollywood’s Race/Ethnicity and Gender-Based Casting: Prospects for a Title VII Lawsuit,” the vast majority of the breakdowns explicitly state the race of the character, with only 8.5% of roles open to any race/ethnicity:

Notice that African Americans and Latinos are particularly under-represented compared to their proportion of the total U.S. population. And while 22.5% of breakdowns specifically said the character should be White, almost half included language that designated the role as implicitly White — for instance, including only White actors in a list of prototypes for the role. In fact, interviews with casting directors indicate that roles are presumed to be White unless the breakdown specifically says otherwise.

Almost all breakdowns specified the sex of the character; 59% of the breakdowns specified the role was for a man, while 35% of roles were for women.

Robinson also analyzed the cast of 171 films released in 2005 that made at least $1 million. The majority of all roles were reserved for men. An overwhelming 73% of leads were men, and even supporting roles were predominantly for men:

Of the leads in those films, 81.9% were White non-Hispanic:

Robinson’s work shows that Hollywood still explicitly uses protected classes in hiring decisions, including race/color, which have been excluded from the BFOQ loophole. For more on this, see our posts on race and roles in recent trailerscasting Whites in Asian roles, Hollywood’s discomfort with Asian lead roles, gendered positioning in promotional posters, race and representation in Hollywood, the Smurfette Principle in movies, who goes to see movies, anyway?, Anita Sarkeesian on male-centric plots, and the lack ofra African Americans on Friends.

Thanks to Dolores R. for the tip about Robinson’s study, which she originally saw at Racialicious.

The U.S. Census Bureau has started releasing data from the 2010 Census. This map shows the change in the racial/ethnic minority (i.e., anything other than non-Hispanic White) population over the last decade:


They released a report, An Overview: Race and Hispanic Origin in the 2010 Census (available here), which includes data on those who reported more than one race. Among those who reported more than one race, the vast majority listed two. Here are the most commonly reported combinations:

AIAN = American Indian/Alaska Native, NHPI = Native Hawaiian/Pacific Islander, and SOR = some other race.

Laura E. pointed out that New Geography posted some maps based on 2010 Census data. Here’s the Hispanic population as a percent of the total population, by county (notice that the legend need to be multiplied by 100 to get the percent):

The African American population (alone or in combination with another race, and again, multiply by 100):


Jamie Keiles, of  the Seventeen Magazine Project and Teenagerie, wondered how racially representative Seventeen was compared to the U.S. population.  So she offered some data based on the May 2010 issue.  Her methods:

There are 332 faces in this month’s issue of Seventeen. I counted a face as a head with at least one visible eye. That is, backs of heads and disembodied mouths or eyes were not included in my data. I researched the races of the models and celebrities that I could identify. Those whose race I could not determine with reasonable certainty I’ve excluded from my data, making for 319 surveyed faces.
She found that Seventeen was not representative (it was nine percent more white than the U.S. population and especially under-represented Hispanics).  Still, she concluded that it was surprisingly representative, considering what she’d heard about the modeling industry.  Her findings actually reflect Ashley Mears’ argument that there is much more diversity among “commercial” models than “high end” fashion models.

Keiles was also surprised by the fact that, compared to the U.S. population, there were many models who identified as bi-racial.  My guess is that it’s because advertisers think (and perhaps know, but I’m not sure) that models whose identities are hard to discern appeal to a larger array of audience members who may see themselves in what is otherwise an “ambiguous” appearance.

Any ideas as to why white Hispanics are particularly underrepresented?  Is it possible that white Hispanic models simply identify publicly as “white”?  Other ideas?

Keiles finds a similar patterns when she looks by gender and by whether it was Seventeen content or advertiser content:

Lisa Wade is a professor at Occidental College and the co-author of Gender: Ideas, Interactions, Institutions. Find her on TwitterFacebook, and Instagram.

Different countries formalize different racial categories.  Below are examples of the “race” questions on the Censuses of 9 different countries.   They illustrate just how diverse ideas about race are and challenge the notion that there is one “correct” question or set of questions.





The categories in the drop down menu include:

South Asian
Black Filipino
Latin American
Southeast Asian
West Asian







The categories in the drop down menu include:

Black, African Am., or Negro
American Indian or Alaska Native
Asian Indian
Other Asian
Native Hawaiian
Guamanian or Chamorro
Other Pacific Islander
Some other race

These images were borrowed from an American Anthropological Association website on race.

Lisa Wade is a professor at Occidental College and the co-author of Gender: Ideas, Interactions, Institutions. Find her on TwitterFacebook, and Instagram.

In this series I have offered five explanations of why people of color are included in advertising. Start with the first in the series and follow the links to the remaining four here.

I am now discussing how they are included.  Already I have shown how people of color are whitewashed. Here I show that, when people of color are included, they are often chaperoned. That is, people of color are usually outnumbered by white people. Here are some examples:

NEW!  This is a two-page ad for Kohls:


I speculate that, if there are more white people than people of color in an advertisement, the inclusion of a non-white person does not threaten the status quo (that whiteness is mainstream and normative) and the product is still clearly marked as mainstream and normative (i.e, white).

NEW (Mar. ’10)!  Caitlan V.d.W. sent in a link to Fergy’s friends paper dolls.  Notice that Fergy has 3 white friends and one black friend.  Anti-racist and also safe!


Next up: Subordination.

Also in this series:
(1) Including people of color so as to associate the product with the racial stereotype.
(2) Including people of color to invoke (literally) the idea of “color” or “flavor.”
(3) To suggest ideas like “hipness,” “modernity,” and “progress.”
(4) To trigger the idea of human diversity.
(5) To suggest that the company cares about diversity.

How are they included?
(6) They are “white-washed.”

I just discovered the entirely excellent website Asian Nation, run by C.N. Le and full of great information about the Asian American community. Here are some tables showing what percent of various Asian American groups are married to spouses of the same or other groups, updated as of October 2007 using Census data (an explanation of the three columns follows):

Ok, now to explain the three columns of numbers. The first one presents data for all marriages that include at least one Asian American spouse–this will include large numbers of immigrants who were married before they moved to the U.S. The second column includes only those marriages where at least one spouse was raised in the U.S., defined as either born here or moved here by age 13. The third column includes only those marriages where both spouses were raised in the U.S. According to Le, this group represents less than 25% of all marriages including an Asian partner, but “…has the advantage of including only those who were raised and socialized within American society and its racial dynamics. It is this U.S.-raised population that best represents young Asian Americans, since they are the ones who have the most exposure to prevailing American cultural images and media.”

Not surprisingly, endogamous (in-group) marriage rates drop off significantly among U.S.-raised Asian Americans. There are other interesting gender patterns as well. Notice, for instance, that Chinese, Vietnamese, Korean, and Filipina women are quite a bit more likely to be married to a White partner (the most common out-group spouses) than are men, and for the remaining groups, women are slightly more likely to be married to a White spouse. You might discuss the social and historical factors that might cause that pattern, and compare it to the trend in marriages with a Black and a White spouse, in which the gender pattern is usually reversed–Black men are more likely to be married to Whites than are Black women. It might also be worth noting that Korean and Filipina women are significantly less likely to marry endogamously than the other Asian American ethnic groups.

In honor of the election, we offer you a summary of all our election 2008 posts.

This election has certainly brought racial tensions front and center. We highlighted two racist caricatures of Obama: on a waffles box and as a cannibal. We also discussed the cover of The New Yorker on which Barack and Michelle Obama were caricatured as terrorists. Whether or not this was racist was widely discussed and offered an interesting opportunity to ask “Who decides what we talk about?” In response to the argument that we were being too sensitive about the caricatures, we offered some evidence that caricatures of black people do not need to be racist.

Anti-Obama propaganda also included comparison with OJ Simpson, a monkey, celebrities, Osama Bin Laden, fascists and communists, a terrorist, a terrorist again, and a “half-breed Muslin.” See here for other racist anti-Obama propaganda.  Gwen asked “So what if Obama is an Arab?” (Note, too, this satirical T-shirt.)

We saw racialization–or the active production of racial meaning–in the fist bump controversy, in calling Michelle Obama a “baby mama,” and in asserting the whiteness of the White House. We discussed the resemblance between Obama and his Grandfather and the meaning of “Main Street” to illustrate the social construction of race.  And we offered examples of white privilege: in one we discuss the option of white ethnics to emphasize their ethnicity; in two we discuss a cartoonist who calls Colin Powell a race traitor for endorsing Obama and a Howard Stern clip that suggests that Blacks only endorse Obama because he’s Black.  We also remark on how easy it is to deride social theories of inequality.

The McCain/Palin ticket was no stranger to derision.  See also our post in which the McCain/Palin ticket is said to be favored by Nazis, another in which Palin effigy is lynched, and a third that discusses ageism in the election.

We’ve also seen plenty of sexism in this campaign. Hillary Clinton has been represented as a nut buster, asked to “iron my shirt,” critiqued for crying, and called a “bitch.” There are more examples here and here.  Also see this montage of sexism among political pundits. Both Hillary Clinton and Sarah Palin were sexualized. See here, here, and here for Clinton and here, here, here, here, and here for Palin. (By the way, Barack Obama was sexualized as well, see here, here, and here.)

We commented, more sociologically, on the gender politics of this election. We discussed the mothering of baby Trig, conservative feminism, the politics of pink, and took a humorous look at the women’s vote with Sarah Haskins.

We also pointed to the way in which Obama and Clinton attempted to appeal to small town people and the ease with which we make fun of them.

For the intersection of race and gender, see our post in which Michelle Obama is called an angry black woman, is said to need to “soften” to be a First Lady, and our post that features the Bros Before Hos T-shirt (scroll to the bottom). For the intersection of race and class, see our post on Obama’s negotiation of the “elitist” label.  And, in making intersectionality invisible, see the SNL skit, “bitch is the new black.”

Looking more broadly at politics and media coverage, we discussed the portrayal of evil in the Reverend Wright scandal, McCain’s trivialization of war, the linking of a Democratic adminstration with a terrorist attack, pundit hypocrisy, political networks, a voter registration campaign that uses bondage imagery, suspiciously delicious polling techniqueshow cell phones shape polling findings, and trends in media coverage of Obama versus Clinton and Obama versus McCain.

In addition, we offered some examples of punditry from alternative media: on young voters, a call for alliance from the labor movement, a call to get your Jewish grandparents to vote for Obama, a political revival of the Budweiser Wassap video, and two examples of art inspired by the election (here and here).

We also put up posts of figures representing public opinion on blacks, a woman president, and politician parents.  And we offered images illustrating how the world would vote.

Finally, our favorite: “We’re not sociologists, we’re Americans!”

Below are two photos — one of Barack Obama as an adult and one of a young Obama and his Grandfather, Stanley Dunham (found here and here).  I tried a little experiment in class. I put up the photo of adult Obama and I had my students make a list of what characteristics made him identifiably Black, in their view. Every one of them put on their list his nose, lips, and hair, and several made comments about his ears or just that “the combination of all his facial features” was “clearly” Black.

Then I brought up the photo of young Obama and his grandfather alongside it:


It led to a really interesting discussion. Because my students think of Obama as Black, they saw all his features through that racial lens. It was obvious to them that he had “Black” facial features. After viewing the photos next to one another, they talked about how the two men look very similar, but their facial features seem “clearly” Black on one person and “clearly” White on the other because we’re used to believing that Blacks and Whites look very different. Because they believe in racial differences, they see them. This activity seemed to really help students grasp what I meant when I talked about the way we identify things as racial differences when they’re really variations that occur in many different groups that we swear are physically distinct from one another.

It’s interesting because I’d halfway expected my students to argue that Obama doesn’t “really” look Black at all — that they would say his skin color and hair, perhaps, were identifiable as African American, but that they would point out that he’s half-White and argue that in fact he doesn’t have stereotypically Black or White features. I mean, that would made sense, right? But I’ve tried this with three classes now, as well as several random individuals I’ve subjected to it to see if it might work as a class activity, and no one has yet failed to identify a number of the facial features Obama and Dunham share as specifically African-American features when they see them on Obama.

What a great example of racialization and the social construction of race.

See also our Pinterest pages: what color is flesh? and the social construction of race.

Gwen Sharp is an associate professor of sociology at Nevada State College. You can follow her on Twitter at @gwensharpnv.