How widespread are automatic prejudiced reactions to African Americans? Experiments have shown such reactions in varied contexts. For example, in clever experiments by Anthony Greenwald and his colleagues (1998, 2000), 9 in 10 White people took longer to identify pleasant words (such as peace and paradise) as “good” when associated with Black rather than White faces. The participants consciously expressed little or no prejudice; their bias was unconscious and unintended. Moreover, report Kurt Hugenberg and Galen Bodenhausen (2003), the more strongly people exhibit such implicit prejudice, the readier they are to perceive anger in Black faces.
critics note that unconscious associations may only indicate cultural assumptions, perhaps without prejudice (which involves negative feelings and action tendencies). But some studies find that implicit bias can leak into behavior:
• In a Swedish study, a measure of implicit biases against Arab-Muslims predicted the likelihood of 193 corporate employers not interviewing applicants with Muslim names (Rooth, 2007).
• In a medical study of 287 physicians, those exhibiting the most implicit racial bias were the least likely to recommend clot-busting drugs for a Black patient described as complaining of chest pain (Green & others, 2007).
• In a study of 44 Australian drug and alcohol nurses, those displaying the most implicit bias against drug users were also the most likely, when facing job stress, to want a different job (von Hippel & others, 2008). In some situations, automatic, implicit prejudice can have life or death consequences. In separate experiments, Joshua Correll and his co-workers (2002, 2006, 2007)
and Anthony Greenwald and his co-workers (2003) invited people to press buttons quickly to “shoot” or “not shoot” men who suddenly appeared onscreen holding either a gun or a harmless object such as a flashlight or a bottle. The participants (both Blacks and Whites, in one of the studies) more often mistakenly shot harmless targets who were Black. (Follow-up computerized simulations revealed that it’s Black male suspects—not females, whether Black or White—that are more likely to be associated with threat and to be shot [Plant & others, 2011].) In the aftermath of London police shooting dead a man who looked Muslim, researchers also found Australians more ready to shoot someone wearing Muslim headgear (Unkelbach & others, 2008).
If we implicitly associate a particular ethnic group with danger, then faces from that group will tend to capture our attention and trigger arousal (Donders & others, 2008; Dotsch & Wigboldus, 2008; Trawalter & others, 2008). In a related series of studies, Keith Payne (2001, 2006) and Charles Judd and colleagues (2004) found that when primed with a Black rather than a White face, people think guns: They more quickly recognize a gun and they more often mistake a tool, such as a wrench, for a gun. Even when race does not bias perception, it may bias reaction—as people require more or less evidence before firing (Klauer & Voss, 2008). Jennifer Eberhardt and her colleagues (2004) demonstrated that the reverse effect can occur as well. Exposing people to weapons makes them pay more attention to faces of African Americans and even makes police officers more likely to judge stereotypical-looking African Americans as criminals. These studies help explain why in 1999,
Amadou Diallo (a Black immigrant in New York City) was shot 41 times by police officers for removing his wallet from his pocket. It also appears that different brain regions are involved in automatic and consciously controlled stereotyping (Correll & others, 2006; Cunningham & others, 2004; Eberhardt, 2005). Pictures of outgroups that elicit the most disgust (such as drug addicts and the homeless) elicit brain activity in areas associated with disgust and avoidance (Harris & Fiske, 2006). This suggests that automatic prejudices involve primitive regions of the brain associated with fear, such as the amygdala, whereas controlled processing is more closely associated with the frontal cortex, which enables conscious thinking.
We also use different bits of our frontal lobes when thinking about ourselves or groups we identify with, versus when thinking about people that we perceive as dissimilar to us (Jenkins & others, 2008; Mitchell & others, 2006). Even the social scientists who study prejudice seem vulnerable to automatic prejudice, note Anthony Greenwald and Eric Schuh (1994). They analyzed biases in authors’ citations of social science articles by people with selected non-Jewish names (Erickson, McBride, etc.) and Jewish names (Goldstein, Siegel, etc.). Their analysis of nearly 30,000 citations, including 17,000 citations of prejudice research, found something remarkable: Compared with Jewish authors, non-Jewish authors had 40 percent higher odds of citing non-Jewish names. (Greenwald and Schuh could not determine whether Jewish authors were overciting their Jewish colleagues or whether non-Jewish authors were overciting their non-Jewish colleagues or both.)
How pervasive is prejudice against women? Here we consider gender stereotypes —people’s beliefs about how women and men do behave. Norms are prescriptive; stereotypes are descriptive.
From research on stereotypes, two conclusions are indisputable: Strong gender stereotypes exist, and, as often happens, members of the stereotyped group accept the stereotypes. Men and women agree that you can judge the book by its sexual cover. In one survey, Mary Jackman and Mary Senter (1981) found that gender stereotypes were much stronger than racial stereotypes. For example, only 22 percent of men thought the two sexes equally “emotional.” Of the remaining 78 percent, those who believed females were more emotional outnumbered those who thought males were more emotional by 15 to 1. And what did the women believe? To within 1 percentage point, their responses were identical. Remember that stereotypes are generalizations about a group of people and may be true, false, or overgeneralized from a kernel of truth.
Do we then conclude that gender stereotypes are accurate? Sometimes stereotypes exaggerate differences. But not always, observed Janet Swim (1994). She found that Pennsylvania State University students’ stereotypes of men’s and women’s restlessness, nonverbal sensitivity, aggressiveness, and so forth were reasonable approximations of actual gender differences. Gender stereotypes have persisted across time and culture. Averaging data from 27 countries, John Williams and his colleagues (1999, 2000) found that people everywhere perceive women as more agreeable, and men as more outgoing. The persistence and omnipresence of gender stereotypes have led some evolutionary psychologists to believe they reflect innate, stable reality (Lueptow & others, 1995). Stereotypes (beliefs) are not prejudices (attitudes). Stereotypes may support prejudice. Yet one might believe, without prejudice, that men and women are “different yet equal.” Let us, therefore, see how researchers probe for gender prejudice.
SEXISM: BENEVOLENT AND HOSTILE
Judging from what people tell survey researchers, attitudes toward women have changed as rapidly as racial attitudes have. the percentage of Americans willing to vote for a female presidential candidate has roughly paralleled the increased percentage willing to vote for a Black candidate. In 1967, 56 percent of first-year American college students agreed that “the activities of married women are best confined to the home and family”; by 2002, only 22 percent agreed (Astin & others, 1987; Sax & others, 2002). Thereafter, the home–family question no longer seemed worth asking. Alice Eagly and her associates (1991) and Geoffrey Haddock and Mark Zanna (1994) also report that people don’t respond to women with gut-level negative emotions as they do to certain other groups. Most people like women more than men.
They perceive women as more understanding, kind, and helpful. A favorable stereotype, which Eagly (1994) dubs the women-are-wonderful effect, results in a favorable attitude. But gender attitudes often are ambivalent, report Peter Glick, Susan Fiske, and their colleagues (1996, 2007) from their surveys of 15,000 people in 19 nations. Gender attitudes frequently mix a benevolent sexism (“Women have a superior moral sensibility”) with hostile sexism (“Once a man commits, she puts him on a tight leash”).
Being male isn’t all roses. Compared to women, men are three times more likely to commit suicide and be murdered. They are nearly all the battlefield and death row casualties. They die five years sooner. And males represent the majority with mental retardation or autism, as well as students in special education programs (Baumeister, 2007; S. Pinker, 2008). One heavily publicized finding of discrimination against women came from a 1968 study in which Philip Goldberg gave women students at Connecticut College several short articles and asked them to judge the value of each. Sometimes a given article was attributed to a male author (for example, John T. McKay) and sometimes to a female author (for example, Joan T. McKay). In general, the articles received lower ratings when attributed to a female. That’s right: Women discriminated against women. Eager to demonstrate the subtle reality of gender discrimination, I obtained Goldberg’s materials in 1980 and repeated the experiment with my own students.
They (women and men) showed no such tendency to deprecate women’s work. So Janet Swim, Eugene Borgida, Geoffrey Maruyama, and I (1989) searched the literature and corresponded with investigators to learn all we could about studies of gender bias in the evaluation of men’s and women’s work. To our surprise, the biases that occasionally surfaced were as often against men as women. But the most common result across 104 studies involving almost 20,000 people was no difference. On most comparisons, judgments of someone’s work were unaffected by whether the work was attributed to a female or a male. Summarizing other studies of people’s evaluations of women and men as leaders, professors, and so forth, Alice Eagly (1994) concluded, “Experiments have not demonstrated any overall tendency to devalue women’s work.” Is gender bias fast becoming extinct in Western countries? Has the women’s movement nearly completed its work? As with racial prejudice, blatant gender prejudice is dying, but subtle bias lives.
Violate gender stereotypes, and people may react. People take notice of a cigar smoking woman and a tearful man and denigrate a White rapper (Phelan & Rudman, 2010). A woman whom people see as power hungry suffers more voter backlash than does a similarly power-hungry man (Okimoto & Brescoll, 2010). In the world beyond democratic Western countries, gender discrimination is not subtle. Two-thirds of the world’s unschooled children are girls (United Nations, 1991).
In some countries, discrimination extends to violence, even prosecuting rape victims for adultery (UN, 2006). But the biggest violence against women may occur prenatally. Around the world, people tend to prefer having baby boys. In the United States, in 1941, 38 percent of expectant parents said they preferred a boy if they could have only one child; 24 percent preferred a girl; 23 percent said they had no preference. In 2011, the answers were virtually unchanged, with 40 percent still preferring a boy ( Newport, 2011). With the widespread use of ultrasound to determine the sex of a fetus and the growing availability of abortion, these preferences are, in some countries, affecting the number of boys and girls. In China, where 95 percent of orphanage children are girls (Webley, 2009), the 118 boys born for every 100 girls has led to an excess of 32 million under-20 males.
These are tomorrow’s “bare branches,” as the Chinese think of them—bachelors who will have trouble finding mates (Hvistendahl, 2009, 2010, 2011; Zhu & others, 2009). This “gender genocide” is not found only in China. Taiwan, Singapore, India, and South Korea likewise have millions of “missing women” (Abrevaya, 2009). In response, China has made sex-selective abortions a criminal offense. To conclude overt prejudice against people of color and against women is far less common today than it was in the mid-twentieth century. Nevertheless, techniques that are sensitive to subtle prejudice still detect widespread bias. And in parts of the world, gender prejudice makes for misery. Therefore, we need to look carefully and closely at the social, emotional, and cognitive sources of prejudice.