Ethnic Images

Tom W. Smith

National Opinion Research Center
University of Chicago

GSS Topical Report No. 19

December, 1990


NOT FOR RELEASE BEFORE 12:00 NOON EDST JANUARY 8TH, 1991

This research was done for the General Social Survey Project directed by James A. Davis and Tom W. Smith. The project is funded by the National Science Foundation, Grant No. SES-87-18467. I would like to thank Lawrence Bobo for his helpful comments.

In recent decades Americans have made considerable progress in the area of race relations. Lynching, de jure segregation, and Jim Crow laws have been abolished. Whites have become more supportive of integration and racial equality (Smith and Sheatsley, 1984; Schuman, Steeh, and Bobo, 1985; Jaynes and Williams, 1989; Smith, 1990). Governments have instituted numerous programs to promote integration and assist minority groups (e.g. busing, affirmative action, minority contracting). Yet almost totally missing from the extensive research that has been done on changes in racial tolerance, has been any study of changes in the stereotypes or, as we shall refer to them, images that people have towards ethnic groups.

In this paper we want to explore two questions: 1) what are the images that people have towards several, prominent ethnic groups on various dimensions or characteristics; and 2) do the images people have about ethnic groups influence other attitudes and behaviors toward the groups.

Measuring Ethnic Images

In our examination of ethnic images, we use "ethnic" as a general term to cover the six groups under examination (Whites, Jews, Blacks, Asian Americans, Hispanic Americans, and Southern Whites) which are defined partly by race, religion, nationality, and region. We utilize "images" rather than stereotypes or prejudices since we wish to avoid some of the baggage that is frequently associated with one or both of these terms. For example, stereotypes and prejudice are often assumed to contain a component of irrationality, including such fallacies as causal misattribution, improper generalization, excessive categorization, and rejecting or ignoring counter-evidence (Allport, 1953; Schuman and Harding, 1964; Jackman, 1973). These traits may well be part of the images we measure here, but we have no direct tests of that and therefore can not assume that irrationality is a notable element in our measures. Furthermore, stereotypes are also sometimes seen as projections of psychological states (e.g. as either id or superego based) and we do not wish to adopt this formulation (Bettelheim and Janowitz, 1950; Pettigrew, 1971). Thus, ethnic images are in general beliefs that people have about cultural groups (and their members) and in particular they are beliefs about group characteristics and attributes.

To measure ethnic images we developed a question that: 1) reduced the likelihood of giving offense, 2) facilitated the reporting of group characterizations, 3) permitted the expression of both positive and negative attributions, 4) allowed comparisons across various groups, and 5) included both in- and out-group evaluations of the reference groups. Both the general survey literature on social desirability and self-presentation effects and the specific literature on prejudice and stereotypes suggested that special care had to be taken in devising questions on ethnic images.

First, we avoided declarative statements of negative attributions as had often been done in the past (e.g. "The trouble with Jewish businessmen is that they are so shrewd and tricky that other people don't have a fair chance in competition" and "Generally speaking, Negroes are lazy and don't like to work hard." Marx, 1967; Selznick and Steinberg, 1969). The problem with using such statements is partly that their offensive nature may lead to a loss of rapport or even a break-off. In addition, because of offensiveness such statements were frequently not asked of the in-group members themselves, thus losing the ability to compare in- and out-group images.

Second, declarative statements discourage the reporting of groups differences because of their violation of norms of politeness and their often absolutist phrasing. Third, we wanted to allow the comparison of several groups on various image dimensions. Some formulations have avoided the problems of offensive declarations by asking whether Blacks or Whites were more likely to have some attribute (Apostle, Glock, Piazza, and Suelzle, 1983; Matthews and Prothro, 1966), but this approach does not readily facilitate multi-group comparisons. Fourth, we wanted to allow people to express positive as well as negative feelings towards a group. Because of their apparent repercussions (discrimination, minority persecution, etc.), negative stereotypes have been given more attention than more general group depictions covering negative, neutral, and positive evaluations. This focus is clearly unbalanced and ignores that fact that many groups are rated positively on at least some dimensions.

To achieve these goals, we developed an instrument that asked people to rate whether people in the designated group were mostly closer to one or the other of two polar statements (e.g. Rich/Poor). (See Appendix for the complete wordings.) The opposing images were fixed at points 1 and 7 with intermediate points of 2, 3, 4, 5, and 6. Point 4 is defined as meaning "you think that the group is not towards one end or another." This allowed people to place a group at any point along the continuum. It also allowed the study of the comparative positioning of groups by studying where people rated one group vs. other groups.

We studied six ethnic groups that covered some of the major cultural groups in American society: Whites, Jews, Blacks, Asian Americans, Hispanic Americans, and Southern Whites. Clearly it would have been desirable to have both covered certain other groups (e.g. American Indians and other religions) and to have separated sub-groups within our broad categories (e.g. Mexicans, Puerto Ricans, Cubans, etc. among Hispanic Americans), but time constraints necessitated selecting only a few, prominent groups.

Similarly, we would have liked to include more than six characteristics. There are many important general characteristics (e.g. honesty, sexuality, frugality) that we might have added. In addition, we might have included specific characteristics that are mostly (but not exclusively) associated with one group in the public's mind (e.g. Jews and business skills, Asians and Hispanics and family, or Blacks and athletics). Still, we were able to cover a number of important dimensions (Wealth: Rich/Poor; Work Ethic: Hard-Working/Lazy; Violence: Violence-Prone/Not Violence Prone; Intelligence: Unintelligent/Intelligent; Dependency: Self-Supporting/Live Off Welfare; and Patriotism: Patriotic/Unpatriotic) that touch upon common-place and vital images held about ethnic groups in contemporary society. In selecting this list, we primarily chose the factual, wealth dimension as our first dimension because we thought that it would be relatively easy for people to rate groups on this dimension and that by getting people used to the idea of rating groups it would improve response to the other, more personality-related characteristics. We chose work ethic, dependency, violence, and intelligence because both traditional and contemporary stereotypes of minorities include these as prominent factors (Bettelheim and Janowitz, 1950; Apostle, Glock, Piazza, and Suelzle, 1983; Karlins, Coffman, and Walters, 1969; Devine, 1989; Pettigrew, 1971). Patriotism is a slight reformulation of the traditional image about the "foreignness" of various ethnic groups and of the basic in-group/out-group dichotomies that are central of all ethnic evaluations.

In our analysis of ethnic images, we took the ratings that people gave Whites and subtracted from it the score they gave each of the other five groups. For example, if a person rated Whites as 3 on wealth and rated Jews as 2 and Blacks as 5, we calculated a Jewish wealth difference score of +1 and a Black score of -2. Scores could range from +6 to -6 (although because Whites were usually rated near the middle, few maximum difference scores actually occurred). For each characteristic, we coded the dimension so that a positive score meant that a group was rated closer to the positive image (Rich, Hard-Working, Not Violence-Prone, Intelligent, Self-Supporting, and Patriotic) than Whites and a negative score meant that a group was rated more towards the negative images than Whites.

We chose to use difference scores between the ratings of Whites and the various other ethnic groups primarily because we were interested in the comparative positioning and advantages that groups were seen as having. In addition, while the difference scores and absolute ratings were substantially correlated, with correlations usually in the .5 to .8 range, the difference scores seemed to perform somewhat better as predictors. When we correlated the absolute and difference scores with measures of racial interactions, racial attitudes, and national images, we found that the difference scores generally had slightly higher correlations than the absolute scores (higher in 63 of 95 correlations).

Ethnic Images in Contemporary America

The belief that Americans are approaching a color and creed-blind society is easily disabused by the ethnic image data collected on the 1990 General Social Survey (Table 1). First, these scores show that people are willing and able to rate group members on the basis of their ethnicity. Blacks seem to be the easiest minority group to evaluate with item non-response almost half as high as for other groups, while Asian Americans receive slightly higher non-response than other groups. We suspect this is partly a function of the size and prominence of these (and the other) groups.

Second, with one exception, minority groups are evaluated more negatively than Whites in general. The one exception is Jews who are rated more favorably than Whites on each characteristic except patriotism. No other group scores above Whites (i.e. has a positive mean) on any characteristic. Looking at how everyone rates groups (including in-group members), we see that Jews are rated most positively overall (first on wealth, industry, non-violence, intelligence, and self-support and third on patriotism - second among minorities). Asian Americans and White Southerners are ranked next (second or third) on almost every dimension. Finally, Blacks and Hispanic Americans are ranked last or next to last on almost every characteristic. Looking at a scale that sums up scores on all items, except the more factually grounded wealth dimension (Table 2), we see that Jews are the only positively rated group (+0.75). Southern Whites (-2.32) and Asian Americans (-2.65) are rated immediately below Whites and Hispanic Americans (-5.70) and Blacks (-6.29) were rated considerably lower. In fact, over 80% of respondents rated Hispanic Americans and Blacks lower than Whites on one or more of the five characteristics.

If we look at only how out-groups rate the ethnic groups, the general pattern is for the exclusion of the in-group to lower the overall rating (except for some wealth ratings). For example, among everyone Blacks where the mean rating of Blacks on being Hard-Working/Lazy was -1.24. The mean rating was -1.35 by non-Blacks and -0.40 by Blacks. This does little to change the overall rankings however.

People see the most inter-group variation (differences in the range of group means) on the socio-economic variables of Self-Support/Welfare, Rich/Poor, and Hard-Working/Lazy. Thus at least in terms of this limited range of variables, people see ethnic groups differing the most on class-related attributes. They see Jews as excelling on the status and achievement variables and other minorities as falling well below the White standards. Next, in terms of inter-group variation comes violence and intelligence. On these variables Jews still exceed the White standard and other minorities fall below, but the spread is not as great. Finally, for patriotism the variation is the smallest. It is also the only dimension on which Whites surpass all other groups.

In/Out-Group Differences

One way of checking the validity of the image scale is to compare the ratings of in- and out-group members. On the five character dimensions in-groups uniformly rate themselves higher than they are rated by out-group members. For Jews and Asians however there are two few in-group members (respectively 27 and 9) for meaningful analysis. Non-Blacks however rate Blacks significantly lower on each dimension except patriotism than Blacks rate themselves. Usually the difference is rather large. For example, 56.7% of non-Blacks rate the intelligence of Blacks below that of Whites, while only 29.6% of Blacks rated themselves lower than Whites. For Hispanic Americans the in/out-group differences are usually not significant, but they are actually similar in direction and magnitude as the Black/White comparisons. For example, only 35.1% of Hispanic Americans rate themselves as less intelligent than Whites, while 55.3% of non-Hispanics rate Hispanic Americans as lower in intelligence than Whites. A similar pattern also applies for White Southerners.

The Impact of Images

The impact of ethnic images can be quite notable, as the following six multiple regressions illustrate (Table 3). In each case, we have selected one or more image items concerning a group, other relevant independent variables, and a dependent variable with a substantive tie to the group and characteristic on the ethnic image item(s). The first two regression use an image scale that combined the Hard-Working/Lazy and Self-Supporting/Welfare dimensions. Concerning both support for special treatment of Blacks by the government ("affirmative action") (3.A) and support for more governmental spending for Blacks (3.B), images about the work ethic of Blacks are significantly related to the programs net of the other variables. A further analysis (not shown), indicated that considering all images about Blacks (i.e. adding patriotism, violence, and intelligence to the two work ethic items) added little or nothing to the predictive power of images.

The next two regressions show the relationship between images about the intelligence and violence proneness of Blacks on support for sending a child to a school with from a few to a majority of the students of the opposite race (Blacks are asked about Whites and Non-Blacks are asked about Blacks) (3.C) and school busing (3.D). Images about intelligence and violence were significant predictors of willingness to send a child to an integrated school, but were not predictive of support for school busing. This suggests that support for school integration is partly shaped by racial images, but that support for busing notably differs (note also the change in the education coefficients) and does not appear to be based on racial images.

The final two regressions show the relation between images of domestic, ethnic groups and foreign countries. In Table 3.E net of other variables liking Israel is significantly related to seeing Jewish Americans as patriotic. Similarly, 3.F indicates that liking Japan is significantly related to believing that Asian Americans are patriotic. Presumably if the ethnic image had focused on Japanese Americans, the association would have been stronger.

Ethnic images also are related to the social distance that people wish to maintain between themselves and other groups (Table 4). As the summary image scale for each group moves from positive to negative, people are less favorable towards living in a neighborhood where half of neighbors are from particular groups and less favorable to having a close relative marry a group member. While all of the associations are significant, the relationships between images and social distance for Jews and White Southerners are modest, while the association for Blacks, Asian Americans, and Hispanic Americans are much more substantial.

Finally, at least in regards to Blacks (the only group for which there is information), there is little relationship between the images that non-Blacks hold and their contacts with Blacks (Table 5). While more positive attitudes are related to living near Blacks, having had a Black as a dinner guest, and having Blacks attend your church, none of the associations are significant. We believe that this is primarily related to two factors: opportunity and receptivity. In many cases not having contact with Blacks merely reflects the absence of Blacks in the locality (e.g. many Northern rural areas, some suburban areas, etc.) and not any intention to avoid Blacks. Conversely, certainly on the integrated neighborhood question and to a lesser extent on dinner guests and church integration, the proximity of Blacks might represent an undesired contact, rather than a desired contact.

In sum, images about ethnic groups are significant predictors of support for racial integration programs, attitudes toward foreign countries, and desired social distance. The lack of an association with school busing indicates however that racial images do not necessarily play a significant role on all race-related issues. The lack of association between ethnic images and Black/White contact also shows that race relations is a complex topic to cover and simple assumptions (e.g. inter-group contact means group tolerance and positive images) may not be warranted.

Conclusion

Despite the demonstrable progress in inter-group tolerance over the last several decades, ethnic images are still common place in contemporary society. On the whole these images are neither benign nor trivial. Most Americans see most minority groups in a decidedly negative light on a number of important characteristics. Only images of Jews are generally positive and even in their case people seem to question their patriotism. All other groups, including the old stock, White Southerners, are seen more negatively than Whites in general. In particular, Hispanic Americans and Blacks receive very low ratings.

These negative ethnic images in turn help shape attitudes towards civil rights and racial integration policies, social distance, ratings of countries, and presumably other group-related issues. However, as the busing example shows, not all group-related issues are necessarily driven by ethnic images.

Ethnic images remain important determinants of inter-group attitudes. They are crucial components of public opinion on such issues as affirmative action, school desegregation, and many other group-related issues and contemporary inter-group relations can not be understood without understanding the images that people have in their minds about the various ethnic groups that make up America.

Table 1

Images of Groups Compared to Whites^a

Characteristic	Group	Mean	Distributionb			Missing
			-	0	+
Rich/Poor	Jews	+0.58	12.8	37.4	49.8	9.1
	Blacks	-1.60	83.2	13.0	3.8	5.3
	Asians	-0.77	52.8	30.9	16.3	10.6
	Hisps.	-1.64	83.4	10.6	6.0	9.5
	So.Whts.	-0.56	46.6	41.2	12.2	9.0
Hard-Working/ Lazy	Jews	+0.38	12.5	47.6	39.9	9.0
	Blacks	-1.24	62.2	31.9	5.9	5.8
	Asians	-0.19	34.2	35.8	30.3	11.4
	Hisps.	-0.99	54.1	37.2	8.7	10.0
	So.Whts.	-0.52	38.8	52.1	9.1	10.1
Violence-Prone/ Not Violence-Prone	Jews	+0.36	12.0	55.2	32.9	11.1
	Blacks	-1.00	56.1	30.0	13.9	6.9
	Asians	-0.15	29.8	45.0	25.1	13.3
	Hisps.	-0.75	49.5	34.0	16.5	10.8
	So.Whts.	-0.23	28.3	56.0	15.7	11.6
Unintelligent/ Intelligent	Jews	+0.15	11.8	76.3	6.9	9.3
	Blacks	-0.93	53.2	40.5	6.3	6.9
	Asians	-0.36	36.3	44.6	19.1	12.3
	Hisps.	-0.96	53.5	40.1	6.4	10.6
	So.Whts.	-0.54	38.4	55.4	6.2	10.4
Self-Supporting/ Live-Off Welfare	Jews	+0.40	9.1	53.2	37.7	8.4
	Blacks	-2.08	77.7	20.4	1.9	5.5
	Asians	-0.75	46.4	37.4	16.2	12.2
	Hisps.	-1.72	72.4	23.7	3.9	10.4
	So.Whts.	-0.71	44.5	49.1	6.5	11.2
Unpatriotic/
Patriotic	Jews	-0.57	34.4	60.3	5.4	11.3
	Blacks	-1.03	50.6	46.6	2.7	9.6
	Asians	-1.16	55.2	38.6	6.2	14.5
	Hisps.	-1.34	60.4	35.6	4.0	12.9
	So.Whts.	-0.31	27.4	61.2	11.3	11.3
aThe scores are based on subtracting the rate assigned to Jews, Blacks, Asian Americans, Hispanics Americans, and Southern Whites from the White rate. All scales are scored so that negative means closer to the unfavorable characterization (poor, lazy, violence-prone, unintelligent, preferring to live-off welfare, and unpatriotic). Thus, if Whites were scored 4 on Rich/Poor and Blacks 5 the score on the Rich/Poor scale for Blacks above would be -1.0.
bThese percentages are based on the exclusion of missing responses. The % missing is given in the last column.

Table 2

Summary Group Difference Scores^a

Group	Mean	Distribution
		-	0	+
Jews	+0.75	25.4	50.4	35.1
Blacks	-6.29	84.7	11.4	4.0
Asian Americans	-2.65	60.1	14.5	25.4
Hispanic Americans	-5.70	83.0	12.6	4.4
Southern Whites	-2.32	61.2	25.7	13.1
aSum of group difference on Hard-Working/Lazy, Violence, Intelligence, Self-Supporting/Welfare, and Patriotism.

Table 3

Multiple Regression Analyses of the Impact of Ethnic Images

A. For Affirmative Action for Blacks (HELPBLK)
	Independent Variables (High Value)	Beta	Prob.
	Govt. Help Poor (Should)	.243	.0000
	Race (Not Black)	-.226	.0000
	Image about Blks. on Work & Welfare (+)	.192	.0000
	Political Ideology (Liberal)	.162	.0000
	Education (More Years of Schooling)	.117	.0002
	Region (Not South)	.075	.0169
	Party Identification (Democrat)	.046	.1687
	Age (Older)	.022	.4790
		r2=.28	(783)
B. More Governmental Spending for Blacks (NATRACE + NATRACEY)
	Race (Not Black)	-.242	.0000
	Image about Blks. on Work & Welfare (+)	.193	.0000
	Govt. Help Poor (Should)	.185	.0000
	Education (Years of Schooling)	.138	.0000
	Region (Not South)	.105	.0018
	Political Ideology (Liberal)	.063	.0712
	Party Identification (Democrat)	.029	.4263
	Age (Older)	.025	.4651
		r2=.22	(740)
C. No Objection to Integrated School (RACFEW+RACHAF+RACMOST)
	Race (Not Black)	-.169	.0000
	Image about Blks. on Intelligence (+)	.102	.0010
	Image about Blks. on Violence (+)	.087	.0044
	Education (Years of Schooling)	.054	.0660
	Age (Older)	-.048	.0991
	Party Identification (Democrat)	.035	.2341
	Region (Not South)	.025	.3823
		r2=.07	(1183)
D. For Busing to Integrate Schools (BUSING)
	Race (Not Black)	-.197	.0000
	Age (Old)	-.125	.0000
	Party Identification (Democrat)	-.112	.0001
	Region (Not South)	-.066	.0205
	Education (Years of Schooling)	.021	.4775
	Images about Blks. on Violence (+)	.028	.3540
	Images about Blks. on Intelligence (+)	-.011	.7225
		r2=.07	(1195)
E. Liking of Israel (ISRAEL)
	Liking of Egypt (+)	.515	.0000
	Religion (Not Jewish)	.161	.0000
	Image about Jews on Patriotism (+)	.151	.0000
	Political Ideology (Liberal)	.085	.0134
	Age (Older)	.061	.0568
	US World Activity (Active)	-.046	.1590
	Fundamentalism (Liberal)	-.045	.1915
	Region (Not Northeast)	.023	.4749
	Education (Years of Schooling)	-.021	.5338
	Party Identification (Democrat)	.014	.6880
	Foreign Aid Spending (More)	-.002	.9616
		r2=.35	(664)
F. Liking of Japan (JAPAN)
	Education (Years of Schooling)	.162	.0000
	Age (Older)	-.128	.0006
	Image about Asians on Patriotism (+)	.118	.0015
	US World Activity (Active)	.078	.0381
	Party Identification (Democrat)	.073	.0645
	Foreign Aid Spending (More)	.053	.1478
	Region (Not South)	-.021	.5742
	Political Ideology (Liberal)	.009	.8094
		r2=.07	(703)

Table 4

Summary Images and Social Distance^a

(prob. /gamma)
Groups	Marriage	Neighbors
Jews	.0000/-.051	.0000/-.114	(1152/1154)
Blacks	.0000/-.374	.0000/-.354	(1202/1208)
Asian Americans	.0000/-.247	.0000/-.257	(1112/1111)
Hispanic Americans	.0000/-.340	.0000/-.350	(1146/1151)
Southern Whites	.0000/-.069	.0000/-.138	( 762/ 764)b
aSummary scale of images on Hard-Working/Lazy; Violence, Intelligence, Self-Supporting/Welfare, and Patriotism associated with two social distance measures listed below.
bOnly people outside the South were asked about Southern Whites.
Neighbors: Now I'm going to ask you about different types of contact with various groups of people. In each situation would you please tell me whether you would be very much in favor of ithappening, somewhat in favor, neither in favor nor opposed to it happening, some what opposed, or very much opposed to it happening?
Living in a neighborhood where half of your neighbors were Jews/Blacks/Asian Americans/Hispanic Americans/Whites raised in the (North/South)?
Marriage: What about having a close relative or family member marry a Jewish person? Would you be very in favor of it happening, somewhat in favor, neither in favor nor opposed to it happening, somewhat opposed, or very opposed to it happening?
A Jewish person/A Black person/An Asian-American person/A HispanicAmerican person/A White raised in the (North/South)?

Table 5

White Images About Blacks and Contact with Blacks

(Prob. /gamma)
	Live Near Blacks	Black Dinner Guest	Blacks Attend Church
Summary Image Scale	.071/.115	.131/.147	.079/.126
	(1019)	( 717)	( 585)

Appendix: Question Wording

Now I have some questions about different groups in our society. I'm going to show you a seven-point scale on which the characteristics of people in a group can be rated. In the first statement a score of 1 means that you think almost all of the people in that group are "rich." A score of 7 means that you think almost all of the people in the group are "poor." A score of 4 means you think that the group is not towards one end or another, and of course you may choose any number in between that comes closest to where you think people in the group stand.

A.	1	2	3	4	5	6	7
	Rich						Poor
	1. Where would you rate whites in general on this scale?
	2. Jews?
	3. Blacks?
	4. Asian Americans?
	5. Hispanic Americans?
	6. Southern Whites?
B.	The second set of characteristics asks if people in the group tend to be hard-working or if they tend to be lazy.
C.	The next set asks if people in each group tend to be violence prone or if they tend not to be violence prone.
D.	Do people in these groups tend to be unintelligent or tend to be intelligent?
E.	Do people in these groups tend to prefer to be self-supporting or do they tend to prefer to live off welfare?
F.	Do people in these groups tend to be patriotic or do they tend to be unpatriotic?

References

Allport, Gordon, The Nature of Prejudice. Reading, MA: Addison- Wesley, 1954.

Apostle, Richard A.; Glock, Charles Y.; Piazza, Thomas; and Suelzle, Marijean, The Anatomy of Racial Attitudes. Berkeley: University of California Press, 1983.

Bettelheim, Bruno and Janowitz, Morris, Social Change and Prejudice. New York: The Free Press, 1964.

Campbell, Angus, White Attitudes Towards Black People. Ann Arbor: Institute for Social Research, 1971.

Devine, Patricia G., "Stereotypes and Prejudice: Their Automatic and Controlled Components," Journal of Personality and Social Psychology, 56 (January, 1989), 5-18.

Glock, Charles Y.; Wuthnow, Robert; Piliavin, Jane Allyn; Spencer, Metta, Adolescent Prejudice. New York: Harper & Row, 1975.

Harris, Louis and Associates, A Study of Attitudes Toward Racial and Religious Minorities and Toward Women. New York: Harris Report, 1978.

Jackman, Mary R., "Education and Prejudice or Education and Response-Set," American Sociological Review, 38 (June, 1973), 327-339.

Jackman, Mary R. and Crane, Marie, "'Some of My Best Friends are Black...': Interracial Friendships and Whites' Racial Attitudes," Public Opinion Quarterly, 50 (Winter, 1986), 459-486.

Jaynes, Gerald David and Williams, Robin M., Jr., eds., A Common Destiny: Blacks and American Society. Washington, D.C.: National Academy Press, 1989.

Lipmann, Walter, Public Opinion. New York: Macmillan, 1922.

Martire, Gregory and Clark, Ruth, Anti-Semitism in the United States: A Study of Prejudice in the 1980s. New York: Praeger, 1982.

Matthews, Donald R. and Prothro, James W., Negroes and the New Southern Politics. New York: Harcourt, Brace, and World, 1966.

Pettigrew, Thomas F., Racially Separate or Together? New York: McGraw-Hill, 1971.

Quinley, Harold E. and Glock, Charles Y., Anti-Semitism in America. New York: The Free Press, 1979.

Schwartz, Mildred A., Trends in White Attitudes towards Negroes. Chicago: National Opinion Research Center, 1967.

Schuman, Howard and Harding, John, "Prejudice and the Norm of Rationality," Sociometry, 27 (Sept., 1964), 353-371.

Schuman, Howard; Steeh, Charlotte; and Bobo, Lawrence, Racial Attitudes in America: Trends and Interpretations. Cambridge: Cambridge University Press, 1985.

Selznick, Gertrude J. and Steinberg, Stephen, The Tenacity of Prejudice: Anti-Semitism in Contemporary America. New York: Harper & Row, 1969.

Simpson, George Eaton and Yinger, J. Milton, Racial and Cultural Minorities: An Analysis of Prejudice and Discrimination. 5th edition. New York: Plenum Press, 1985.

Smith, Tom W., and Dempsey, Glenn R., "Ethnic Social Distance and Prejudice," Public Opinion Quarterly, 47 (Winter, 1983), 584-600.

Smith, Tom W. and Sheatsley, Paul B., "American Attitudes Towards Race Relations," Public Opinion, 7 (Oct./Nov., 1984), 14-15, 50-53.

Williams, Robin M., Jr., Strangers Next Door: Ethnic Relations in American Communities. Englewood Cliffs, NJ: Prentice-Hall, 1964.

Yankelovich, Skelly, and White, Anti-Semitism in the United States. New York: American Jewish Committee, 1981.