• Prejudice is ubiquitous; it affects all of us—majority group members as well as minority. People are prejudiced against many aspects of identity: nationality, ethnicity, gender, sexual preference, religion, appearance, physical state, and even professions and hobbies.

A. Prejudice and Self-Esteem

• Prejudice is dangerous, fostering negative consequences from lowered self-esteem to genocide.

• Clark and Clark (1947) showed that African-American children as young as 3 were already convinced that it was not desirable to be black, choosing to play with white rather than black dolls. This evidence led to the 1954 Brown v. Board of Education decision to desegregate schools.

• Goldberg (1968) showed that women had learned to consider themselves intellectually inferior to men, rating the same article higher when it was written by “John McKay” than by “Joan McKay.”

B. A Progress Report

• Over the past decades, blatant discrimination has been reduced; the previous two findings no longer replicate. However, prejudice still exists in subtle—and sometimes blatant—forms.


• Prejudice is an attitude and thus has affective, cognitive, and behavioral components.

Prejudice is a hostile or negative attitude toward a distinguishable group of people based solely on their membership in that group. Prejudiced people direct their prejudice towards members of the group as a whole, ignoring distinguishing characteristics.

A. Stereotypes: The Cognitive Component

• Journalist Walter Lippman introduced the term stereotype in 1922. A stereotype is a generalization about a group of people in which identical characteristics are assigned to virtually all members of the group, regardless of actual variation among the members. Stereotypes are not necessarily emotionally laden and do not necessarily lead to discrimination. Frequently stereotyping is merely a way to simplify a complex world—Allport’s (1954) “law of least effort.”

1. Sports, Race, and Attributions

• The potential abuse engendered by stereotyping can be subtle as well as blatant, and involve positive as well as negative characteristics (e.g., the stereotype that African-Americans are good basketball players). The abuse involves ignoring the overlap of distributions and ignoring individual differences in characteristics. For example, Stone, Perry, & Darley (1997) found that those students who believed a student was African-American rated him as having better athletic ability than those who thought he was white, who rated him as having greater “basketball sense.”

2. Stereotypes, Attribution, and Gender

• Gender stereotypes are still pervasive in our society. Women are seen as more nurturant and less assertive than men; this may be due to their involvement in the homemaker role. Evolutionary psychologists argue that the difference is due to a basis in the behaviors required for reproductive success. Whatever the cause of the difference, this stereotype does have some basis in truth. Work by Eagly, Wood, and Swim shows that there are indeed behavioral differences between men and women such that women are more concerned with the welfare of others and men are more independent and dominant.

• Nonetheless, gender stereotyping often does depart from reality and can cut deeply. For example, people tend to see men’s ability and women’s motivation as responsible for their success, and men’s lack of effort and women’s lower ability as responsible for failure. These results, originally found in the 1970s, continue to be replicated in work in the late 1990s.

• Research shows that girls are more likely to blame themselves for their failures, and boys are likely to blame bad luck. Jacobs and Eccles (1992) showed that daughters of women who held gender stereotypic beliefs were most likely to hold such self-defeating beliefs themselves.

B. Discrimination: The Behavioral Component

Discrimination is an unjustified negative or harmful action towards a member of a group, simply because of his or her membership in that group. For example, Bond, DiCandia, & McKinnon (1988) compared how white vs. black patients in a psychiatric hospital (run by an all-white staff) were treated. They found that, in the first 30 days of a stay, there appeared to be an assumption that blacks would be more violent than whites, as their offenses were more likely to be treated with physical restraints or drugs (Figure 13–1). However, eventually the staff did notice that there was no racial difference in violent incidents and began to treat whites and blacks equally.

1. Discrimination against Homosexuals

In a study by Hebl et al. (2002) confederates applied for jobs in the community. In some job interviews the confederates portrayed themselves as homosexuals and in other interviews they did not. Hebl found that in the cases where the confederates were portrayed as homosexuals the potential employers were less verbally positive and spent less time interviewing them. However, the employers did not formally discriminate against them (e.g., not calling them back as often for follow-up interviews as the other candidates).


• Whether or not there is a biological root to prejudice is unknown; in any case, it is clear that prejudice occurs between biologically similar people who hold different beliefs.

• Prejudices are easy to learn, although childhood prejudices are not necessarily maintained. For example, Rohan and Zanna (1996) found the greatest similarity of beliefs for parents and their children with egalitarian values. Children whose parents hold prejudices may be exposed to competing views and not hold their parents’ prejudices.

• A schoolteacher (Jane Elliot) in Riceville, Iowa, divided her class by eye color, telling the blue-eyed students that they were better than the brown-eyed students and giving them special privileges; in less than half an hour, the formerly cohesive class was split along eye-color lines, with the blue-eyed students taunting and punishing the others, and the brown-eyed students feeling so low that their academic performance was depressed. The next day, the eye-color roles were reversed, and the day after that, the class was debriefed. Even 20 years later, the students claimed the exercise had a life-long impact (see Eye of the Storm and A Class Divided in the film list).

A. The Way We Think: Social Cognition

• One explanation for prejudice is that it is the inevitable byproduct of categorization, schemas, heuristics, and faulty memory processes in processing information.

1. Social Categorization: Us versus Them

• The first step in prejudice is the creation of group categorizations. Once we have mental categories, we group stimuli into them by similarities, downplaying differences between members of a group and exaggerating differences between members of different groups.

2. In-group bias

In-group bias is the especially positive feelings and special treatment we reserve for people we have defined as being part of our in-group (the group with which a person identifies and of which he or she feels a member), and the negative feelings and unfair treatment we reserve for others simply because we have defined them as being in the out-group (groups which an individual does not identify with).

• Tajfel postulates that the underlying motive behind in-group bias is self-esteem maintenance and enhancement. To study this, he invented the minimal group paradigm, in which arbitrary groups were formed by putting strangers together on the basis of trivial criteria. Even in these minimal groups, people still displayed in-group bias by rating in-group members more highly, liking them better, and rewarding them more. People even preferred to take less money as a reward for their own group, if it meant beating the out-group, rather than taking more money but being beat by the out-group.

3. Out-Group Homogeneity

• Another consequence of social categorization is the out-group homogeneity bias, the perception that those in the out-group are more similar (homogenous) to each other than they really are, as well as more similar than the members of the in-group are (i.e., the belief that “they’re all alike”). Quattrone and Jones (1980) showed that Rutgers and Princeton students watching videos of other students (purportedly from Rutgers or Princeton) making decisions would judge the student’s selection as typical of others at his school when the person went to the rival school but not if they went to the student’s own school (Figure 13–2).

4. The Failure of Logic

• There are two reasons why it is almost impossible to get a person holding a deep-seated prejudice to change his or her mind. First, it is primarily the emotional aspect of attitudes that makes a prejudiced person hard to argue with; logic is not effective in countering emotions—people will ignore or distort any challenge to their belief. Second, people with strong prejudices have a firmly established schema for the target group(s); this will lead them to pay attention to, and recall more often, information that is consistent with their beliefs than that which is inconsistent. Thus stereotypes become relatively impervious to change.

5. The Persistence of Stereotypes

• Table 13–1 displays the beliefs of students about Americans, Japanese, Jews, and African-Americans from 1933 to 1969. Over 30 years, the stereotypes remained fairly stable, becoming somewhat less negative. By 1969 many students felt uncomfortable with the task and only agreed to do it if it was made clear they were displaying their knowledge of societal stereotypes and not their own beliefs.

6. The Activation of Stereotypes

• Greenberg and Pyszczynski (1985) conducted a study to find out whether knowing a stereotype will affect the processing of information about a target person even for unprejudiced people. Observers watched a staged debate between a white and a black student; which student performed better in the debate was manipulated. Additionally, a planted confederate in the audience either made a racist remark, a nonracist remark, or no remark about the black student. When this student was the poorer debater, the racist remark activated the negative stereotype and led to lower ratings of him than in the other conditions (Figure 13.4). Similarly, Henderson-King and Nisbett (1996) found that it took only one negative action by one African-American to activate the negative stereotype against blacks and discourage participants from wanting to interact with a different African-American. These findings suggest that stereotypes exist in most of us and are easily activated to have negative effects on the perception and treatment of out-group members.

7. Automatic and Controlled Processing of Stereotypes

• Patricia Devine (1989) developed a theory about how stereotypical and prejudiced beliefs affect information processing. Her theory is based on the distinction between automatic and controlled information processing. According to her theory, when we process information about another, a two-step process takes place: first the stereotypes that we know about are automatically triggered, then in the controlled process we decide whether or not to accept the stereotype; unprejudiced people will use the controlled process to override it. However, if a person is distracted, overwhelmed, or not attending, the controlled processing will not be initiated, and the stereotype will prevail (see Figure 13–5). In a test of this theory, Devine showed that high and low prejudiced Ss showed equal knowledge of the stereotype of African-Americans; in a second part of the study, she displayed either stereotypical or nonstereotypical words to Ss subliminally; then she asked them to rate an ambiguous story about “Donald.” Those Ss who had been subliminally exposed to the stereotypical words rated Donald more harshly, regardless of level of prejudice. Finally, in a third study, Devine showed that, when processing consciously, high prejudice students listed significantly more negative words than low prejudice students in describing black Americans.

8. The Justification-Suppression Model of Prejudice Crandall and Eshleman (2003) offer a model of the expression of prejudice. They content that people struggle between the urge to express prejudice and their need to maintain a positive self-concept. If we find valid justification for holding a negative attitude toward a group, we can act against them and still feel as though we are not bigots.

9. The Illusory Correlation

• An illusory correlation is the tendency to see relationships, or correlations, between events that are actually unrelated. Illusory correlations are most likely to occur when the events or people are distinctive or conspicuous; minority group members are so by definition. Once formed, an illusory correlation increases attention to confirming information and decreases attention to disconfirming information. The media create illusory correlations by their stereotypical presentations of women and minorities.

10. Can We Change Stereotypical Beliefs?

• Kunda and Oleson (1997) found that when people are presented with examples that strongly challenge their existing stereotypes, they tend to dismiss the disconfirming example as “the exception that proves the rule,” and some actually strengthen their stereotypic belief.

• Nonetheless, there are some situations when stereotypes can change.

B. How We Assign Meaning: Attributional Biases

1. Dispositional versus Situational Explanations

• Stereotypes are negative dispositional attributions. Thomas Pettigrew has called our making dispositional attributions about a whole group of people the ultimate attribution error.Bodenhausen (1988) found that students were more likely to find a defendant guilty of a crime (ignoring extenuating circumstances) when his name was Carlos Ramirez than when it was Robert Johnson. In an earlier study, Bodenhausen and Wyer (1985) had found that when a crime was consistent with a group stereotype, Ss were less lenient in parole decisions, ignoring other relevant information, than when the crime was inconsistent with a group stereotype. Thus when people act in a way that confirms our stereotype, we make dispositional attributions and ignore possible situational causes.

2. Stereotype Threat Steele and J. Aronson (1995) have shown that at least one major contributing factor is situational. They define stereotype threat as the apprehension experienced by members of a minority group that they might behave in a manner than confirms an existing cultural stereotype. This worry in turn interferes with their ability to perform well in these situations. For example, Steele and Aronson found that when white and black students were told that a difficult test they were taking was just in the development phase and thus not valid, there were no differences in performance; but when the students were told that the same test was a valid measure of intellectual ability, the blacks performed more poorly than the whites.

• Stereotype threat applies to gender as well as race. Spencer and Steele (1996) found a similar phenomenon among women taking math tests. Even white males can display the phenomenon—when compared to Asian males on a math exam (J. Aronson et al., 1999, 2000).

• The more conscious individuals are of the pertinent stereotype, the greater the effect on their performance (Brown & Pinel, 2002).

• Research indicates that providing a counter-stereotypic mind-set (e.g., I’m a student at a top university) can eliminate the effects of stereotype threat.

3. Expectations and Distortions

• When an out-group member behaves in a way that disconfirms our stereotypes, we are likely to make a situational attribution for his or her performance, leaving the stereotype intact. For example, Ickes et al. (1982) told college men that the person they would interact with was either extremely friendly or extremely unfriendly. In both conditions, the Ss went out of their way to be nice to their partners and their partner returned their friendliness. However, those who expected their partner to be unfriendly explained his friendly behavior away as being a phony response due to their own pleasant behavior.

4. Blaming the Victim

Blaming the victim is the tendency to blame individuals (make dispositional attributions) for their victimization; ironically, it is motivated by a desire to see the world as a fair and just place where people get what they deserve. Believing that people get what they deserve leads one to blame victims for their outcomes. Negative attitudes toward the poor, including blaming them for their own plight, are more prevalent among individuals who display strong belief in a just world (Furnham & Gunter, 1984).

5. Self-Fulfilling Prophecies

• The self-fulfilling prophecy is a process in which we find confirmation and proof for our stereotypes by unknowingly creating stereotypical behavior in out-group members through our treatment of them.

• Word, Zanna, and Cooper (1974) conducted a set of experiments that demonstrates the phenomenon. In the first study, they asked white undergraduates to interview job applicants who were either white or black. The students tended to display discomfort when interviewing the African-Americans: for example, they sat further away, stammered, and terminated the interview earlier. In a second experiment, the researchers varied the behavior of the interviewers so that they acted towards a job candidate either the way that the interviewers had acted towards whites or the way that the interviewers had acted towards blacks in the first study. They found that those applicants who had been interviewed in the way that African-Americans had been interviewed were judged to be more nervous and less effective than the others (Figure 13–6).

C. Prejudice and Economic Competition: Realistic Conflict Theory

Realistic conflict theory is the theory that limited resources lead to conflict between groups and result in increased prejudice and discrimination.

1. Economic and Political Competition

• Several historical studies document that discrimination against out-groups covaries with the scarcity of jobs or other resources.

• Although correlational data is supportive of the theory, it still does not allow a causal inference. To allow this, an experiment is essential, such as that conducted by Sherif et al. (1961). In the classic “Robber’s Cave” experiment, two groups of 12-year-old boys at a summer camp were randomly assigned to one of two groups, the Eagles or the Rattlers. In the first phase of the study, the groups were isolated and placed in situations designed to increase group cohesiveness. In the second phase of the study, the researchers set up a series of competitive activities in which the two groups were pitted against each other. Hostility between the two groups rapidly escalated. In the next phase of the study, researchers tried to eliminate hostility by eliminating competitive games and increasing contact. This failed to reduce the hostilities (the final resolution follows later in the chapter).

2. The Role of the Scapegoat

Scapegoating, the tendency for individuals, when frustrated or unhappy, to displace aggression onto groups that are disliked, visible, and relatively powerless, may occur when people are frustrated (for example, by scarcity of resources) but there is no clear target to blame the frustration on. It may occur even in the absence of direct competition.

• Such scapegoating may be seen in recent years with homosexuality.

D. The Way We Conform: Normative Rules

• Through both explicit and implicit socialization, we are trained in the norms of our culture. Stereotypes and prejudiced attitudes are part of this normative package.

1. When Prejudice is Institutionalized Institutionalized racism refers to the idea that racist attitudes are held by the vast majority of us because we live in a society where stereotypes and discrimination are the norm; institutionalized sexism is the idea that sexist attitudes are held by the vast majority of us for the same reason. In societies in which racism and sexism are institutionalized, normative conformity leads to the tendency to go along with the group in order to fulfill their expectations and gain acceptance. Pettigrew (1958) argues that the greatest determinant of prejudice is this slavish conformity to social norms. For example, he showed that ministers in Little Rock, Arkansas, in the 1950s were personally in favor of desegregation but kept these fears to themselves. Other studies show that people’s prejudice and discrimination changes when they move to an area with different norms, or even, in a study of miners in West Virginia, when they are underground and when above. Over the past 50 years, American norms for attitudes such as that towards desegregation have changed drastically.

2. “Modern” Prejudice

• Although American norms have changed and the blatant expression of prejudice has diminished, prejudice is still with us. Modern racism is prejudice revealed in subtle, indirect ways because people have learned to hide prejudiced attitudes in order to avoid being labeled as racist. For example, many parents protest against their children being bussed only when the busing is interracial. Because of the nature of modern prejudice, it can best be studied using subtle or unobtrusive measures. For example, the bogus pipeline technique uses an impressive-looking machine labeled as a lie detector; the machine is a fake. People who are hooked up to the machine and believe that their true attitudes can be detected showed higher levels of racism and sexism than those completing the paper scales than white males.

3. Subtle and Blatant Prejudice in Western Europe • Pettigrew and Meertens (1995) examined blatant and modern racism in France, the Netherlands, and Great Britain. They found that those who scored as racist on both scales wanted to send immigrants back; those who scored low on both wanted to improve their rights and were willing to take actions to do so, and those who scored as nonracist on the blatant scale but racist on the subtle scale did not want to take action to send immigrants back but nor were they willing to support any actions to help improve their rights.

E. Subtle Sexism

Subtle forms of prejudice can also be directed toward woman. Many men have feelings of ambivalence toward women and as Glick and Fiske (2001) have shown this ambivalence can take one of two forms: hostile sexism or benevolent sexism. Hostile sexism suggests that women are inferior to men while benevolent sexism tends to idealize women romantically.


• The hope that prejudice can be reduced by education has proven naive. Change requires more.

A. The Contact Hypothesis

• The contact hypothesis is the idea that merely bringing members of different groups into contact with each other will erode prejudice. This idea lay at the basis of the 1954 Supreme Court decision on school desegregation. For example, Deutsch and Collins (1951) had shown that white and black families randomly assigned to an integrated housing unit showed reductions in racism compared to those assigned to segregated units. However, things did not work so smoothly in school desegregation: there was tension, and, in more than half of the studies, prejudice actually increased. In a quarter of the studies, the self-esteem of African-American children was found to have decreased after desegregation. Mere contact does not work.

B. When Contact Reduces Prejudice: Six Conditions

• Allport (1954) suggested that six conditions are necessary for inter-group contact to reduce prejudice: (1) mutual interdependence, or the existence of situations where two or more groups need each other and must depend on each other in order to accomplish a goal; (2) a common goal that is important to both of them; (3) equal status of group members; (4) having informal interpersonal contact; (5) multiple contacts with several members of the out-group so that individuals can learn that their beliefs are wrong; and (6) social norms in place that promote equality. When these conditions are met, suspicious or even hostile groups will reduce their stereotyping, prejudice, and discrimination. Sherif’s Robber’s Cave study, described above, ultimately resolved the intergroup hostility by fostering each of these six conditions (Figure 13.7).

C. Why Early Desegregation Failed • In most classrooms, the environment is very competitive; when minority students who have had deficient preparation are bussed in, they are guaranteed to lose the competition. The situation is ripe for the creation of self-fulfilling prophecies by both majority and minority group members. Thus Stephen (1978) found a general decrease in self-esteem of minority students following desegregation. To change the atmosphere of the classroom so that it meets the six conditions outlined above, Aronson and his colleagues developed the jigsaw classroom. This is a classroom setting designed to reduce prejudice and raise the self-esteem of children by placing them in small desegregated groups and making each child dependent on the other children in his or her group to learn the course material and do well in the class. Formal studies demonstrate that children in jigsaw classrooms perform better and show greater increases in self-esteem than those in traditional classrooms; further, they show more evidence of true integration and better abilities to empathize with and see the world through the eyes of others.

D. Why Does Jigsaw Work?

• Gaertner et al. (1990) suggest that the process is effective because it breaks down in-group and out-group categorization and fosters the notion of the class as a single group.

• Another reason is that it places people in a “favor-doing” situation, which leads people to like those they do favors for.

• A third reason why the jigsaw process is effective is that it encourages the development of empathy. Bridgeman thus showed that 10-year-old students who had spent two months in a jigsaw classroom were more likely to successfully take the perspective of a story character and correctly answer questions from this character’s point of view than were students who had not had the jigsaw classroom experience.

1. The Gradual Spread of Cooperative Learning

• The cooperative learning movement has been widely accepted by researchers as one of the most effective ways of improving race relations, building empathy, and improving instruction in schools. However, the educational system, like all bureaucracies, resists change, and the slowness of change can have tragic consequences.