SciELO - Scientific Electronic Library Online

vol.22 número109Funcionamiento diferencial del ítem en pruebas de matemática para educación mediaPrejudice, discrimination and intergroup conflict índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Actualidades en psicología

versão On-line ISSN 0258-6444

Actual. psicol. v.22 n.109 San José  2008




Intergroup prejudice: its causes and cures1



Thomas F. Pettigrewr

University of California, Santa Cruz

Dirección para correspondencia



Defining prejudice

Intergroup conflict is apparent throughout the globe today, fueled by prejudice and discrimination. But prejudice, like most human phenomena, is more complex than it first appears. For years, social psychology followed Gordon Allport’s straightforward definition: intergroup prejudice consists of negative opinions against an outgroup without sufficient evidence (Allport, 1954). In other words, prejudice is being down on something you are not up on. Note that this view holds prejudice to involve both negative emotions and irrational beliefs. But my teacher’s definition turns out to be too simple.

Consider prejudice against women. Most opinions of men about women are in fact favorable. Prejudice becomes evident only when women step out of the roles that society prescribes for them. For example, there is often resistance to even a competent woman becoming an airline pilot. Such resistance arises because there is a perceived “lack of fit” between the generally positive stereotype of women and that of airline pilots. Women are supposed to lack the technical skills required of pilots.

Thus, prejudice becomes evident when there is a perception of “role incongruity” (Eagly & Dickman, 2005). When large numbers of women challenge these perceptions and attempt to assume previously all-male roles.

&– as in the feminist movement, anti-female beliefs and emotions rise to the surface. My wife, who became a medical doctor at a time when women constituted less than 10% of all American physicians, remembers this phenomenon all too well.

Note that this view of prejudice also challenges the irrationality of beliefs part of the old definition. If discrimination routinely keeps all women from being trained as airline pilots, then the stereotyped conception that women are not able to fly airplanes is in one limited sense true. Prejudice is involved in this belief in that it assumes it is innate feminine qualities, rather than the inability to obtain the needed training, that makes women unqualified to fly.


Blatant and subtle prejudice

In research throughout the world, social psychologists have sought to capture this complexity by uncovering more subtle forms of intergroup prejudice. In simplest terms, think of two rather different &– though highly intercorrelated &– forms of prejudice: Blatant and Subtle Prejudice (Pettigrew & Meertens, 1995). Blatant prejudice is the more traditional form &– close, hot, and direct. Our measure of it has two components &– one a threat and rejection factor (e.g., Nicaraguans have jobs Costa Ricans should have.), the other an intimacy factor (e.g., I would object if a Nicaraguan who had a similar economic background as mine joined my close family by marriage.).

By contrast, subtle prejudice is the modern form &– distant, cool, and indirect. It has three components that share an ostensibly non-prejudicial focus. The first consists of a traditional values factor (e.g., Nicaraguans living here teach their children values and skills different from those required to be successful in Costa Rica.). The second component concerns exaggerated views of the outgroup as extremely different culturally from the ingroup. The final subtle component involves the denial of sympathy and admiration for the outgroup. Note this last component tests for the denial of positive emotions rather than the expression of negative emotions &– fear, envy, hatred &– that are associated with blatant prejudice. In short, Allport’s old definition still describes blatant prejudice; while the new lack-of-fit, role incongruity conception of prejudice more closely describes the new subtle form.

Critics of social psychology have heatedly argued that subtle prejudice, as I have described it, is not prejudice at all. But the fact remains that throughout the world subtle prejudice correlates highly with blatant prejudice and predicts discriminatory intentions and behavior (Pettigrew & Meertens, 2001). In other words, it acts just like blatant prejudice. So I maintain that if it waddles like a duck, quakes like a duck, and swims like a duck &– then it is a duck! Subtle prejudice is prejudice.


Prejudice and discrimination

Intergroup discrimination is even more complex than prejudice; but the two are clearly related. Discrimination can be defined as the restriction of a minority’s life chances and choices in comparison to those of the dominant group.

Major sources of intergroup discrimination are structural &– that is, our institutions (schools, businesses, government agencies) often have built-in arrangements that operate to discriminate against certain groups. Such institutional discrimination can even occur without there being any intention to discriminate by those who head the institutions. That is, non-prejudiced people often unwittingly lead highly discriminatory institutions.

Often institutional discrimination can also be quite subtle. For years in the United States, surgery required all training residents to work full-time. This seemingly innocuous requirement effectively kept all married women out of surgery. The residency came right at the critical child-bearing time for women in medicine. They needed a part-time residency. Similarly, in a court case involving the firemen of San Francisco, California, many Asian Americans and women in general were effectively denied employment by a requirement that they must be big and strong enough to move large, heavy hoses. For years, this requirement seemed reasonable. But it turned out that modern fire hoses are far smaller and lighter; so the Federal Court ruled that the Fire Department had to use the newer hoses and start hiring the previously excluded groups.

Discrimination can also occur at the face-to-face level in intergroup situations. Here the operating norms of the situation govern what happens largely apart from the prejudices of the participants. If discrimination against minorities and women is tolerated, even encouraged, by the larger institution, norms soon develop for discrimination to flourish in situations throughout the institution. And even the unprejudiced may become involved. By contrast, when discrimination is not tolerated by the institution, even extreme bigots avoid discriminating for fear that they would lose their jobs were they to act out their beliefs.

It is at the individual level of analysis that prejudice becomes a major factor in intergroup discrimination. Studies conducted by social psychologists throughout the world find that prejudice strongly predicts discriminatory behavior and intentions when the situational norms do not relate to intergroup behavior. “Would you be willing to buy a car from a Turk?” is a question often asked in German surveys; and individual prejudice predicts the results far better than any other variable (Wagner, Christ & Pettigrew, 2006).


The causes of prejudice

Like intergroup discrimination, the causes of intergroup prejudice involve all three levels of analysis &– structural, situational, and individual. Historical conflict and such institutions as slavery create at the societal level norms of discrimination, hatred and resentment that last for years. Most people conform to these long-established societal norms and develop prejudicial attitudes simply as part of their adaptation to their life situation. In other words, many people who are prejudiced are simply conforming to prejudicial norms &– it is in the air they breathe. If the societal norms are changed, conforming people often shift their intergroup attitudes with surprising ease. This fact is the basis of most remedial procedures for reducing prejudice &– a topic we shall discuss at the close of the talk.

Conforming prejudice is enhanced by societal barriers to intergroup contact. The classic examples are Apartheid in South Africa and racial segregation in the southern United States. I grew up in the American South in the 1930s and 1940s when racist beliefs and practices against Black Americans were at their height. I soon learned that opposing these racist norms led to swift punishment. For example, if I objected to racist remarks by teachers, I would be immediately dismissed from school. Punishments for Black Americans violating the norms were, of course, much more severe.

These same negative societal norms also poison intergroup situations. The dominant group is expected to be commanding, the less powerful group subservient. Blatant forms of discrimination are likely to become established. If there is threat &– especially threat to the dominant group itself &– then greater discrimination and prejudice are typically invoked. Such situations are likely to increase the power of the societal norms and make them seem justified.

But not all prejudice derives from conformity to societal norms. Social psychologists have shown that some people seem especially prone to intergroup prejudice. One clear indication of this fact is that these individuals tend to be prejudiced against an array of quite different types of groups &– social class, racial, religious, national, even the infirmed. Prejudice for them obviously serves deep personality needs apart from the norms that operate for intergroup relations.

Two personality syndromes have been uncovered that are especially prone to intergroup prejudice in countries all over the world. The first is authoritarianism (Adorno, Frenkel-Brunswik, Levinson & Sanford, 1950; Altemeyer, 1988, 1998). Three personality traits characterize this syndrome: deference to authorities, aggression toward outgroups, and rigid adherence to cultural conventions. Typical items used to measure authoritarianism are: “Crime should be punished more harshly” and “Two of the most important characteristics should be obedience and respect for one’s superiors.”

Authoritarianism rises in times of societal threat, and recedes in times of calm. Crises invoke authoritarian leadership and encourage equalitarians to accept such leadership &– as has been true since the 9/11 tragedy in the United States. And authoritarians are more likely than others to favor extreme right-wing politics. Three key questions arise: Just what is authoritarianism? What are its origins? And why does it universally predict prejudice against a wide variety of outgroups?

The remarkable global consistency of research results suggests that the authoritarian personality is a general personality syndrome with early origins in childhood that center on universal issues of authority. Recent work on the syndrome’s origins connects authoritarianism with attachment theory. Rejection by an early caregiver, often the mother, leads to an avoidance attachment style that closely resembles the authoritarian personality. Recent survey data with a probability sample of German adults reveal a strong relationship between the syndrome and a desire to avoid interpersonal closeness (Pettigrew, Wagner, Christ, & Stellmacher, 2006).

Developed early in life, authoritarianism later leads to conditions and behaviors that in turn generate intergroup prejudice. For example, authoritarians more often feel politically powerless and that modern life is too complex and bewildering &– both predictors of prejudice. Situational factors are also involved. Authoritarians tend to associate with others who are prejudiced. And they tend to avoid contact with outgroup members - a major means for reducing prejudice. (Pettigrew, Wagner, Christ, & Stellmacher, 2006)

Thus, the authoritarian personality concept is a useful tool for social psychology to understand a range of important social phenomena. It has stood the test of time and an abundance of research. But it operates at the individual level of analysis. Writers often erroneously employ it to explain societal phenomenon &– a compositional fallacy that incorrectly assumes societal processes are mere composites of individual behavior (Pettigrew, 1996).

A second personality correlate of generalized prejudice &– social dominance orientation &– is positively related with authoritarianism (Sidanius & Pratto, 1999). Individuals with this orientation regard the world as organized in a strict hierarchical fashion &– with some groups deservedly superior and dominant. Prejudice for these people erupts sharply when a threatened group attempts to maintain its superior status. A typical item to tap this syndrome is: Groups at the bottom of society should stay there. In many studies, this social dominance orientation predicts prejudice even better than authoritarianism.


How to reduce prejudice

These many considerations raise the critical question: How can intergroup prejudice be reduced? One major remedy is through intergroup contact under optimal conditions. Social psychologists repeatedly find that intergroup contact reduces prejudice in research conducted throughout the world &– from Northern Ireland, India and South Africa to Australia and Costa Rica. Five interrelated contact conditions greatly facilitate the effect (Allport, 1954; Pettigrew, 1998a).

Equal status. The groups share equal status and power in the situation. And, just as important, their members perceive equal group status to be operating.

Interdependence. The groups share common goals and work cooperatively to achieve these goals. Group against group competition in zero-sum games &– in which what one side wins, the other loses - is a certain recipe for increased intergroup hostility and conflict. By contrast, group interdependence builds cross-group bonds; in time it can even create a single, overarching group identity. In this situation, cooperation between the groups wins rewards for both that are unattainable for each group working alone.

Cross-group friendship potential. Contact that fosters cross-group friendships achieves maximally positive effects. Such friendships not only improve attitudes toward the outgroup, but also tend to reduce prejudice against other outgroups not even involved in the contact. Moreover, just having ingroup friends who have outgroup friends &– so-called indirect contact &– also tends to lower prejudice.

A positive experience that counters prior negative group stereotypes. Enjoyable intergroup contact relieves prior anxiety about the interaction and counters negative stereotypes held about the outgroup.

Authority sanction. Finally, positive effects are enhanced when the intergroup contact has the sanction and support of authorities. Illicit contact carries the stigma of a norm violation and is unlikely to be generalized to the groups at large.

These contact conditions foster positive intergroup interaction that reduces prejudice and conflict between groups. Increasing evidence indicates, however, that these conditions are not essential; they simply facilitate improved intergroup relations. Contact can lead to positive outcomes even in situations that lack many of these conditions.

One of the largest research literatures in social psychology supports these contentions. A recent meta-analysis collected 516 separate studies, 714 independent samples with 250,000 subjects from 39 different countries around the globe (Pettigrew & Tropp, 2006). The principal findings of this analysis are:

Intergroup contact typically reduces intergroup prejudice. The average mean effect across these varied studies is a correlation of -.21. There is, however, great variation in effect sizes across the many studies even though only 6% of all these studies found that contact had no effect or led to increased prejudice. Just as intergroup contact theory predicts, this vast heterogeneity of effects is shaped by the varying conditions under which the contact takes place. For example, intergroup contact that occurs in recreational settings &– such as playing on the same baseball team - diminish prejudice significantly more than contact in tourist settings. And intergroup contact programs specifically designed to meet the optimal conditions yield more robust reductions in prejudice.

This empirically established link between intergroup contact and less prejudice is not an artifact. The relationship between contact and decreases in prejudice might occur because prejudiced people simply avoid intergroup contact. But investigations that test this possibility consistently find that the stronger causal path involves contact diminishing prejudice. Or, perhaps, the negative association between contact and prejudice merely reflects a publishing bias for research that supports the contact theory. This possibility is rendered unlikely by the finding that unpublished studies in the metaanalysis actually yield larger effects than published studies.

Intergroup contact effects typically generalize beyond the specific outgroup members in the immediate situation to the entire outgroup. Extensive generalization to the entire outgroup normally occurs. The average effect for reduced prejudice toward the entire outgroup approaches that of reduced prejudice toward the outgroup members in the immediate contact situation.

The more rigorous the research, the more it demonstrates that contact reduces prejudice. Hence, experimental studies obtain significantly larger effects than surveys.

Many different manifestations of prejudice are reduced by intergroup contact. Changes wrought by contact are broad. The emotional components of prejudice are especially improved; but substantial improvements are also recorded for biased beliefs and negative stereotypes (Tropp & Pettigrew, 2005a).

The effectiveness of intergroup contact to lessen prejudice varies across different types of groups. Contact reduces prejudice against homosexuals and ethnic and racial groups significantly more than it reduces prejudice against the mentally ill. In addition, the impact of contact is less for minorities than for majorities (Tropp & Pettigrew, 2005b). Research suggests that minorities may less often than majorities view contact situations as involving equal status .

The policy implications of these worldwide findings are straightforward even if their implementation is difficult. Creating optimal intergroup contact situations in a society requires many of the same remedies needed to deter intergroup discrimination and conflict. This is hardly surprising for we are dealing with a tightly interwoven system of intergroup relations that involves prejudice, discrimination and conflict. So the same policies that foster optimal intergroup contact will also serve to decrease discrimination and conflict.

Four critical societal indices. Policy makers must keep a close eye on four important intergroup indices: measures of intergroup segregation in the labor force, intergroup residential segregation, intergroup educational segregation, and intergroup marriage. Increasing intergroup separation on two or more of these indices generally signals future intergroup conflict. Such trends also indicate increasing group discrimination and a dangerous decline in optimal intergroup contact situations. When groups live largely segregated existences residentially, educationally, and in the work force, optimal contact is severely restricted. Intergroup friendships are limited, and intergroup marriages remain rare and stigmatized.

Once again, Apartheid in South Africa and racial segregation in my native southern United States highlight this process. By making equal status contact and intermarriage illegal, these regions guaranteed massive discrimination and conflict. Although these systems of separation have formally ended, their negative legacies still distort current interracial relations. Hence, Black and White Americans are more than any other groups residentially segregated in modern urban America. And intermarriage between Blacks and Whites remains rare in a nation that otherwise witnesses rapidly increasing intermarriage rates. By 1992, half of all Native American marriages and almost a third of all Asian American marriages involved a White partner. The comparable figure for African Americans was only 6.5% (U.S. Bureau of the Census, 1998).

Group differences other than race also can lead to sharp separation and act to impede optimal contact. In Latin America, social class forms a key differentiating variable. In Western Europe, culture and citizenship become focal lines of division. A continent more accustomed to out- than inmigration has found it difficult to embrace new cultures in its midst. Prejudice, discrimination, anti-immigrant political parties, and violence have erupted throughout Europe for many years and finally hit our headlines recently after the riots and car-burnings throughout France and flagrantly anti-Muslim cartoons in a major Danish newspaper (Pettigrew, 1998b).

In addition to thwarting beneficial intergroup contact, intergroup separation triggers a series of interlocking processes that inflame group conflict. Negative stereotypes are magnified; distrust cumulates; and awkwardness typifies the limited intergroup interaction that does take place. The powerful majority comes to believe that segregated housing, low-skilled jobs and constrained educational opportunities are justified, even “appropriate,” for the minority.

Two broad policy generalizations emerge from this intergroup contact perspective: [1] Separation in housing, schools and employment limits intergroup contact and triggers a series of interlocking processes at the psychological, situational and societal levels that enhance intergroup prejudice, discrimination, and conflict. [2] Social policies that increase minority access to wider societal opportunity automatically create more optimal situations for intergroup contact. Such efforts then trigger a constructive cycle that leads to less intergroup prejudice, discrimination, and conflict.


A last word

In sum, intergroup conflict is a serious world problem. And prejudice is an important component in such conflict. Furthermore, we know the social policies that can reduce prejudice and improve intergroup relations. The unanswered question is whether the nations of the world have the will to pursue these policies.



Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. New York: Harper, 1950.        [ Links ]

Allport, G. W. (1954). The nature of prejudice. Reading, MA: Addison-Wesley.        [ Links ]

Altemeyer, B. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco, CA: Jossey-Bass.        [ Links ]

Altemeyer, B. (1998). The ‘other” authoritarian. In: M. P. Zanna (Ed.), Advances in experimental social psychology. New York, NY: Academic Press.

Eagly, A. H., & Dickman, A. B. (2005). What is the problem? Prejudice as an attitude-incontext. In J. F. Dovidio, P. Glick, & L. P. Rudman (Eds.), Reflecting on the nature of prejudice (pp. 263-277). Oxford, U.K.: Blackwell.        [ Links ]

Pettigrew, T. F. (1996). How to think like a social scientist. New York: Harper Collins, 1996.        [ Links ]

Pettigrew, T. F. (1998a). Intergroup contact theory. Annual Review of Psychology, 49, 65-85.        [ Links ]

Pettigrew, T. F. (1998b). Responses to the new minorities of Western Europe. Annual Review of Sociology, 24, 77-103.        [ Links ]

Pettigrew, T. F., & Meertens, R. W. (1995). Subtle and blatant prejudice in Western Europe. European Journal of Social Psychology, 57, 57-75.        [ Links ]

Pettigrew, T. F., & Meertens, R. W. (2001). In defense of the subtle prejudice concept: A retort. European Journal of Social Psychology, 31, 299-309.        [ Links ]

Pettigrew, T. F., & Tropp, L. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, in press.        [ Links ]

Pettigrew, T. F., Wagner, U., Christ, O., & J. Stellmacher, Y. (2006). Why does authoritarianism predict prejudice? The mediators of a global phenomenon. Manuscript under review.        [ Links ]

Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup theory of social hierarchy and oppression. New York, NY: Cambridge University Press.        [ Links ]

Tropp, L., & Pettigrew, T. F. (2005a). Differential relationships between intergroup contact and affective and cognitive dimensions of prejudice. Personality and Social Psychology Bulletin, 31 (8), 1145-1158.        [ Links ]

Tropp, L., & Pettigrew, T. F. (2005b). Relationships between intergroup contact and prejudice among minority and majority status groups. Psychological Science, 16, 651-653.        [ Links ]

U.S. Bureau of the Census (1998). Race of wife by race of husband: 1960, 1970, 1980, 1991, and 1992. Recuperado el 7 de mayo de 2003 de: htpp://        [ Links ]

Wagner, U., Christ, O., & Pettigrew, T. F. (2006). Prejudice and group-related behavior in Germany. Manuscript under review.        [ Links ]



Dirección para correspondencia
Psychology Dept., Social Sciences II
University of California, Santa Cruz
Santa Cruz, CA 95064 U.S.A.



1Conferencia presentada el 3 de setiembre de 2006 en la Universidad de Costa Rica, San José Costa Rica.

Creative Commons License