Social dominance theory posits that although some members of dominant groups will devote considerable energy and resources to making the world more equitable, dominant groups as a whole tend to act in ways that maintain their advantage. As the comedian Mel Brooks said in his satirical film History of the World, Part I, “It’s good to be the king.” Efforts to perpetuate the existing system, as we noted earlier, take the form of individual discrimination, institutional discrimination, and behavioral asymmetries. When we combine social dominance theory with some of the specific tenets of the stereotype content model, we can gain insight into exactly how social and economic imbalances are preserved.
Instances of active prejudice and discrimination against members of nondominant groups tend to command our attention (and receive the most media scrutiny) and are therefore the most familiar to most people. We therefore begin this section by discussing some of the most common forms of prejudice and discrimination that involve acts of commission—that is, individuals engaging in actions that disadvantage or harm members of certain groups. We then discuss ways in which members of certain groups are disadvantaged through omissions—the absence of things, such as resources, opportunities, or mere attention, that are available to members of dominant groups.
In the aftermath of the murder of George Floyd by Minneapolis police officer Derek Chauvin, awareness of police misconduct against people of color has increased. Figure 11.3 presents data on the demographic breakdown of police killings in the United States, showing that Black American, Indigenous American, and Hispanic individuals are killed by police more often than members of other groups are. Although the killing of Black Americans by police has gone down steadily over the past 50 years, Black Americans have nevertheless been the victims of 26 percent of recent police killings, although they make up only 13 percent of the U.S. population. Indigenous Americans make up only 0.8 percent of the population but are the victims of 1.9 percent of police killings (Males, 2014).
These are pure demographic statistics, of course, so their interpretation must be guided by the usual caution accompanying any correlational result. But social psychologists go beyond such demographics to try to understand the extent to which they reflect police bias by both conducting finer analyses and carrying out controlled experiments. For example, an analysis of policing in Oakland, California, found that 60 percent of police stops involved Black Americans even though only 28 percent of the population of Oakland is Black—a result that, crucially, held after controlling for more than two dozen variables potentially related to the decision to stop someone, such as the crime rate and demographic composition of the specific location where the stop occurred (Hetey et al., 2016). In an even more telling sign that some police interactions with Black individuals might start off in a biased fashion before ending in tragedy, an analysis of the speech recorded from body cameras revealed that law enforcement officers, both Black and White, used more respectful language (“Have a safe night”; “Sorry to stop you”) when speaking to White motorists than when speaking to Black motorists (Voigt et al., 2017).
Social psychologists have also utilized their favorite tool, the randomized controlled experiment, to examine the extent to which preexisting stereotypes of Black Americans as more likely to cause “trouble” or engage in crime might channel police officers’ or other establishment figures’ interactions with Black people down a troublesome path. In one study, educators watched a videotape of preschool children engaged in typical classroom activities. Some were told to expect to see some challenging behavior on the part of some of the students; others weren’t given this instruction. Those led to expect trouble spent more time looking at the Black children, especially the Black boys, than those not told to look for signs of trouble (Gilliam et al., 2016).
In another controlled experiment that highlighted the biasing effects of the stereotype that links Black people with crime, participants were shown a series of degraded images of various objects, starting with one so fuzzy that it was nearly impossible to decipher and then gradually proceeding to clearer images. Some of the images, like the gun shown in Figure 11.4, were crime-related, and others were not, and the participants’ task was to signal as soon as they recognized what the gradually clarified image depicted.
FIGURE 11.4 STEREOTYPES AND BIASED PERCEPTION (A) A degraded image of an object (Frame 1), gradually becoming clearer (Frames 20 and 41). How quickly would participants realize that the object, in this case, was a gun? (B) This graph shows how long it took participants to recognize the object (i.e., the frame number when they indicated that they knew what it was) when it was related to criminality or not and when the presentation of the object was preceded by a series of White faces, Black faces, or a non-face line drawing. The participants recognized the crime-related objects more quickly when they had just seen a series of Black faces.Source: Adapted from Eberhardt et al., 2004.
DATA EXPLORATION
Before doing so, participants were primed with a series of Black faces, White faces, or meaningless line drawings. As you can see in Figure 11.4, being primed with Black faces led the participants to detect the crime-related objects, but not the other objects, more readily (Eberhardt et al., 2004). The Black/crime stereotype thus guided participants’ attention—a result that is unlikely to come as a surprise to most Black people. This finding has implications for how Black people’s interactions with others, including those with the police, can play out.
Members of marginalized groups also aren’t treated as well as dominant-group members are in the workplace. When researchers submit applications for job openings—some with stereotypically Black names (Lakisha, Jamal) and others with stereotypically White names (Emily, Greg)—the former receive fewer follow-ups for interviews than the latter do. In one of the earliest and best-known studies of this type, applications with stereotypically White names received 50 percent more callbacks than those with Black names did (Bertrand & Mullainathan, 2004). Subsequent studies have established that some of this difference is due to the stereotypically Black names being associated with lower social socioeconomic status (Simonsohn, 2016), itself a source of hiring bias (Rivera & Tilcsik, 2016).
Other studies using this experimental paradigm have uncovered bias against applicants with foreign-sounding names in Canada (Oreopoulos, 2011), Turkish-sounding names in Germany (Kass & Manger, 2012), and Chinese, Italian, Middle Eastern, and Indigenous Australian names in Australia (Booth, Leigh, & Varganova, 2012), as well as bias against applicants in the United States who listed Muslim student organizations on their résumés (Wright et al., 2013). In contrast to the consistent evidence of bias against members of racial and ethnic minority groups, attempts to assess gender bias and bias against members of sexual minority groups have yielded mixed and equivocal evidence of discrimination (Bertrand & Duflo, 2016; Quadlin, 2018), with some studies actually indicating preferential treatment for women (Veit et al., 2022; W. M. Williams & Ceci, 2015).
When we think about prejudice directed at members of particular groups, we usually think of active mistreatment—using ethnic slurs, for instance, or being more likely to suspend a Hispanic student from school or to pull over a person of color in a traffic stop. But harm can also result from what people don’t do—from omissions as well as commission. This is clearest when it comes to attention. It’s painful when people don’t grant us their attention (K. D. Williams, 2007), and members of marginalized groups often find that the world does not pay as much attention to them as it does to members of more dominant groups. We saw in Chapter 9 that unattractive people struggle to be recognized, as do obese individuals, older adults, and people with disabilities. Not only do individuals who belong to these groups have to live in a world where they don’t attract as much attention as their peers do, but they also must watch as others actively turn their attention away from them. Former U.S. senator Robert Dole experienced this vividly after returning from Germany in World War II with severe injuries that left him without the use of his right hand and with only limited use of his left. When he attended events in uniform, people treated him as the war hero he was; when he went out in public without his uniform, people treated him as an invalid, often turning away from him.
You are well aware that attitudes toward Black people in the United States have changed over the past two centuries—and are changing still. But did you know that Americans have had many ethnic prejudices that have been greatly mitigated or eliminated over time?
Beginning in the middle of the nineteenth century, Irish people—who were treated little if any better by the British than Black people were treated in the United States at the time—began coming to the United States by the millions to escape the hunger caused by the potato famine. They had not been here long before extreme stereotyping and prejudice began. The Irish had little schooling; fewer than half of adults could read. Alcoholism and spousal abuse were common among Irish men, knife fights were ordinary occurrences, and people lived in squalor in the cities, often sharing their house with their cattle. Houses owned by Irish people were called “shanties.” The slang term “Paddy wagons” (for police vans) was coined from “Paddy,” a pet name for an Irish person called Patrick. Depictions of Irishmen in newspaper cartoons often looked more like apes than humans. Well into the twentieth century, it was common for job postings to include the sentence “No Irish need apply.” Sometimes, there was an addition: “Colored man preferred.” And in fact, free Black people in northern cities were preferred to the Irish both as employees and neighbors. Over time, however, the lives of the Irish improved. But it wasn’t until around the time that John F. Kennedy was elected president (the first Irish American to hold that office) that the old stereotypes largely disappeared. There’s a good chance that a number of your classmates are of Irish descent. This would likely be of little import or interest to you; indeed, it might even be of little import to your Irish American classmates.
The story of the Irish has been repeated many times. In the late nineteenth century, the Irish, along with the Italians, Poles, and Jews, were not regarded as White by many people. (A book published in 1995 had the title, written only slightly tongue in cheek, How the Irish Became White.) In the Upper Midwest, the phrase “dumb Swede” was heard in every community until at least the middle of the twentieth century. Do the Swedes’ descendants today know about this?
Was every group of immigrants treated so badly? Not at all. Germans, English, and the Dutch, mostly middle class, were quickly stirred into the melting pot. But remember that the Statue of Liberty stands next to Ellis Island in New York, where millions of immigrants entered the country. There is a poem by Emma Lazarus at the base of the statue. It does not say, “Welcome all you cultivated, skilled people.” The poem reads, “Give me your tired, your poor, Your huddled masses yearning to breathe free.” A lot of people so described came to the United States, and they turned themselves into prototypical Americans. It sometimes just took a while.
The pain of others’ inattention is familiar to people with “intersectional” identities—that is, those who have multiple nondominant identities. As we noted in Chapter 10, no one is simply Asian: One is an Asian man, woman, or nonbinary person; upper, middle, or working class; straight, gay, bisexual, or asexual; and so on. Valerie Purdie-Vaughns and Richard Eibach (2008) have argued that individuals who have multiple nondominant identities tend to be relatively invisible to others because they are not prototypical members of their different identity groups, and people tend to think in terms of prototypes. This is reflected in the Black feminist slogan that captures so much of Black women’s experience trying to promote social justice: “All the women are White and all the Blacks are men.” The voices and images of Black women, in other words, tend to be especially underrepresented.
In one study that illustrates this tendency, participants were first shown photos of the faces of Black and White men and women. After a brief delay, they were then shown all the photos a second time, interspersed with photos they hadn’t seen (foils), and asked whether they had seen each photo before. Consistent with the idea that intersectional individuals can be relatively invisible, the photos of Black women were significantly less likely to be recognized than any of the other types of photos. A follow-up study examined participants’ memories of the comments made in a simulated group discussion and similarly found that participants were less likely to correctly tie the comments made by the Black women to the women who said them (Sesko & Biernat, 2010). Studies like these lend credence to intersectional individuals’ reports that they feel more invisible to others and that they have received more discriminatory treatment (at work and in stores and restaurants) because of their identities (Remedios & Snyder, 2018).
Members of marginalized groups also face challenges that stem not from direct mistreatment by others but from systemic inequities that can make life seem like an unending uphill struggle (Salter, Adams, & Perez, 2018). When a severe heat wave struck Portland, Oregon, in the summer of 2021, for example, the ambient temperature in the poorer neighborhoods of East Portland, which are populated mostly by people of color, was 9 degrees Fahrenheit (5 degrees Celsius) hotter than the city average and 25 degrees Fahrenheit (14 degrees Celsius) (!) hotter than the leafier, more upscale neighborhoods of Northwest Portland (Peel, 2021). Parks are more likely to be located in higher-income areas of a town, and highways and industrial sites are more common in lower-income areas, making the latter neighborhoods much warmer during heat waves. Indeed, a systematic analysis of summertime temperatures in urban areas across the United States found that surface temperatures were, on average, 4.7 degrees Fahrenheit (2.6 degrees Celsius) higher in lower-income neighborhoods (Hoffman, Shandas, & Pendleton, 2020). As a consequence, when a heat wave strikes today, the poor have to ride it out in a substantially hotter environment than their wealthier neighbors do.
This is but one example of the pervasive, often less visible types of discrimination that members of marginalized groups confront. The legal system is rife with examples of white-collar crime, which tends to be committed by members of the dominant groups in society, being punished much less severely than crimes committed by members of marginalized groups. The substantially different penalties for possession of crack versus powder cocaine mentioned earlier are but one example. Consider also the statistics presented in Figure 11.5. Although White Americans outnumber Black Americans by a wide margin both in the general population and in the ranks of current or recent drug users, as many Black as White individuals are sent to state prison for their drug offenses, and more Black than White individuals are sent to federal prison (Hinton, Henderson, & Reed, 2018). Again, these are pure demographic statistics, so they need to be interpreted with caution. Nevertheless, the argument that these results may be the result of systemic bias in the legal system is supported by data showing that although Black Americans make up only 13 percent of the population, they represent 47 percent of those exonerated for wrongful convictions (Equal Justice Initiative, n. d.).
FIGURE 11.5 DRUG USE AND PENALTIES FOR DRUG USE FOR WHITE AND BLACK AMERICANS (A) White people outnumber Black people in the general population, (B) in the number of people who use illegal drugs, and (C) in the number of people arrested for drug possession and distribution. Nevertheless, (D) as many Black people as White people are sentenced to state prison for drug offenses, and (E) more Black people than White people are sentenced to federal prison for drug-related crimes. Source: Adapted from Hinton et al., 2018.
Members of dominant groups in society also enjoy systemic financial advantages. To support this point, one need look no further than the fact that the highest tax rates on labor (that is, on time and energy exerted in performing of one’s job) are higher than the tax rates on capital gains (that is, on the money passively earned by having money in the bank or owning stock in companies). These tax rates work disproportionately to the benefit of wealthy people, for whom relatively more of their income is in the form of capital gains. Of course, many wealthy people benefit further by employing accountants and tax lawyers to reduce their already low tax rate even more. Changes in the tax systems in countries around the world are responsible for much of the widening gap between rich and poor, as the ratio of money going to capital versus labor has increased markedly in recent decades (Piketty, 2014).
The election of Barack Obama as the 44th president of the United States highlights the often-ambiguous nature of race. Although the child of a White mother and a Black father, Obama is almost always referred to as the first African American president, not the first biracial president. This is no doubt a legacy of the “one-drop rule”: Historically, individuals were considered Black if they had any Black ancestry at all. Various Southern states used this standard to back the notorious Jim Crow laws that enforced racial segregation and restricted the rights of Black Americans. But now, with society having moved beyond the one-drop rule, we are left with the difficult issue of “who counts” as Black, White, Asian, Hispanic, and so on. Indeed, many biologists question whether racial categories make any sense at all—that is, whether race really exists (Bamshad & Olson, 2003). Genetic analyses enabled by the Human Genome Project have shown that there is as much genetic variation within a population in, say, Asia as there is between that population and one in, say, Europe.
The psychology behind the one-drop rule notwithstanding, race-based judgments about others often differ in intensity depending on the degree to which a person’s physical features conform to a stereotype. Black faces with more stereotypically African features (darker skin, fuller lips, more flared nostrils) elicit prejudiced reactions more readily than faces with less stereotypical features do (Livingston & Brewer, 2002; Ma & Correll, 2011). Furthermore, both Black and White individuals with more stereotypically African features are assumed to have traits associated with common stereotypes of Black Americans (I. V. Blair et al., 2002). In the most consequential manifestation of this tendency, both Black and White convicts with stereotypically African features tend to receive harsher sentences than those with less stereotypically African features do (I. V. Blair, Judd, & Chapleau, 2004), and Black people accused of capital crimes are more likely to end up on death row if they have stereotypically African features (Eberhardt et al., 2006). Moreover, testifying to the utility of the motivational perspective on stereotyping and prejudice discussed in Chapter 10, people are more likely to categorize ambiguous faces as Black when they are primed with thoughts of economic scarcity (Krosch & Amodio, 2014; Rodeheffer, Hill, & Lord, 2012).
The list of institutional disadvantages faced by members of marginalized groups could go on and on. For instance, people in predominately non-White communities in the United States have to wait in line to vote nearly a third longer than people in predominately White communities do, and they are 74 percent more likely to have to wait for more than half an hour (M. K. Chen et al., 2019).
But this is a psychology textbook, not a sociological or legal text, so let’s consider a more fundamentally social psychological example of advantage (to men) and disadvantage (to women) in mass media. An analysis of photographs of men and women in U.S. magazines and newspapers revealed that the person’s face was more prominent in the photographs of men (that is, the face took up more of the overall photograph) than in the photographs of women (which tended to include more of the women’s bodies). The same was true in publications from 11 countries around the world, in the artwork produced in over six centuries of art history, and in college students’ drawings of men and women (D. Archer et al., 1983; Nigro et al., 1988). This difference works to men’s advantage: People are seen as more competent when their faces are more prominent in photographs (Schwarz & Kurz, 1989).
Just as members of marginalized groups can feel the sting of inattention from other individuals, so too can they suffer from a more pervasive inattention and underrepresentation in society at large. The idea is embedded in the very term underrepresented group, which connotes an absence. There would be no need for such a term if members of marginalized groups were represented in boardrooms, Hollywood studios, and academic faculties in proportion to their demographic numbers. In reality, most marginalized groups in the United States are underrepresented in the ranks of CEOs and NFL coaches and in the clinical trials that determine recommended medical treatments. And their voices are also underrepresented in the academic journals that report the research we have drawn on in writing this book (S. O. Roberts et al., 2020).
This sort of underrepresentation is even more pronounced in most other parts of the world. Consider this (admittedly extreme) example: The Golden Globes have been around since 1943, handing out awards for the best that film and television have to offer as determined by the 87 members of the Hollywood Foreign Press Association. The Golden Globe Awards ceremony was a major event in the film industry’s awards season until it lost its television sponsor in 2022—after it became known that not a single member of its voting body was a person of color. Talk about underrepresentation.
This kind of absence, which can lead to disidentification with society and its institutions more broadly, is particularly pronounced for Indigenous people in the United States (Fryburg & Eason, 2017). An analysis of prime-time television shows from 1987 to 2008 found only three characters (among more than two thousand) who were Native American (Tukachinsky, Mastro, & Yarchi, 2015). Things aren’t much better outside the world of entertainment: A majority of the annual proclamations made by the president of the United States to commemorate Thanksgiving Day have failed to make any mention of Indigenous people, even though a meal shared by the Pilgrims and the Wampanoag people is such a big part of the holiday’s origin story (Kurtiş, Adams, & Yellow Bird, 2010).
And don’t forget the 1979 Church Rock uranium spill. Actually, you probably can’t forget it because you’ve probably never heard of it. Four months before the widely covered Three Mile Island nuclear accident, the largest release of radioactive material in U.S. history took place in Church Rock, New Mexico, contaminating the groundwater and an essential river used by the Navajo Nation. But in part because the damage was done to the Navajo Nation and did not, like the Three Mile Island accident, threaten residents of Harrisburg, Pennsylvania’s capital, it did not attract much press attention and is largely absent from the history books. Some difference in press coverage is to be expected because Three Mile Island is located in a more populous area. But the extent to which the Church Rock uranium spill was underreported is striking.
Being largely absent from the broader culture has psychological consequences. Native American students who are exposed to Native American role models report feeling that they belong more in school than do Native American students who aren’t exposed to such role models. In contrast, European American students who are similarly exposed to role models of European heritage do not experience a comparable boost in feelings of belonging, presumably because European American role models are already so present in their schools (Covarrubias & Fryberg, 2015). Over time, the chronic omission of Native American role models in education can lead Native American students to disidentify with school altogether, perpetuating educational disparities between Native and European American students.
As we noted earlier, the cultural invisibility that is a big part of the Native American experience is also a frequent fact of life for people with intersectional identities. Note that a multiracial category was only added to the U.S. Census in 2000, meaning that, before then, a half-Black, half-Asian woman would have had to choose one race or the other. Her intersectionality was simply not recognized—demographically invisible, in other words. Recall from earlier in the chapter that Purdie-Vaughns and Eibach (2008) have posited that intersectional individuals tend to be relatively invisible because they aren’t prototypical members of any of their constituent identity groups. The authors called attention to this source of invisibility through what they referred to as the librarian’s dilemma: Imagine that you are a librarian, and your library receives a book on Black women’s history. Where do you put it: in the Women’s History section or the Black History section? You have to choose one or the other, and your choice makes part of the book’s contents hard for patrons to access—invisible, in other words.
Purdie-Vaughns and Eibach discussed how people with intersectional identities are treated as marginal members of marginal groups and thus tend to be relatively invisible in historical narratives (histories of slavery feature men more than women), in political discourse and advocacy (Black women’s concerns can get lost in both women’s movements and racial justice movements), and in the eyes of the law (plaintiffs who bring intersectional claims of discrimination don’t fare as well in the courts as do those who claim a single form of discrimination; Best et al., 2011).
Of course, there are times when invisibility can be a blessing rather than a curse, and, accordingly, there are times in which intersectional individuals benefit from not being a prototypical member of either of their identity groups. For example, because Black women are not seen as prototypical of either the category of women or of Black people, there can be less resistance to their acting in ways that depart from the stereotypes of those groups. Black women can display more dominance than White women or Black men without paying a price in terms of how likable or hirable they seem to others. In one study, participants were shown a photo of a White or Black man or woman and told that he or she was a senior vice president of a Fortune 500 company (Livingston, Rosette, & Washington, 2012). Participants were further told about a meeting between the vice president and an employee who had performed below expectations. Half were told that the vice president displayed dominance in the meeting, saying such things as “I demand that you take steps to improve your performance.” The other half were told that the vice president struck a more communal tone, saying such things as “I encourage you to take steps to improve your performance.” Participants then rated how well they thought the vice president handled the situation and how effective the vice president was as a leader. The White male vice president was rated highly whether he was dominant or communal, whereas the White female vice president was seen as significantly less effective as a leader when she acted in a dominant manner. Notably, however, the Black female vice president did not pay the same price in perceived effectiveness when she acted assertively (see Figure 11.6).
When you think of politicians, CEOs, or film directors, who comes to mind? You shouldn’t fault yourself too much for being sexist if you mostly thought of men. After all, 75 percent of members of parliaments around the world are men (UN Women, 2021), as are 94 percent of CEOs of S&P companies and 93 percent of the directors of the 250 top grossing films (Catalyst, 2022; Hunt et al., 2017). The instances that spring to mind most readily are often those we’ve been exposed to most often. That can make gender inequality self-fulfilling, as boys may have an easier time seeing themselves in those roles and therefore may be more inclined to take the steps necessary to achieve them. This increases the chance that such gender imbalances will be perpetuated in the future.
But as we saw in Chapter 4, other things beyond real-world frequency determine which instances of different categories spring to mind. One such determinant is language. Many languages are androcentric, often using masculine terms (such as “mankind” or even “humanity”) to refer to everyone. The use of such generic masculine language, as opposed to the use of gender-inclusive terms, makes it more likely that people will think of a man as representing the category in question (Bigler & Leaper, 2015). The language we use, in other words, has consequences: It makes it more likely that some things will be present and other things absent in our minds, which can have the effect of empowering some groups and disadvantaging others.
A recent study of language’s impact in influencing who comes to mind and who doesn’t took advantage of the fact that, in 2016, Yale University changed the title of an academic/residential leadership role from master to head (Bailey, Dovidio, & LaFrance, 2022). The investigators noted that master can be considered a generic masculine term (it has a feminine counterpart of mistress), whereas head cannot (it has no feminine counterpart). Did the name change influence how the students thought about the individuals in that leadership role? It did indeed. When students in 2015 were asked to think of a typical master and to assign the person they imagined a name, 77 percent thought of a man; in 2018, when asked the same questions about a typical head, only 56 percent thought of a man. The students in both cohorts were also shown photos of different Yale faculty members, one at a time, and asked to indicate whether each was a master or not (the 2015 wave of student participants) or a head or not (the 2018 participants). Half of the photos were of faculty who were in those leadership roles, and half were not. The 2015 students more accurately identified which men were house masters than which women were, but the 2018 students were equally accurate at identifying which men and women were house heads. Language matters: It influences what (and whom) we attend to and think about.
There are some stereotypes that you don’t know about. (Did you know that there’s a stereotype that Mormons love trampolines?) But many of them, accurate and inaccurate ones alike, are widely shared. And that opens up another way in which they influence social life. If I am a member of a marginalized group, knowing that you have a stereotype about people like me is likely to grab my attention, occupy my mind, and possibly influence my behavior. And my awareness that you’re likely to think that people like me tend to hold stereotypes about people like you is also likely to influence my behavior—if nothing else, by making me try to convince you that I’m not one of those people who believes in such stereotypes.
All of this knowledge about stereotypes (their content) and meta-knowledge of stereotypes (who holds them) affects our inner lives in various ways. Let’s consider a few of the most important.
To function effectively, people need to understand the causes of events happening around them (see Chapter 4). But this understanding is not so straightforward for members of marginalized groups because they can’t always tell whether their experiences have the same causes as the experiences of majority-group members do or whether their experiences are instead the result of prejudice. In other words, they suffer from attributional ambiguity: “Did my officemate get the promotion instead of me because I’m a woman?” “Would the state trooper have pulled me over if I were White?” Questions like these can be distressing even when it comes to positive outcomes: “Did I get that fellowship because I’m a Latino?” When someone has to wonder whether an accomplishment is the product of an affirmative action policy, for example, it can be difficult to completely “own” it and reap the full measure of pride that the achievement would ordinarily inspire.
In one study that examined this type of attributional predicament, Black and White students received flattering or unflattering feedback from a White student in an adjacent room (Crocker et al., 1991). Half the participants were led to assume that the White student could see them through a one-way mirror, and half thought they couldn’t be seen because a blind covered the mirror. Whether or not they could be seen had no effect on how White students reacted to the feedback, but it did affect how Black students reacted. When Black students thought the other person could not see them—and therefore didn’t know their race—their self-esteem went down from the unflattering feedback and was boosted by the positive feedback. When they thought the other person could see them, in contrast, their self-esteem was not injured by the negative feedback (presumably because they did not know whether to attribute this feedback to their own failings or to the other’s prejudice), nor was it enhanced by the flattery (presumably because they did not know whether to attribute the positive feedback to their own skill or to the other’s condescension). This study indicates that members of marginalized or stigmatized groups live in a less certain world, not knowing to what causes they can attribute their experiences.
An extensive program of research initiated by Claude Steele and his colleagues highlights a second difficulty for members of stigmatized groups (Steele, 1997; Steele, Spencer, & Aronson, 2002). Their performance can be impaired by stereotype threat, the fear that they will confirm the stereotypes others have about them and their group. In one study, researchers examined the effect on women’s math test scores of bringing to mind the stereotype that women don’t perform as well as men in mathematics (Spencer, Steele, & Quinn, 1999). In one condition, participants were told there was no gender difference on a particular test they were about to take. Other participants were told that there was a gender difference in favor of men.
FIGURE 11.7 STEREOTYPE THREAT AND PERFORMANCE This study shows the performance of men and women on a math test when they thought the test tapped gender differences and when they did not. Source: Adapted from Spencer et al., 1999.
DATA EXPLORATION
As Figure 11.7 shows, men and women performed equally well when they thought there was no gender difference on the test, but women performed worse than men when they thought there was a gender difference.
It’s not necessary to blatantly invoke stereotype threat for it to have an effect—although, not surprisingly, the effect tends to be greater when the threat is more obvious (Liu et al., 2021; Shewach, Sackett, & Quint, 2019). Michael Inzlicht and Talia Ben-Zeev (2000) had undergraduate women take a math test in the company of either two other women or two men, but the researchers did not say a word about any gender differences on the test. Nonetheless, those who took the test with other women got an average of 70 percent of the problems correct. Those who took the test alongside men got 55 percent correct on average.
In another study, Steele and Joshua Aronson examined Black students’ sensitivity to stereotype threat (Steele & Aronson, 1995). Playing on a stereotype that questions Black people’s intellectual ability, they gave Black and White college students a difficult verbal test taken from the Graduate Record Exam. Half the students were led to believe that the test could measure their intellectual ability, and half were told that the investigators were in the early stages of developing the test and that nothing could be learned about intellectual ability from the participants’ scores. This information had no effect on the performance of White students. In contrast, the Black students did as well as the White students when they thought it was the test that was being tested, but they performed much worse than the White students when they thought their intellectual ability was being tested. Again, a blatant manipulation was not required to produce a significant effect on the performance of the Black students: Even without directly priming any stereotypes about Black Americans’ intellectual performance, the students still felt the effects of that pervasive stereotype, and it affected their scores accordingly. In a follow-up study, it was enough simply to have participants indicate their race at the top of the page to cause Black students’ performance to be worse than in a control condition in which they did not indicate their race (Steele & Aronson, 1995).
It seems that no one is safe from stereotype threat. Another research team showed that the math performance of White men deteriorated when they were reminded of Asian Americans’ proficiency in math (J. M. Aronson et al., 1999). And in a particularly clever experiment, Jeff Stone and his colleagues had college students perform a laboratory golf task described as a measure of “natural athletic ability,” “sports intelligence,” or “sports psychology” (Stone et al., 1999). White and Black students performed equally well in the “sports psychology” condition. But Black students performed significantly worse when it was described as a test of “sports intelligence,” and White students performed worse when it was described as a test of “natural athletic ability.” In still another telling study, Asian American women did worse on a math test than control participants did when their gender was made salient, but they performed better than control participants did when their race was highlighted (Shih, Pittinsky, & Ambady, 1999).
Stereotype threat appears to undermine performance in a number of ways. It leads to increased arousal, which can directly interfere with performance on complex tasks (see Chapter 12; Ben-Zeev, Fein, & Inzlicht, 2005) and serve as a source of distraction that interferes with concentration on the task at hand (Cheryan & Bodenhausen, 2000). Furthermore, knowing that one’s group is “suspect” in the eyes of others can both directly undermine performance and lead individuals to “play it safe” by being more obsessed with avoiding failure than striving for success (Cadinu et al., 2005; Seibt & Förster, 2004). In accordance with the idea that stereotypes affect health, stereotype threats to women’s math performance have been shown to increase physiological markers of stress (John-Henderson, Rheinschmidt, & Mendoza-Denton, 2015).
Although all people are vulnerable to some type of stereotype threat based on their group memberships, Steele (1997) maintains that the vulnerability of Black Americans has particular potential for damage. Stereotype threat can result in poorer overall academic performance, which undermines confidence, rendering the individual still more susceptible to stereotype threat. This vicious cycle can result in disidentification from academic pursuits, as students who feel the threat most acutely often opt out of academics altogether and identify other ways to invest their talent and energy and build their self-esteem. The same process appears to play a role in the underrepresentation of women in STEM fields (science, technology, engineering, and mathematics) (Deemer, Lin, & Soto, 2015; Walton et al., 2015). Fortunately, social psychologists have developed a number of low-cost, highly effective interventions that can help people deal with the debilitating effects of stereotype threat in schools (see Application Module 3).
Australia is known for its powerhouse Olympic swimming teams, but no Aussie swimmer has had a bigger hold on the country’s imagination than Ian Thorpe. Nicknamed “the Thorpedo” for the speed and grace with which he cut through the water, Thorpe won five Olympic gold medals in his career. Throughout his career, Thorpe was dogged by rumors that he was gay, which he steadfastly denied. “You know, I’m a little bit different to what most people would consider being an Australian male,” he said. “That doesn’t make me gay. I mean I’m straight, so people want to claim me as part of a minority group and put labels on you and that’s not what I’m about, and I don’t understand why people are like that” (Magnay, 2002). After retiring from swimming, however, Thorpe announced during a television interview that he was in fact gay.
Sadly, Thorpe’s experience is not unusual. Members of stigmatized groups throughout history have often felt compelled to hide their true identity. LGBTQ individuals have often chosen to remain “in the closet,” light-skinned Black people have sometimes tried to “pass” as White, and many older adults get plastic surgery, tummy tucks, and toupees in an effort to hide their true age. The ubiquity of such underground experiences makes one wonder what sort of toll they exact.
A big one, it turns out. Physically, the concealment of sexual orientation is associated with cardiovascular stress, and gay men who conceal their sexual orientation show more rapid progression of HIV symptoms (Cole et al., 1996; Pérez-Benítez et al., 2007). Psychologically, being open about one’s sexual orientation is associated with a variety of indicators of better mental health, including reduced depression, less anger, and higher self-esteem (Legate, Ryan, & Weinstein, 2012; Miranda & Storms, 1989; M. W. Ross, 1990; Szymanski, Chung, & Balsam, 2001).
Concealment can also take a cognitive toll. In one study, researchers Clayton Critcher and Melissa Ferguson (2013) instructed half of their participants, all straight, to conceal their sexual orientation during a mock interview, whereas the control participants were free to say whatever they wanted. The investigators predicted that the act of concealment would be mentally taxing, making participants in the first group less able to perform well on subsequent tasks. Indeed, across several experiments, they found that those who were asked to conceal their sexual orientation did less well on tests of spatial ability, self-control, and physical stamina. The monitoring necessary to conceal a part of oneself is demanding, and meeting those demands can have unfortunate consequences down the road.
Other research has shown that concealing any of a variety of significant identities (e.g., being a religious person attending a secular college) leads to the development of something of a “divided self,” whereby aspects of oneself that are public are mentally organized separately from aspects that are private. This public/private schematization is associated with depressive symptoms (Sedlovskaya et al., 2013).
LOOKING BACKMembers of underrepresented and stigmatized groups face discrimination from other individuals, often receiving harsher treatment from members of law enforcement, classroom instructors, and company recruiters and interviewers. They also face institutional discrimination: They often have a harder time landing housing in desirable neighborhoods, receive harsher penalties for similar crimes, are disfavored by the tax code, and have to wait longer to vote. Discrimination against members of marginalized groups can involve acts of commission (for example, harsh treatment by law enforcement) or omission (for example, being overlooked in social interactions or underrepresented in the media). Victims of stereotyping can suffer attributional ambiguity—that is, not knowing whether others’ feedback on their performance is genuine or based on their group membership. They can suffer from stereotype threat, performing worse than they would otherwise because they are afraid of confirming a stereotype that exists about their group. Members of some marginalized groups feel compelled to try to cover up their status, an effort that can exact a physical and psychological toll.
