The substantial difference between these two lists of traits supports the hypothesis that asking respondents solely about women in general generates images that do not hold for minority women. In addition, the traits selected to describe black women suggest that Anglo students primarily view black women as threats: loud, aggressive, argumentative, stubborn, and so on. This image, it should be noted, strongly resembles the image of “black matriarchs,” although without specifying black men as the victims of these threats. Although Table II provides useful data regarding the proportion of Anglo students who believe that various individual traits characterize black women, it cannot tell us how students mentally combine these traits into more complex images of black women. For this purpose, we ran a factor analysis to determine which combinations of traits respondents most often selected as characteristic of black women. Using maximum-likelihood extraction and orthogonal rotation and seeking the most parsimonious solution, three factors emerged: a threatening factor (consisting of loud, dishonest, and argumentative), a “good mother/wife/daughter” factor (intelligent, family-oriented, and loyal to family ties), and a “welfare mother” factor (too many children, fat, and lazy). Virtually all respondents (95%) selected one or more of the traits which comprise the threatening factor, 28% selected one or more of the good mother/wife/daughter traits, and 19% selected one or more of the welfare mother traits.
Neither sex, age, religion of origin, current religion, being a born-again Christian, nor self-described political orientation was significantly associated with any of these three factors. The racism and sexism scales, however, were significantly correlated with the three factors. Both racism and sexism were negatively correlated with the good mother/wife/daughter factor (for racism, r = -.2158, p |is less than~ .001; for sexism, r = -.2647, p |is less than~ .001) and positively correlated with the welfare mother factor (for racism, r = .3229, p |is less than~ .001; for sexism, r = .2020, p |is less than~ .001). The threatening factor, however, was not significantly associated with either sexism or racism.
Both sexism and racism are highly correlated with each other (r = .6334, p |is less than~ .001). To determine whether sexism had an impact independent of racism, we used multiple regression analysis. When both variables were entered into the equation, only racism significantly correlated with characterizing black women as welfare mothers and only sexism significantly (and negatively) correlated with characterizing black women as good mothers/wives/daughters (Table III).(6) Racism and sexism in combination, as when investigated separately, were not correlated with characterizing black women as threatening. For all three factors, adding the interaction between racism and sexism as a third independent variable did not significantly affect the results.
TABULAR DATA OMITTED
Up to this point, our analysis relied solely on the revised Katz/Braly scale and on our first survey for dependent variables. A major limitation of this scale, however, is its implicit assumption that a given trait has the same meaning when used to describe different groups. Data from the second survey enabled us to test this assumption. We calculated the mean rating for each trait among the 25 respondents whose target group was black women and among the 54 whose target group was American women in general (with “positive” given a score of 1, “neutral” a score of 2, and “negative” a score of 3). We then used t tests for differences of means to compare the average rating of each trait for black women black women with the average rating of that trait for American women in general.
The differences in average ratings were significant for 10 traits (Table IV). A striking pattern emerged: of these 10 traits, the 5 rated most positively for women in general– determined, attractive, assertive, independent, and meditative- -were all considered less positive for black women. At the same time, the 5 considered most negative for women in general–very religious, loud, asexual, having too many children, and weak- -were all considered less negative for black women. Because these data are based on a small sample, their reliability is limited.
However, the consistency of the results adds to their credibility, as does the fact that data we collected on attitudes towards Mexican and Jewish women yielded identical patterns (Weitz, 1992).
Further analysis suggested that the same people who rate black women ess negatively on negative traits rate them less positively on positive traits. Scores on the negative traits are negatively correlated with scores on the positive traits, i.e., as individuals’ scores on the negative traits go down or become less negative, their scores on the positive traits go up or become less positive (r = -.1690, p |is less than~ .05). Perhaps because of the small size of this second survey, however, these results are not statistically significant and these findings have to be considered tentative.
Our findings suggest two conclusions: that Anglo students assign different emotional evaluations to the same trait depending on whether they are characterizing women in general or black women and that Anglo students believe that black women are generally characterized by a different and substantially more negative set of traits than women in general.
TABULAR DATA OMITTED
The first of these two conclusions is the more unexpected and requires the most explanation. We would suggest that Anglo students consider positive traits less positive for black women and negative traits less negative because of their expectations regarding how black women should and would behave. Those who have low expectations for black women’s behavior–expecting them to be weak or to have too many children, for example–will not be surprised when black women meet those expectations. Consequently, they will not judge these traits less harshly when exhibited by black women, perhaps believing at some level that black women cannot help acting in these ways. Conversely, such respondents will be not only surprised but disturbed when black women act independently, assertively, or even thoughtfully (one possible interpretation of meditative).
An alternative explanation would hold that Anglo students rate black women less harshly when characterized by negative traits because these students want to “given them the benefit of the doubt.” Anglo students may believe that social factors make it more likely that black women will, for example, have many children, be loud, or be weak. They thus may not blame black women who exhibit such traits and, thus, may rate them less harshly than other women who exhibit the same traits. Logically, however, such respondents should also view black women more positively than others when they exhibit positive traits, in the belief that black women would have had to overcome substantial odds to do so. As the data show, however, the reverse is true. Thus, the first explanation for the discrepancy in trait ratings seems the more parsimonious and hence more likely explanation.
The particular image that emerged most strongly in this study, in both the analysis of frequencies and the factor analysis, is the image of black women as threatening. This image was so common that even those who otherwise appeared nonracist and nonsexist on our measures still subscribed to it.
Although the image of black women as threatening does not have deep historic roots, the image of black men as threatening is an old and familiar one in American iconography. For decades, whites used this image to justify lynchings of black men (Hall, 1979; Zangrando, 1980). More recently, George Bush masterfully manipulated this image, in the form of Willie Horton, to bolster his 1988 presidential campaign. The image of black women as threatening, which appears in our students’ responses, probably draws in part on ideas about the threatening nature of black men. Anglo students who have never encountered a clear, specific, and widely held stereotype of black women may have extrapolated beliefs about black women from popular ideas about black men.
The image of black women as threatening also may reflect our respondents’ status as recent high-school graduates. Adolescence is probably the age during which whites and blacks are most likely to harm each other. Given the absence of true integration in this country, most of these students will have graduated either from high schools with virtually no blacks (and hence assumed, in the absence of a popular image of black women, that the image of black men also applied to black women) or from high schools which were racially torn. The latter schools are often segregated enough so that one has few if any classmates or friends of other races but integrated enough so that one does encounter members of other races in public areas. In these situations, both blacks and whites may fear and sometimes encounter physical harm from the other.
Finally, the image of black women as threatening may derive in part from the myth of the black matriarchy. The entertainment industry has often portrayed black women as domineering, beginning at least as early as the Amos and Andy radio show, with its “Sapphire” character (Sims-Wood, 1988; Ely, 1991). The news media, meanwhile have reinforced this idea by publicizing social scientific research on the black matriarchy and the supposed death of the black family. Students’ ideas about black women may thus stem in part from these sources.
In contrast, the good mother/wife/daughter image seems the descendant of the “Mammy” or “Aunt Jemima” stereotype. Whereas white culture in the past, however, had depicted Mammy and Aunt Jemima as always and willingly taking care of others’ families, the good mother/wife/daughter image depicts them taking care of their own families.
The third image that emerged in this study, of black women as welfare mothers, has more obvious roots in contemporary American culture. Stories of poor black mothers, often unmarried and on welfare, appear frequently in newspapers–probably more often than any other images of black women. Whether or not reporters intend it, white readers often seem to conclude from these stories that most black women not only live in poverty, but also deserve their poverty. To many white American readers, the problems of poor black women seem to stem from their uncontrolled sexuality, which leads them to give birth often and early, and to their laziness, which leads them to rely on welfare rather than to better themselves through education or a job.
The problems that black women face when they are viewed as threats or as welfare mothers are obvious. In contrast, the image of black women as good mothers, wives, and daughters, at first glance, seems benign. Yet this image, too, may handicap black women in white society, for it stands in striking contrast to Anglo students’ image of American women in general. That latter image places American women not within the family but within the broader world–career-oriented, ambitious, independent, and intelligent (the latter trait chosen by these Anglo students almost twice as often for American women in general as for black women). Although in some ways positive, therefore, the image of black women as good wives, mothers, and daughters will not help black women who aspire to success in the public sphere.
The emotional evaluations of traits taken from the second survey also suggest some pessimistic conclusions. Several decades ago, Robert K. Merton (1957) described how ethnic groups can be “damned if they do and damned if they don’t”: condemned as pushy or sly when they succeed despite discrimination and as lazy or stupid when they do not.
Thus an Englishman gets labeled “thrifty,” while a Jew who behaves identically gets labeled “stingy.” Similarly, various cognitive psychologists have suggested that majority-group members often attribute minority successes to luck or other situational factors, while attributing minority failures to laziness, stupidity, or other internal factors (Greenberg & Rosenfield, 1979; Hamilton, 1979; Pettigrew, 1979).
These theories suggest that whites will view black women who succeed (or, in our terms, exhibit positive traits) less positively than they would equally successful whites, since white observers would attribute black women’s successes to forces outside the women’s control rather than to hard work, intelligence, and so on. The theories leave open whether those who fail will be judged more harshly (because they are presumed to have had only themselves to blame for their failures) or less harshly (because those failures were caused by purported insurmountable, inborn weaknesses). Our data support the latter interpretation.
These data have significant implications for black women’s ability to succeed. Our data come only from college students. It is from the ranks of these students, however, that future teachers, employers, and others who hold positions of authority will be drawn. Extrapolating from this study, therefore, we would hypothesize that white teachers and supervisors, for example, will less often reward young black women for their successes than reward young white women and will less often chastise black women for their failures. As a result, black women’s incentive to achieve may decline. Moreover, those black women who do achieve, whether in school or the marketplace, will likely find that their achievements either are not rewarded or are even held against them by whites as signs that these black women are stepping out of their “places” or benefiting undeservedly due to good luck.
Previous research has provided us with considerable data regarding the nature and sources of cultural images of minority groups and of women. No longer, however, can we assume that ethnic images hold across gender lines or that gender images hold across ethnic lines. We are currently conducting qualitative interviews to obtain more detailed data on image of minority women. Our future plans call for conducting parallel survey research with broader samples on other groups of minority women and investigating images of minority women within various minority communities. We hope, for example, to study stereotypes of black women among both black men and black women. Such research will enable us to investigate the extent to which gender and ethnic groups reject or internalize oppressive cultural images.
2 To whom correspondence should be addressed at Department of Sociology, Arizona State University, Tempe, AZ 85287-2101.
3 In addition, Bayton and Muldrow (1968) provide some minimal data about images of black women. Most of their article, however, focuses on how black respondents’ skin color affects their assessment of light-skinned and dark-skinned blacks.
4 We used the term “black” rather than “African American” on this scale because it is the term most often used by Anglos. The term “African American” might have suggested to respondents that the researchers held “politically correct” or liberal views and, thus, might have decreased respondents’ willingness to give “politically incorrect” answers. Because we used the term “black” in this survey, we use it in this paper to increase clarity and consistency.
5 Although a scale based on more items might have been a stronger measure, scales based on only three items are common in sociological research and provide stronger measures of a given concept than do individual items.
6 Because our purpose was to investigate whether racism and sexism each affect the dependent variables and are worth studying in future analyses, our focus was on the betas rather than the explained variance. We recognize, however, that the explained variance is small and that the final word on this awaits a better specified model.
Astin, A. W., Korn, W. S., & Berz, E. R. (1989). American freshman: National norms for fall 1989. Los Angeles: American Council on Education, Cooperative Institutional Research Program, University of California.
Baptiste, D. A. (1986). The image of the black family portrayed by television: A critical comment. Marriage and Family Review, 10, 41-63.
Bayton, J. A., & Muldrow, T. R. (1968). Interacting variables in the perception of racial personality traits. Journal of Experimental Research in Personality, 3, 39-44.
Broverman, I., Vogel, S. R., Broverman, D. M., Clarkson, F. E., & Rosenkrantz, P. S. (1972). Sex role stereotypes: A current appraisal. Journal of Social Issues, 32, 119-133.
Clark, M. L., & Pearson, W. (1982). Racial stereotypes revisited. International Journal of Intercultural Relations, 6, 381-393.
Collins, P. H. (1990). Black feminist thought: Knowledge, consciousness, and the politics of empowerment. Boston: Unwin Hyman.
Eagley, A. H., & Kite, M. (1987). Are stereotypes of nationalities applied to both men and women? Journal of Personality and Social Psychology, 53, 451-62.
Ely, M. P. (1991). The adventures of Amos and Andy. New York: Free Press.
Frazier, E. F. (1947). The Negro family in the United States. Chicago: University of Chicago Press.
Gordon, L. (1986). College student stereotypes of blacks and Jews on two campuses: Four studies spanning 50 years. Sociology and Social Research, 70, 200-201.
Gordon, L. (1991). Race relations and attitudes at Arizona State University. In P. G. Altbach & K. Lomotey (eds.)., The Racial Crisis in American Higher Education (pp. 233-248). New York: State University of New York Press.
Greenberg, J., & Rosenfield, D. (1979). Whites’ ethnocentrism and their attributions for the behavior of blacks: A motivational bias. Journal of Personality, 47, 643-657.
Gutman, H. (1976). Black family in slavery and freedom. New York: Pantheon.
Hall, J. D. (1979). Revolt against chivalry: Jessie Daniel Ames and the women’s campaign against lynching. New York: Columbia University Press.
Hamilton, D. L. (1979). A cognitive-attributional analysis of stereotyping. Advances in Experimental Social Psychology, 12, 53-84.
Hooks, B. (1981). Ain’t I a woman: Black women and feminism. Boston: South End Press.
Katz, D., & Braly, K. (1933). Racial prejudice and racial stereotypes. Journal of Abnormal and Social Psychology, 30, 175-193.
King, M. C. (1973). The politics of sexual stereotypes. Black Scholar, 4, 12-23.
Landrine, H. (1985). Race x class stereotypes of women. Sex Roles, 13, 65-75.
Maykovich, M. K. (1972). Stereotypes and racial images–white, black, and yellow. Human Relations, 25, 101-120.
Merton, R. K. (1957). Social theory and social structure. New York: Free Press.
Miller, A. G. (1982). Historical and contemporary perspectives on stereotyping. In A. G. Miller (ed.), In the eye of the beholder: Contemporary issues in stereotyping (pp. 1-40). New York: Praeger.
Morton, P. (1991). Disfigured images: The historical assault on Afro-American women. Westport, CT: Greenwood.
Moynihan, D. P. (1965). The Negro family: The case for national action. Washington, DC: GPO.
Ogawa, D. M. (1971). Small group communication: Stereotypes of black Americans. Journal of Black Studies, 2, 273-281.
Pettigrew, T. F. (1979). The ultimate attribution error: Extending Allport’s cognitive analysis of prejudice. Personality and Social Psychology Bulletin, 5, 461-476.
Ruble, D. N., & Ruble, T. L. (1982). Sex stereotypes. In A. G. Miller (ed.), In the eye of the beholder: Contemporary issues in stereotyping (pp. 188-252). New York: Praeger.
Sims-Wood, J. (1988). The black female: Mammy, Jemima, Sapphire, and other images. In J. Smith (ed.), Images of blacks in American culture (pp. 235-256). Westport, CT: Greenwood.
Staples, R. (1973). Black women in America: Sex, marriage, and family. Chicago: Nelson-Hall.
Stephan, W. G., & Rosenfield, D. (1982). Racial and ethnic stereotypes. In A. G. Miller (ed.), In the eye of the beholder: Contemporary issues in stereotyping (pp. 92-136). New York: Praeger.
Weitz, R. (1992, August). College students’ stereotypes of black American, Mexican American, and Jewish American women. Paper presented at the meetings of the American Sociological Association, Pittsburgh, PA.
Wilkinson, D. (1987). The doll exhibit: A psycho-cultural analysis of black female role stereotypes. Journal of Popular Culture, 21, 19-29.
Zangrando, R. L. (1980). The NAACP crusade against lynching, 1909-1950. Philadelphia: Temple University Press.