This article was written for the academic journals in 1987 but was not accepted for publication
DO WE STEREOTYPE STEREOTYPING? STEREOTYPING AND RACISM
University of N.S.W., Australia
Ethnocentrism theory suggests that anti-outgroup prejudice is a product of ingroup favoritism and that the view of outgroups will be monolithic -- i.e. only their "in-out" status will matter. The literature on stereotyping, however, shows that attitudes to outgroups are highly complex and differentiated -- particularly among prejudiced people -- and it also shows that attitude towards the ingroup is a poor predictor of attitude towards outgroups. Both major postulated aspects of ethnocentrism theory are thus undermined. It is further shown that there is considerable evidence for the view that stereotyping is adaptive and that stereotypes can have considerable truth value. Stereotyping is seen as a strategy of successive approximations towards valuable generalizations in an environment of restricted information. Stereotyping does not cause racism.
Race and racism have been and continue to be major factors in the politics of many nations. Gaining an understanding of racism is therefore one of the major tasks confronting political psychology. A notion that seems to figure prominently in most explanations of racism is the notion of stereotyping. It certainly seems to occur in almost all elementary psychology and social psychology textbook accounts of racism. It will be submitted here, however, that even a desultory reading of the social cognition literature pushes us towards the view that stereotyping neither causes racism nor has any useful role in its explanation.
A hard-to-dislodge theory about hard-to-dislodge theories
The history of the concept of stereotyping has often been reviewed by others (e.g. Bond, 1986; Driedger & Clifton, 1984) so will be treated summarily here. Suffice it to say that, as Weber & Crocker (1983) point out, the old Lippman view of stereotypes as being mythical, rigidly held and highly resistant to change still seems to be widely believed among psychologists. More careful writers (e.g. Allport, 1954) admit that stereotypes may often have a "kernel of truth" and one does also sometimes find the point made that there can be positive (laudatory) as well as negative (pejorative) stereotypes but these seem generally to be mere riders to the basic concept. The basic concept, then, is that stereotypes are hard-to-dislodge theories. Since the evidence that they are not is abundant and goes back a long way it is something of a paradox that this concept itself is so hard to dislodge.
Stereotyping and ethnocentrism theory
For a start, let us look at how stereotyping concepts interact with ethnocentrism theory as enunciated by Sumner (1906) and as elaborated by Adorno, Frenkel-Brunswik, Levinson & Sanford (1950). This theory predicts that people who place a high value on their own group will tend to scorn outgroups. In apparent confirmation of this, some writers appear to find that people with high (favorable) autostereotypes have low (unfavourable) heterostereotypes (e.g. Driedger & Clifton, 1984). Further, both ethnocentrism and stereotyping are seen as seriously damaging oversimplifications. Stereotyping might even be seen as essential to ethnocentrism. We should therefore be able to expect that the research findings on stereotyping confirm ethnocentrism theory. Stereotypes of outgroups should be found to be highly monolithic and undifferentiated. If all that matters about a group is its in-out status, the attitudes held towards various outgroups should be highly correlated. Further, autostereotypes should be strongly negatively correlated with heterostereotypes (i.e. positive sentiment towards the ingroup should strongly predict negative sentiment towards outgroups). What do we in fact find?
Outgroup sentiment a mirror of ingroup sentiment?
The study by Driedger & Clifton (1984) has already been mentioned. It is a large study wherein many comparisons between autostereotypes and heterostereotypes are possible. The authors draw fairly conventional conclusions from their findings but a look at what they actually found is instructive. From their Table III we find that the correlation between autostereotype and heterostereotype is in only one case out of 24 above .2 and is in general below .1. Thinking well of your own group, in other words, has virtually nothing to do with thinking ill of other groups. Similar conclusions can even be found in the group dynamics literature (on the occasions where the research is set up so that the question can be examined). For example: "Not only is ingroup favoritism in the laboratory situation not related to outgroup dislike, it also does not seem causally dependant on denigration of the outgroup" (Turner, 1978, p. 249). See also Brewer & Collins (1981, p. 350) and Brown, Condor, Matthews, Wade & Williams (1986). Public opinion polls using standard attitude scales also support this conclusion (e.g. Ray & Furnham, 1984; Heaven, Rajab & Ray, 1985 and Ray & Lovejoy, 1986). This is clearly an inauspicious start for ethnocentrism theory. Attitude to outgroups can be shown from many sources and research modalities not to be a mirror of attitude to the ingroup. There is, furthermore, a substantial body of thought which sees pro-ingroup sentiment as something like self-esteem --i.e. a positive influence and a basis for a healthy, adaptive and positive view of the world. It is hard to think well of outgroups if you do not think well of your own group. This view is most strongly put by sociobiological writers (e.g. Mihalyi, 1984/85) but is not confined to them. See Cairns (1982), Furnham & Kirris (1983) and Elwert (1982). This is all a very far cry from the simplistic conceptualizations of Sumner (1906) and Adorno et al (1950).
Truth in Stereotyping
We must not be surprised, therefore, when we find that other aspects of ethnocentrism theory come into question from stereotyping studies. The view that attitude to outgroups is monolithic also seems to be at variance with reality. Stereotyping studies in fact almost invariably find that responses to different ethnic groups are highly differentiated (e.g. Kippax & Brigden, 1967; Gallois, Callan & Parslow, 1982; Callan & Gallois, 1983; Houser, 1979). From outside the stereotyping literature see also Newman, Liss & Sherman (1983), Ray (1974) and Ray & Lovejoy (1986) -- where attitudes towards different outgroups are shown as only weakly correlated. In other words, what is believed of one ethnic group is not believed of others. Jews, for instance, are seen as different from blacks. It seems hard in retrospect to believe that anyone ever proposed otherwise but that appears to be the implication of classical ethnocentrism theory, at least as enunciated by Adorno et al (1950).
What is in fact found, then, is that people have highly consistent opinions about the characteristics of various ethnic groups but such opinions are not all the same -- e.g. Asians might be seen as "industrious" and blacks as "dirty". There also seems to be a not inconsiderable tendency for agreement between the in- and the out-group concerning the characteristics of the in-group. For instance, Callan & Gallois (1983) found that Anglo-Australians, Greek-Australians and Italian-Australians all showed a high level of agreement that Anglo-Australians were "sportsmanlike", "happy-go-lucky" and "pleasure-loving". As Triandis & Vassiliou (1967) have argued, this sort of occurrence tends to suggest considerable truth-value for the stereotypes concerned. See also Kippax & Brigden (1977), where Australian and American opinions about one-another and various other nationalities are shown to have a lot in common. Stereotyping research, in other words, gives a picture of attitudes to outgroups that is as complex as ethnocentrism theory would have us believe it to be simple. Where Adorno et al (1950) would have us believe that those with negative racial attitudes are very oversimplifying and undifferentiating in their attitudes, we find from authors such as Gallois, Callan & Parslow (1982) that people who discriminate on ethnic grounds do so in highly differentiated ways. The negative response (or even positive response) to a given ethnic group, has a lot to do with the different ways in which different groups are conceived. On matters of interest to him (i.e. racial and ethnic matters) the racially discriminatory person is cognitively complex rather than cognitively simple (See also Ray, 1972a). This is reminiscent of the finding that authoritarians are cognitively complex when matters of interest to them (i.e. matters concerning authority) are in question (Ray, 1972b).
Are positive stereotypes a good thing?
Perhaps the greatest element of complexity introduced to research on race-relations by stereotyping studies, however, is the role of positive stereotypes. One would think that a positive stereotype of a group would mean that all sorts of negative attitudes toward such a group would be minimal. This appears not to be so. Viljoen (1974) found that South African Blacks thought higher of English-speaking whites than they did of themselves yet those same blacks still liked their own group best in other ways. In particular, blacks preferred more social distance from the English-speaking whites than from other blacks. To put it plainly, the blacks thought that the English- speakers were admirable but still did not like the thought of their daughters marrying one. A positive stereotype went with greater rather than lesser social distance. Similarly, Miller (1985) found that older Australian schoolchildren (whites) who had large numbers of blacks (Australian Aborigines) in their classes resented black welfare programs most when they had positive stereotypes of blacks. Far from a positive stereotype of blacks implying that positive discrimination by the government in favour of blacks would be applauded, it meant that such interventions were resented. Conversely, the people who accepted affirmative action programs uncritically were those whites who thought very poorly of blacks (i.e. those whites who stereotyped blacks most unfavorably). It was "prejudiced" people who most accepted the need to help blacks overcome their handicaps!
The simple idea that positive stereotypes are good and negative stereotypes are bad is thus revealed as an oversimplification. After all, if you think that blacks are "pretty good", why should you think that they need help? Only if you think that blacks are "in a mess" would you think that the government should help them. Negative judgments of blacks can have positive results as far as black welfare is concerned.
This does tend to suggest that the characteristic American policy of stressing that blacks are basically equal to whites but still in need of special help from the government is in trouble. The two arms of policy may be pulling in opposed directions. It may be possible in the long run to gain public acceptance of only one of the policy arms -- not both at the same time. In other words, if you really wish to succeed in causing blacks to be perceived as basically equal to whites you may not be able to have affirmative action programs. Your insistence on the need for affirmative action programs, on the other hand, might tend to be seen as implying that you do not really at heart accept black equality. People do sometimes seem to have the habit of behaving in ways that do not suit theorists.
What goes on in stereotyping?
Since much of what has been found out via stereotyping research appears to be inconsistent with what many of us always thought we knew about racism, it surely behooves us at this point to take a closer look at the processes of stereotyping generally. What really does go on in stereotyping? Are stereotypes really as writers such as Simpson & Yinger (1965) suppose, rigid and exaggerated inventions that preclude recognition of individual variation? Or are they more benign and even useful?
At least as long ago as Schutz (1932) the benign functions of stereotypes have been stressed. Schutz pointed out that people seek to typify each other in social interactions in order to simplify their role-taking efforts. If you can categorize people, you have to make less effort in order to interact constructively with them. You do not have to "feel your way" so much. Among more recent writers, Berry (1970) is one of many who concede that stereotypes can indeed have a useful role. He found that stereotypes are an aid in accurately knowing what the key (i.e. different) traits of various groups are. He found, in short, that they are useful truths. They fulfil the function of enabling us to deal with difficult and ambiguous data. Oversimplifications can, in other words, have their place.
This can also be seen in the work of Eisenberg (1968). Eisenberg studied the fact that people will give descriptions of non-existent groups such as "Yurasians" and "Lagesi". Eisenberg found that these supposedly "nonsense" names were not nonsense at all and that they reminded respondents of various real outgroups ("Yurasians" = Eurasians?). Respondents in fact most of all seem to have concluded that the names were names of various primitive tribes. What this highlights is the subtlety of the cues that human beings use in dealing with their world. It is a great human strength that we can make great use of even the tiniest amounts of information. We use every aid we can to reduce the uncertainty in our world and hence to enhance our control over it. To put it plainly, people will stereotype at the drop of a hat (Read, 1983). In fact, Hamill, Wilson & Nisbett (1980) found that people will generalize from a single instance even when they are specifically told in advance that the instance concerned is an a-typical one.
Doing so, however, does not mean that some rigid mental structure has been adopted. Quite to the contrary, stereotypes are approximations. They are continually modified as information comes in and may be abandoned entirely as the situation to be dealt with becomes more highly specified. Thus Locksley and her colleagues (See Locksley, Hepburn & Ortiz, 1982) have done a series of studies which show that when a target person is being evaluated by Ss, the provision of case information about that individual target person will substantially reduce the role of stereotypes in the evaluation made of the target person by the Ss. Similarly, Galper & Weiss (1975) found that stereotyping was not used where the situation was more fully specified and Braithwaite, Gibson & Holman (1985-86) found that stereotyping diminishes as the experimental situation becomes more realistic. Further Forgas (1983) showed that information consistent with cultural stereotypes is more readily processed than information not so consistent.
The overall picture, then, is clearly one of stereotypes being useful and accurate up to a point. Beyond the point where better information than what is contained in the stereotype becomes available, however, the stereotype is steadily abandoned as a guide to action. Where stereotypes persist, however, are those situations where specific information will seldom be adequate or available soon enough. For instance, when confronted by an unfamiliar black, a white does not conclude that he has no information to guide him in the interaction. He instead uses his stereotypes (generalizations from past experience, both personal and vicarious) to guide his initial responses. If continued interaction shows those generalizations not to apply to the given black, the behaviour ceases to be guided by the generalizations concerned.
Thus a white who encounters a large black coming towards him on a dark street late at night will not normally approach the encounter with an empty mind. He may have a stereotype (founded in some reality) that blacks are often muggers and accordingly keep a very wary eye on the approaching black. If, however, the black simply says "Nice day" when he passes, the stereotype will no longer have any role in the interaction and some pleasantry in reply may be uttered. Racial antagonism does exist but blaming it on stereotyping is not at all justified. Stereotyping is simply what we all resort to when we have inadequate information to go on. Since most of us probably structure our world to reduce uncertainty and unpredictability wherever reasonably possible, such situations are probably rare for most of us in most of our lives and stereotyping should therefore have a minimal role for most of us for most of the time. Other studies which support this general view of the typical stereotyping process are not hard to find but perhaps special mention should be given to papers by Stein, Hardyck & Smith (1965), McCauley, Stitt & Segal (1980) and Bond (1986).
Perhaps the earliest research study supporting a view of stereotyping similar to that advocated in the present paper was by Bayton, McAlister & Hamer (1956). These authors described a person to students simply as "black" and got the usual stereotypes back: "dirty", "lazy" etc. They then modified the description to "educated black" and instantly got greatly changed responses. The educated black was in fact described in terms very similar to an educated white. So we see that, far from being rigidly held beliefs that stand in the way of recognition of individual variation, stereotypes are in fact supremely flexible and responsive to new information.
Stereotypes, generalizations and categorization
It may have been observed that the above discussion of stereotypes has attempted no distinction between stereotypes and valid generalizations. Valid generalizations must surely exist. The general view has seemed to be that stereotypes are invalid generalizations so how can we know on any given occasion what sort of generalization we are dealing with? The reason this problem was not raised earlier is that it seems generally to be slighted in the literature. The word "stereotype" seems on many past occasions to have been used as a synonym for any generalization about any group. The implication is that all generalizations in such fields are dangerous. This is, however, a thoroughly unscientific orientation. What is psychology if it is not an attempt to make generalizations about classes of human behaviour? To reject the possibility of generalization or categorization is to reject the possibility of science (and even of language).
As it happens, however, this confusion over what is or is not a valid generalization is rendered unimportant by the account of stereotyping given so far in this paper. It has been seen that it not a question of "either or". Rather, stereotyping is a process of successive approximation towards accurate judgments. The stereotype may start out containing very little in the way of accurate information but as knowledge of and experience with the particular class of person accumulates, the information will become progressively more accurate. Even an accurate generalization, however, can surely only ever be a preliminary guide to any interaction with a particular person so one would hope that use of the stereotype would soon fade away as the interaction progresses and as information about the specific person accumulates. The evidence indicates that this is exactly what happens (Locksley et al, 1982; Galper & Weiss, 1975; Braithwaite et al, 1985- 86; McCauley et al, 1980; Stein, Hardyck & Smith, 1965). Stereotypes are temporarily useful tools, not mental straitjackets.
Studies which show stereotypes as rigid
In case it should seem that an impossibly rosy picture of stereotyping has been presented so far, it should be acknowledged that stereotypes are not mindlessly fluid. There are a variety of studies (e.g. Pettigrew, 1979; Johnson & Judd, 1983; Darley & Gross, 1983) which show that new information is not always flawlessly integrated into new generalizations. Pettigrew, for example, speaks of the "attribution error" whereby a white man might see good behaviour by a black man as an exception to the rule while the same behavior by a white man would be seen as confirming the rule. Pettigrew attributes such cognitions to the inherent rigidity of stereotypes.
This is, however, unfair. Perfect generalizations are rare in any situation. There always seem to be at least some exceptions. This does not mean that we must abandon generalizations. We would have to abandon science if we did. What we do, therefore, is treat exceptions to any rule carefully. We do not immediately abandon or revise the rule but instead wait until several or maybe many exceptions build up. If blacks are generally seen by whites as lazy, one diligent black man will not disturb that stereotype. "One swallow does not make a summer". If, however, lots of diligent black men are encountered, cognitive change will eventuate (Weber & Crocker, 1983). Writers such as Pettigrew simply fail to consider adequately how many exceptions (to a rule) will be tolerated.
Stereotyping is not the cause of racism
We are now in a position to say, then, that the old mention of stereotyping in our psychology textbooks as one of the causes of racism is quite incautious. Stereotyping may be involved as a step in the formation of racially antagonistic attitudes but it is involved as a step in the formation of all attitudes. To say that stereotyping causes negative racial attitudes is to confuse the cause with the process. It is not only those who dislike other races who are stereotypers. We are all stereotypers.
Interesting support for this conclusion is to be found in a recent paper by Devine (1989). Devine showed that "tolerant" people do not differ in their awareness of stereotypes from non-tolerant people but that the tolerant people deliberately suppress their use of stereotypes. Tolerance has to be learned and deliberately practiced.
Perhaps one final paper that should be mentioned is that by Smith, Griffith, Griffith & Steger (1980). These authors studied stereotypes of Germans held by American students who had been living in Germany for some time. They found that the students had stereotypes that were realistic and positive and concluded that stereotyping is of little use in explaining racial and ethnic antagonisms.
Given the lack of usefulness that stereotyping may now be seen as having in explaining racial attitudes, it is reasonable to ask what the alternatives are. How do we explain racial attitudes? It must be conceded that all the theories that have been advanced so far do have serious problems (Record, 1983; Studlar, 1979; Ray, 1984). Perhaps the most hopeful line of enquiry for psychologists, however, may be those theories and findings which portray racial preferences as just another instance of a more general human tendency to prefer the familiar and thus to prefer people who are similar to themselves (Rokeach, 1960; Stein, Hardyck & Smith, 1965; Levine & Campbell, 1972; Liebowitz & Lombardo, 1980; Taylor & Guimond, 1978; Byrne, Clore & Smeaton, 1986; Marin & Salazar, 1985; Ray, 1983).
The present paper has of course been only one in a long line of studies that have moved towards revision of our notion of stereotyping. It is, however, perhaps the most radical of the theory papers in that it finally urges the complete irrelevance of stereotyping to any understanding of racial conflict. This is a decisive break with the past but one that may be long overdue. An example of the tenacity with which the old views are clung to is the paper cited earlier by Driedger & Clifton (1984). These authors interpreted correlations averaging around .1 as support for the notion that dislike of outgroups is the mirror of liking for the ingroup! With friends like that, the older theories hardly need enemies!
What the textbooks say
In the presence of so much evidence against the traditional view of how stereotypes work, someone might wonder whether this paper has any point at all. Is it not simply rehashing what has gone before? It is. Unfortunately, however, such "rehashing" seem needed. Let us in closing look at what some of the elementary textbooks are telling our students about stereotyping.
What we find is that the large band of people who write introductory psychology and social psychology textbooks seem generally not yet to have integrated well the findings discussed in the present paper. They seem generally to be like Pettigrew (1979) in seizing on any small sign of resistance to change in beliefs as an indication that stereotypes are hopelessly rigid. They tend to have "stereotyped" (in their sense, i.e. rigid) views of stereotyping. Two of the worst offenders write as follows: "In-group/out-group biases lead us to conclude that we are better than they are. Our stereotypes reinforce these biases, stand resolute against disconfirmation, and function as self-fulfilling prophecies" (Forsyth, 1987, p. 233); and "Stereotypes are a major mechanism in sustaining prejudice. Once people agree on prejudicial labels, such labelling becomes resistant to change" (Gergen & Gergen, 1986, p. 146).
While obsolete views are being purveyed to our students, therefore, there will surely be a continuing need for papers such as the present one.
Adorno,T.W., Frenkel-Brunswik, E., Levinson, D.J. & Sanford, R.N. (1950) The authoritarian personality. N.Y.: Harper.
Allport, G.W. (1954)The nature of prejudice. Cambridge: Addison Wesley
Bayton, J.A., McAlister, L.B. & Hamer, J. (1956) Race-class stereotypes. J. Negro Education 25, 75-78.
Berry, J.W.(1970) A functional approach to the relationship between stereotypes and familiarity. Australian J. Psychol. 22, 29-33.
Bond, M.H. (1986) Mutual stereotypes and the facilitation of interaction across cultural lines.International J. Intercultural Relations 10, 259-276.
Braithwaite, V., Gibson, D. & Holman, J. (1985/86) Age stereotyping: Are we oversimplifying the phenomenon? Internat. J. Aging & Human Development 22, 315-325.
Brewer, M.B. & Collins, B.E. (1981) Scientific enquiry and the social sciences San Fran.: Jossey Bass.
Byrne, D., Clore, G.L. & Smeaton, G. (1986) The attraction hypothesis: Do similar attitudes affect anything? J. Personality & Social Psychology 51, 1167-1170.
Cairns, E. (1982) Intergroup conflict in Northern Ireland. Ch. 10 in: H. Tajfel (Ed.) Social identity and intergroup relations Cambridge, U.K.: U.P.
Callan, V.J. & Gallois, C. (1983) Ethnic stereotypes: Australian and Southern European youth. J. Social Psychol. 119, 287-288.
Crocker, J. & Schwartz, I. (1985) Prejudice and ingroup favoritism in a minimal intergroup situation: Effects of self-esteem. Personality & Social Psychology Bulletin 11, 379-386.
Darley, J.M. & Gross, P.H. (1983) A hypothesis-confirming bias in labelling effects. J. Personality & Social Psychology 44, 20-33.
Devine, P.G. (1989) Stereotypes and prejudice: Their automatic and controlled components. J. Personality & Social Psychology 56, 5-18.
Driedger, L. & Clifton, R.A. (1984) Ethnic stereotypes: Images of ethnocentrism, reciprocity or dissimilarity? Canadian Rev. of Sociology & Anthropology 21, 287-301.
Eisenberg, S. (1968) Ethnocentrism and the face of the stranger. J. Social Psychol. 76, 243-247.
Elwert, G. (1982) Probleme der Auslaenderintegration. Gesellschaftliche Integration durch Binnenintegration? Koelner Zeitschrift fuer Soziologie und Sozialpsychologie 34, 717-731.
Forgas, J.P. (1983) The effects of prototypicality and cultural salience on perceptions of people. J. Res. Personality 17, 153-173.
Forsyth, D.R. (1987) Social psychology Monterey, Ca.: Brooks/Cole.
Furnham, A. & Kirris, R. (1983) Self-image disparity, ethnic identity and sex-role stereotypes in British and Cypriot adolescents. J. Adolescence 6, 275-292.
Gallois, C., Callan, V.J. & Parslow, L.A. (1982) Evaluations of four ethnic groups: Level of ethnocentrism, favourability, and social distance. Australian J. Psychol. 34, 369-374.
Galper, R.E. & Weiss, E. (1975) Attribution of behavioural intentions to obese and normal-weight stimulus persons. European J. Social Psychology 5, 425-440.
Gergen, K.J. & Gergen, M.M. (1986) Social psychology N.Y.: Springer
Hamill, R., Wilson, T.D. & Nisbett, R.E. (1980) Insensitivity to sample bias: Generalizing from a-typical cases. J. Personality & Social Psychology 39, 578-589.
Heaven, P.C.L., Rajab, D. & Ray, J.J. (1985) Patriotism, racism and the disutility of the ethnocentrism concept. J. Social Psychol. 125, 181-185.
Houser, B.B.(1979) Content and generality of young white children's ethnic attitudes. J. Social Psychol. 109, 69-77.
Johnson, J.T. & Judd, C.M. (1983) Overlooking the incongruent: Categorization biases in the identification of political statements. J. Personality & Social Psychology 45, 978-996.
Kippax, S. & Brigden, D. (1977) Australian stereotyping -- a comparison. Australian J. Psychol. 29, 89-96.
LeVine, R. & Campbell, D.T. (1972) Ethnocentrism: Theories of conflict, attitudes and group behavior N.Y.: Wiley.
Liebowitz, S.J. & Lombardo, J.P. (1980) Effects of race, belief, and level of prejudice on responses to black and white strangers. J. Social Psychol. 110, 293-294.
Locksley, A., Hepburn, C & Ortiz, V. (1982) On the effects of social stereotypes on judgments of individuals: A comment on Grant & Holmes's "The integration of implicit personality theory schemas and stereotypic images". Social Psychology Quarterly 45, 270-273
Marin, G. & Salazar, J.M. (1985) Determinants of hetero- and autostereotypes: Distance, level of contact and socioeconomic development in seven nations. J. Cross-cultural Psychol. 16, 403-422.
McCauley, C., Stitt, C.L. & Segal, M. (1980) Stereotyping: From prejudice to prediction. Psychological Bulletin 87, 195-208.
Mihalyi, L.J. (1984/85) Ethnocentrism vs. nationalism: Origin and fundamental aspects of a major problem for the future. Humboldt J. Social Relations 12(1), 95-113.
Miller, B. (1985) Resource competition, authoritarianism and contact as determinants of attitudes to Aborigines and Torres Strait Islanders. Dissertation for the Diploma of Sociology, James Cook University of N. Queensland, Townsville, Australia.
Newman, M.A., Liss, M.B. & Sherman, F. (1983) Ethnic awareness in children: Not a unitary concept. J. Genetic Psychol. 143, 103-112.
Pettigrew, T.F. (1979) The ultimate attribution error: Extending Allport's cognitive analysis of prejudice. Personality & Social Psychology Bulletin 5, 461-476.
Read, S.J. (1983) Once is enough: Causal reasoning from a single instance. J. Personality & Social Psychol. 45, 323-334.
Record, W. (1983) Race and ethnic relations: The conflict continues. Sociological Quarterly 24, 137-149.
Rokeach, M. (1960) The open and closed mind N.Y.: Basic Books. Schutz, A. (1932) The phenomenology of the social world Evanston, Ill.: Northwestern Univ. Press.
Simpson, G. & Yinger, J. (1965) Racial and cultural minorities: An analysis of prejudice and discrimination N.Y.: Harper.
Smith, R.J., Griffith, J.E., Griffith, H.K. & Steger, M.J. (1980) When is a stereotype a stereotype? Psychological Reports 46, 643-651.
Stein, D.D., Hardyck, J.A. & Smith, M.B. (1965) Race and belief: An open and shut case. J. Personality & Social Psychol. 1, 281-294.
Studlar, D.T. (1977) Social context and attitudes towards coloured immigrants. British J. Sociology 28, 168-184.
Sumner, W.G. (1906) Folkways N.Y.: Ginn. Taylor, D.M. & Guimond, S. (1978) The belief theory of prejudice in an intergroup context. J. Social Psychol. 105, 11-25.
Triandis, H.C. & Vassiliou, V. (1967) Frequency of contact and stereotyping. J. Pers. Social Psychol. 7, 316-328.
Turner, J.C. (1978) Social categorization and social discrimination in the minimal group paradigm. In: H. Tajfel (Ed.) Differentiation between social groups: Studies in the social psychology of intergroup relations European Monographs in Social Psychology, No. 14. London: Academic.
Viljoen, H.G. (1974) Relationship between stereotypes and social distance. J. Social Psychol. 92, 313-314.
Weber, R. & Crocker, J. (1983) Cognitive processes in the revision of stereotypic beliefs. J. Personality & Social Psychology 45, 961-977.