Attitudes and Norms

By Howard Schuman, Charlotte Steeh, Lawrence Bobo, and Maria Krysan

An excerpt from Racial Attitudes in America: Trends and Interpretations, winner of the 2005 AAPOR Book Award.

A frequent concern raised about racial attitudes is that they may represent little more than the superficial verbalization of socially approved norms. If it has become less acceptable over the past half century for whites to express negative attitudes toward blacks, then are we measuring how whites actually feel, or are we instead measuring how whites think they should report feeling, especially in the interview situation? The term "social desirability" is often used to characterize answers to survey questions that conform to current norms, rather than to what people truly feel. [There are] several kinds of evidence indicating that at least some responses to attitude questions do appear to vary depending on the nature of the interview situation. In one case, attitudes shift in relation to the degree of privacy provided to respondents, with some white attitudes becoming more negative toward blacks when the survey setting is more private. In another case, the assumptions that respondents make about an interviewer's attitudes appear to influence the respondents' own expression of attitudes.

The issue is more complex than this line of thinking suggests, however. It is useful to recognize that attitudes themselves can be distinguished as primarily normative in origin or primarily a matter of personal preference. For example, what to order from a restaurant menu can be considered largely a matter of personal taste, and each person is free to make his or her own choices within a wide range. But it would be naïve to think that the transformation of white racial attitudes over the last half century has occurred simply because a great many Americans have each altered their personal views. Up until at least the 1940s, segregation, discrimination, and openly verbalized prejudice toward minorities of all kinds were entirely acceptable throughout much of the United States. But today very few people would express open support for any of these. Norms calling for equal treatment regardless of race are now highly salient in America, not only in much of the legal structure, but in more intangible ways as well. In this sense, certain. trends that we report. for attitude questions can be regarded as revealing substantial changes in American norms.

Thus our book might well be called Racial Norms in America: Trends and Interpretations. This would not make the book any less significant. On the contrary, tracing such a momentous change in norms is a very important effort. The elimination of hotel advertisements stating "Only White Christians Welcome" and the inclusion in job advertisements of "Equal Opportunity Employer" are not trivial signs of a cultural transformation. Even if we do not entirely understand what has led to the changes or what their effects are, the trends say something quite meaningful about American life. But beyond that larger transformation, norms have real force, else why should a respondent care what an interviewer thinks? It is exactly the power of norms that is revealed when we say that a person gives a different answer because he or she is in the presence of an interviewer, especially since the same pressure is likely to occur in a number of other social situations.

Indeed, that norms are powerful is the paradoxical implication of certain social psychological experiments that were carried out to demonstrate just the opposite. For example, Gaertner and Dovidio (1986) show the importance of covert prejudice by drawing on Latane and Darley's (1970) classic finding that people are less likely to help in an emergency when other potential helpers are available than when they are alone. Gaertner and Dovidio "predicted that the belief that other bystanders are present would have a greater inhibiting effect on the subject's response when the emergency involved a black victim than when it involved a white victim. Failure to help a black person in this situation could be justified or rationalized by the belief that the victim is being helped by someone else. Bystanders believing themselves to be the sole witness, however, were not expected to discriminate against black victims relative to white victims. [because when a bystander is] alone, the failure to help a black victim could be more readily attributed to bigoted intent." Although the experiment is presented to show the hypocrisy that lurks behind a liberal front, it also points to the power of the norm of equal treatment to influence behavior when the norm is made salient (the "alone condition"). If the white subjects show their "true feelings" when normative forces are removed, then the condition that produces a different behavior when norms of equal treatment are brought to the fore tells us that norms are really quite important in shaping actions.

Still, there is the question of the degree to which norms are "internalized" and become personal attitudes, so that they operate even when interviewers or other observers are not present. There is good reason to think that this varies greatly across individuals, with some having made the norm an integral part of their personality and thus attempting to live in accord with it all of the time. For others, the norm functions more as an external constraint, shaping their behavior to the extent that they feel observed by those assumed to uphold the norm. But even for the latter people, we can see the norm as meaningful, since it does influence the way they act in certain public situations. It is by no means unimportant, for example, that many people try to use nonracist or nonsexist language because of external normative constraint, even if this does not reflect their inner preferences. In addition, widely supported social norms are available to be called on in critical situations, as the extremely negative nature of the charge of "racism"--whether deserved or undeserved--indicates.

From a larger standpoint, if there were not important changes in racial attitudes more generally, why should respondents assume interviewers to be so different in their expectations in the 1980s than in the 1940s? Norms do not exist in thin air, and in the absence of legal or other coercion they must receive some support from personal attitudes. Just as many attitudes are shaped by social norms, so individual attitudes support social norms by being called forth when there is a violation of the norm. The circularity of this statement is no accident: a norm can be thought of as a kind of collective attitude or evaluation, and its efficacy depends on its receiving sufficient support in the form of individual attitudes toward those who violate the norm. When attitudes at the individual level no longer support a norm--for example, when attitudes toward a presidential candidate are no longer affected by his having been divorced--then the norm itself is well on its way to disappearing.

Excerpted from Racial Attitudes in America: Trends and Interpretations, by Howard Schuman, Charlotte Steeh, Lawrence Bobo, and Maria Krysan. Revised edition. Copyright © 1997 by Harvard University Press. Reprinted by permission and posted electronically by special arrangement with Harvard University Press.