t . .. ; gag 3.3 an; . L‘ «9.8%.. I: fr . 4- a V ‘ . 5;... . .. . f . , ‘ ymawfi & . h 1.. 4 5.: 31.131! .uanV £.‘, 5. In“. «E i. .. .fium .. i i no fr .5; 1:... ..; : aqua? .. .i. 9.5....9unn . mm« 3 a ”up... Alina ‘5' 51?... a: . ‘ ,. am“... 7W1“: L. i. m 16.3. .- fiw.’ hrwhnwnuk a... 395.1 .‘ «We»... £1 ‘ ‘ ‘ y 1kg? vi: stain . “Mm” fit. 9 :&...(W5. Jig. ..., Inf 1 . nu: gal} 52,11. 5. . «3.. . 3 \ II kl {.2713 .2 V am»: 5., .nr .5 3 ,3. \\ b» I tyviIAA'u. ill 3 v .5 .413. nuuufiflfiifi: , clwlal‘vclifilmz ans. . A 7 . .J . fur I in?! . ol}:.fal I“ g v). ,%%a§a?% fl I: 11014.. I .3 .Hu. a". try 6%.. “15:3. 1 3006 LIBRARY Michigan State University This is to certify that the dissertation entitled Blacks’ Racial Attitudes: Perceived Racial Context, Racial Threat, and Self-Reported Prejudice presented by David C. Wilson has been accepted towards fulfillment of the requirements for the Doctoral degree in Political Science \ \ i Q“ lip. a..." i; Major P/olessor’s Signature i l; (0“ ‘4..—--—' 'c. I c 0.: MSU is an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE I DATE DUE DATE DUE _§EP 33mm) L8 2/05 cfimjndd-DJS BLACKS" RACIAL ATTITUDES: PERCEIVED RACIAL CONTEXT, RACIAL THREAT AND SELF-REPORTED PREJUDICE By David C. Wilson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Political Science 2005 ABSTRACT BLACKS’ RACIAL ATTITUDES: PERCEIVED RACIAL CONTEXT. RACIAL THREAT AND SELF-REPORTED PREJUDICE By David C. Wilson The study of racial attitudes in America has mainly been focused on the attitudes and opinions of Whites. This is particularly true for research on racial prejudice. Historically. Blacks have been the targets of racial prejudice in American; yet. there is little doubt that Blacks are also prejudiced towards Whites. Bearing this in mind, there are a still number of unanswered questions surrounding Blacks attitudes towards Whites because there are very few studies. or national data sets addressing this prospect. Using three independent years of data collected by the Gallup Polling Organization. I study potential aspects of Black prejudice. I test the proposition that Black prejudice is a response to perceived racial threats. which are shaped by racial perceptions and experiences signaling Blacks social. economic. and racial standing relative to Whites. The data not only show that are Blacks who perceive a racial threat more likely to be prejudice. but also that the group most likely to be prejudice are those least expected: the Black middle class. Of the demographic categories assessed, the highest income and most educated Blacks were most likely to perceive a negative racial context, perceive that Whites arc prejudiced. and self-report that they. themselves. are prejudiced. The results imply that Black prej udicc differs in very important ways from White prejudice. and that perceived solutions for resolving problems in race relations should possibly be reconsidered. Dedicated to Wilson Q. Welch iii ACKNOWLEDGMENTS The completion of this dissertation and consequently, my doctoral studies, would not have been possible without the encouragement. mentoring, support. and wisdom of many individuals and groups. Over the course of the past fifteen years, I have journeyed from an undergraduate, unsure of my interest and taken out of school in support of one War -- Operation Desert Stomi -- to a scholarly thinker. certain of my interests, one year returned from a second War -- Operation Iraqi Freedom. This is my opportunity to thank the individuals and entities most responsible for this successful journey. First, to my dissertation committee. thanks for enduring many questions about my topic. and providing me with sound professional feedback to support my future research ideas. Thanks to my chairperson. Darren W. Davis. and committee members, Paul Abramson, Brian Silver, and Richard Hula. Special thanks to Ric, who made it possible for me to attend Michigan State University by helping me secure financial support, and also encouraged my development in Urban Affairs Programs. The hours logged to complete this study have been greatly supported by the Gallup Organization. Gallup allowed me unfettered use of the Gallup Race Relations data, as well as went to great lengths to make sure that my research findings would be supported and heard throughout the company. In temis of individuals, I would like to thank Troy Arnold, Susan Nugent, Max Larsen, Jack Ludwig. Gale Muller, and Ted Hayes for their support and encouragement throughout. I would especially like to thank Eric P. Olesen, who read this manuscript and provided thoughtful comments and encouragement throughout. iv As an undergraduate at Western Kentucky University (WKU). l was influenced by three individuals. George Bluhm. now deceased. was the first professor to personally take interest in my potential, and refused to allow me to accept mediocrity as an academic standard. Howard Bailey. Dean of Student Life. was a role model. and professional constant throughout my tenure at WKU. His guidance was greatly under-appreciated by the students at WKU. but he was always professional and forthright. and I thank him for being a source of wisdom. Finally. I would have never found an enduring interest in Political Science were it not for Saundra Ardrey, who both set a high bar for academic standards. and provided the best opportunities for future growth in scholarship. She introduced me to many topics that to this moment I apply to my daily thinking about women. minorities, and politics. I would like to thank Michigan State University for their support. Over the years, the Department of Politic Science has provided excellent support throughout my tenure. most notably Dr.s Paula Keams. Kenneth Williams. and Larry l-Ieiman. Also. the former Urban Affairs Program provided me with outstanding intemship. assistantship, and research opportunities. Special thanks to the following: Maxie Jackson, Rex LaMore. and Kenetha McFadden. Cynthia Jackson-Elmore. who as an Assistant Professor, took a supportive interest in my scholarly and professional development stands as the main reason for my becoming a doctoral student. Cynthia. thank you for your mentorship. I would like to thank my family for their support. including my mother and father, Eton and Jean Wilson. my sister, Natalie Wilson. my aunt Jackie Welch. and my past and present grandparents. Ann S. Welch, Mary McKelvy. Angela Wilson, Eton R. Wilson. Sr., and Wilson Q. Welch. I would like to thank my best friend Samuel C. Watkins for his thoughtful support, and the brothers ofthe Epsilon Rho Chapter. of Kappa Alpha Psi Fraternity. for giving me an additional reason not to fail. I would also like to thank some researchers in the field of racial attitudes who have greatly influenced my work and interests. Thanks to Darren W. Davis. James Jones. Lawrence Bobo. Howard Schuman. David Sears. and Maria Krysan. None of this would be possible without my mentor. and friend, Darren W. Davis. who has also looked out for my interests and encouraged me to stay focused on the prize. Darren is a true scholar who sacrifices a lot for the profession. his family. and his peers. Thanks Darren, I look forward to many partnerships in the future. Finally, my wife Rosalind and my son Dalind provide my primary motivation for being. They have very willingly given me the time and space I needed to complete both my research, and this doctoral program. They are the reason I am optimistic about anything that exists in this world. and I owe them both more than I can ever give. vi TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. ix LIST OF FIGURES .......................................................................................................... xii LIST OF FIGURES .......................................................................................................... xii KEY TO ABBERVIATIONS .......................................................................................... xiii BLACK AMERICANS AND RACIAL ATTITUDES ....................................................... 1 Limited Research on Blacks‘ Prejudice Attitudes .................................................. 5 Blacks Racially Prej udiced Attitudes Towards Whites .......................................... 8 Blacks’ Awareness of Their "Group Position" ....................................................... 9 Blacks’ Responses to Prejudice ............................................................................ 10 Prejudice as a Response to Perceptions of a Negative Racial Environment ......... 12 Perceived Threat and Intolerance .......................................................................... 14 Focus and Goals of this Research ......................................................................... 16 The Concept of Prejudice ...................................................................................... l7 THEORY ........................................................................................................................... 20 Competing Theories of Blacks’ Prejudice ............................................................ 24 Ethnocentrism ....................................................................................................... 24 Interracial Contact ................................................................................................. 26 Cultural Mistrust ................................................................................................... 27 Racism Reactions to Racial Experiences .............................................................. 28 Sense of Group Position and Racial Threat .......................................................... 31 The Plan of the Analysis ....................................................................................... 35 General Hypotheses .............................................................................................. 35 The Data ................................................................................................................ 36 KEY MEASURES ............................................................................................................. 41 Blacks’ Prejudice .................................................................................................. 41 Black Preference/Ethnocentrism/In-Group Bias ................................................... 43 Perceived Racial Threat ........................................................................................ 46 Perceived Negative Context .................................................................................. 49 Overall Negative Context ..................................................................................... 55 Race of Interviewer ............................................................................................... 57 Demographics ....................................................................................................... 60 Summary ............................................................................................................... 62 A DEMOGRAPHIC PROFILE OF BLACKS’ PREJUDICE ........................................... 63 Demographics and Racial Attitudes ...................................................................... 63 Studying Blacks’ Racial Attitudes: Demographics and Prejudice ........................ 65 Perceived Racial Environment as Context ............................................................ 69 Segmenting the Black Population ......................................................................... 71 Data Analysis ........................................................................................................ 72 vii Education .............................................................................................................. 74 Income ................................................................................................................... 76 Employment Status ............................................................................................... 77 Age ........................................................................................................................ 79 Urbanicity ............................................................................................................. 81 Black Density ........................................................................................................ 84 Gender ................................................................................................................... 86 Gender and Education ........................................................................................... 88 Demographic Profile of Blacks’ Prejudice ........................................................... 90 The Role of Negative Context .............................................................................. 94 SENSE OF GROUP POSITION, THREAT, AND BLACKS PREJUDICE .................. 101 Theoretical Perspective: Sense of Group Position and Racial Threat ................ 104 Hypotheses .......................................................................................................... l 05 The Relationship between Threat and Context ................................................... 105 Blacks‘ Prejudice ................................................................................................ 108 Prejudice and Context ......................................................................................... 1 1 l Prejudice and Threat ........................................................................................... 1 l 1 The Relationship between Context, Prejudice, and Threat ................................. l 12 Multivariate Models of Prejudice ....................................................................... 1 14 General Conclusions ........................................................................................... 1 19 HOW BLACK ATTITUDES TOWARDS WHITES VARY WITH RACIAL COMPOSITION .............................................................................................................. 122 Hypotheses .......................................................................................................... 127 Data and Measures .............................................................................................. 128 Analysis ............................................................................................................... 131 Multivariate Analysis .......................................................................................... 140 Discussion and Conclusions ............................................................................... 143 DISCUSSION AND CONCLUSION .............................................................................. 147 Limitations .......................................................................................................... 155 APPENDIX A .................................................................................................................. 157 REFERENCES ................................................................................................................ 1 6O viii LIST OF TABLES TABLE 1. Data Characteristics for Black Respondents; 1997-1999 Gallup Race Relations Social Audit ....................................................................................................... 38 TABLE 2. Distribution of Self-Reported Prejudice Item; 1997-1999 Gallup Race Relations Social Audit ....................................................................................................... 42 TABLE 3. Percentage of Blacks indicating they are Prejudiced: 1997-1999 Gallup Race Relations Social Audits ................................................................................. 42 TABLE 4. Blacks Ethnocentrism; 1997 Gallup Race Relations Social Audits ................ 44 TABLE 5. Correlation between Blacks Preference and Blacks’ Prejudice; 1997 Gallup Race Relations Social Audits ................................................................................. 46 TABLE 6. Distribution of Perceived Racial Threat Item; 1997-1999 Gallup Race , Relations Social Audit ....................................................................................................... 47 TABLE 7. Percentages of Blacks indicating Whites are Prejudice; 1997-1999 Gallup Race Relations Social Audits ................................................................................. 48 TABLE 8. Negative Context - Percentage of Blacks indicating Group Unfair Treatment; 1997-1999 Gallup Race Relations Social Audits ............................................ 51 TABLE 9. Negative Context - Percentage of Blacks indicating Individual Unfair Treatment; 1997-1999 Gallup Race Relations Social Audits ............................................ 53 TABLE 10. Negative Context - Percentage of Blacks indicating Perceived Limits on Social and Economic Opportunities; 1997-1999 Gallup Race Relations Social Audits ................................................................................................................................. 54 TABLE 11. Distribution of Responses to the Negative Context Index; 1997—1999 Gallup Race Relations Social Audits ................................................................................. 56 TABLE 12. Actual Race of Interviewer (AROI): Distribution of Interviews by the Interviewers Race; 1997-1999 Gallup Race Relations Social Audits .......................... 58 TABLE 13. Perceived Race of Interviewer (PROI): Distribution of Interviews by the Interviewers Race; 1997-1999 Gallup Race Relations Social Audits .......................... 59 TABLE 14. Gallup Race Relations Social Audit Demographics and Measurement Levels by Year; 1997-1999 Gallup Race Relations Social Audits .................................... 60 TABLE 15a. Education Category by Racial Context Variables; 1997-1999 Gallup Race Relations Data ............................................................................................... 75 ix TABLE 15b. Education Category by Perceived Threat. and Prejudice: 1997-1999 Gallup Race Relations Data ............................................................................................... 75 TABLE 16a. Income Category by Racial Context Variables; 1997-1999 Gallup Race Relations Data ........................................................................................................... 77 TABLE 16b. Income Category by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data ............................................................................................... 77 TABLE 17a. Employment Status by Racial Context Variables; 1997-1999 Gallup Race Relations Data ........................................................................................................... 78 TABLE 17b. Employment Status by Perceived Threat. and Prejudice; 1997-1999 Gallup Race Relations Data ............................................................................................... 78 TABLE 18a. Age Category by Racial Context Variables; 1997-1999 Gallup Race Relations Data .................................................................................................................... 80 TABLE 18b. Age Category by Perceived Threat. and Prejudice; 1997-1999 Gallup Race Relations Data ............................................................................................... 80 TABLE 193. Urbanicity by Racial Context Variables; 1997-1999 Gallup Race Relations Data .................................................................................................................... 82 TABLE 19b. Urbanicity by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data ........................................................................................................... 82 TABLE 19c. Urbanicity by Education Categories; 1997-1999 Gallup Race Relations Data .................................................................................................................... 83 TABLE 19d. Urbanicity by Income Categories; 1997-1999 Gallup Race Relations Data .................................................................................................................... 83 TABLE 20a. Black Density by Racial Context Variables; 1998-1999 Gallup Race Relations Data .................................................................................................................... 84 TABLE 20b. Black Density-Urban by Racial Context Variables; 1998-1999 Gallup Race Relations Data ............................................................................................... 85 TABLE 20c. Black Density by Perceived Threat. and Prejudice; 1997-1999 Gallup Race Relations Data ............................................................................................... 86 TABLE 20d. Black Density-Urban by Perceived Threat, and Prejudice; 1998- 1999 Gallup Race Relations Data ...................................................................................... 86 TABLE 21a. Gender by Racial Context Variables; 1997—1999 Gallup Race Relations Data .................................................................................................................... 87 TABLE 21b. Gender by Perceived Threat. and Prejudice; 1997-1999 Gallup Race Relations Data ........................................................................................................... 88 TABLE 22a. Most Prejudiced Segment and Negative Context; 1997-1999 Gallup Race Relations Social Audit .............................................................................................. 97 TABLE 22b. Least Prejudiced Segment and Negative Context; 1997-1999 Gallup Race Relations Social Audit .............................................................................................. 98 TABLE 23. Correlations between Negative Racial Contexts and Perceived Racial Threat; 1997-1999 Gallup Race Relations Social Audit .................................................. 107 TABLE 24. Percentage of Blacks who Perceive Whites as Prejudiced For Levels of Context; 1997-1999 Gallup Race Relations Social Audits ......................................... 108 TABLE 25. Percentage of Blacks indicating they are Prejudiced; 1997-1999 Gallup Race Relations Social Audits ............................................................................... 1 10 TABLE 26. Logistic Regression Odds Ratios for Regression of Prejudice Response on Selected Demographics, Context, and Perceived Threat; 1997-1999 Gallup Race Relations Social Audits ............................................................................... 1 15 TABLE 27. Correlations between Racial Population Context and Individual Level Racial Perceptions; 1997 Gallup Race Relations Social Audit ............................. 132 TABLE 28. Central Tendency Measures for Key Race Items: 1997 Gallup Race Relations Social Audits .................................................................................................... 136 TABLE 29. Logistic Regression Odds Ratios for Regression of Prejudice Response; 1997 Gallup Race Relations Social Audits .................................................... 141 xi LIST OF FIGURES FIGURE 1. Percentage of Prejudice Respondents By Education and Gender: 1997-1999 Gallup Race Relations Social Audit ............................................................... 89 FIGURE 2. Percentage of Prejudice Respondents By Income and Gender; 1997- 1999 Gallup Race Relations Social Audit ........................................................................ 89 FIGURE 3. Most Prejudiced Segment; 1997-1999 Gallup Race Relations Social Audit ................................................................................................................................. 92 FIGURE 4. Middle Prejudiced Segment: 1997-1999 Gallup Race Relations Social Audit ...................................................................................................................... 92 FIGURE 5. Least Prejudiced Segment; 1997-1999 Gallup Race Relations Social Audit ................................................................................................................................. 93 FIGURE 6. Demographic Segmentation Results 1997-1999 Gallup Race Relations Social Audit ...................................................................................................... 95 FIGURE 7. Conceptual Model ofBlacks‘ Prejudice towards Whites ........................... 105 FIGURE 8. Percentage of Blacks Who are Prejudiced By Negative Context: 1997-I999 Gallup Race Relations Social Audits ............................................................ l 1 1 FIGURE 9 - Percentage of Blacks Who are Prejudiced By Perceived Threat: 1997-1999 Gallup Race Relations Social Audits ............................................................ l 12 FIGURE 10. Proportion of Respondents Who Are Prejudiced By Context and Threat; 1997-1999 Gallup Race Relations Social Audits ............................................... 1 13 FIGURE 1 1. Proportion of Prejudice Respondents by Perceived Group Size; 1997 Gallup Race Relations Social Audit ...................................................................... 134 xii KEY TO ABBERVIATIONS RDD — RANDOM DIGIT DIALING GRRSA — GALLUP RACE RELATIONS SOCIAL AUDIT ROI — RACE OF INTERVIEWER AROI — ACTUAL RACE OF INTERVIEWER PROI — PERCEIVED RACE OF INTERVIEWER xiii Chapter 1 BLACK AMERICANS AND RACIAL ATTITUDES Relative to Whites. the public opinions. and prejudice attitudes of Black Americans have been ignored in the literature on racial attitudes and American politics. While there is little doubt of the relationship between Whites’ negative racial attitudes towards Blacks. and their public opinionns about race relations and race based policies (Bobo. 2001; Bobo and Klugel, 1993; Jones. 1997; Kinder and Sanders, 1986; McConahay, 1986), Blacks’ negative attitudes towards Whites are rarely analyzed. except for a handful of studies (e.g.. Livingstone, 2002; Marx, 1967; Monteith and Spicer, 2001; Noel. 1964; Paige. 1970: Sigelman and Welch, 1991; Snidemtan and Piazza, 2002; Wojniusz, 1979). Studying Black Americans’ prejudice towards Whites aids our understanding of how target group attitudes and opinions can shape perceptions of race relations. The problem of race relations in America is one deeply ingrained in the American psyche. Racial differences are continually reinforced through stereotypes, cultural neglect, and cultural misunderstanding (Jones, 1997). and thus a necessary step towards understanding the nature of race relations should be to discard the notion that the effects of racial prejudice are one directional. The undersized literature on Blacks‘ racial attitudes promotes a one-sided view of public opinion on race relations. and potentially misdirects the approaches to resolving issues of race in America. The problem of race relations was initially thought to be a problem for Whites to resolve (Myrdal, 1944; Si gleman and Welch, 1991; Sheatsley. 1966; Shelton. 2000). In essence. Whites are expected to determine how. and what course of action should be taken. to alleviate racial prejudice. When. in the 19605 and 19705. Black Americans made the conscious effort to publicly expedite racial equality. through the Civil Rights and Black Power Movements (Hamilton and Charmichael, 1967; Paige, 1970), it became apparent that segments of Blacks actually had crystallized opinions about what they wanted. and that their resolve was so strong that some were presumably willing to die for their opinions to be heard (Young. 1996). In this sense. Blacks were fighting for democratic freedom and equality, both in the public. and in their own minds. The Civil Rights equality of the 19605 that Blacks sought was publicly centered on equal outcomes. Supreme Court decisions and race based public policy helped pave the way for Blacks to move toward more equal education, economic. social, housing. and employment opportunities. as well as a psychologically better way of life. This better way of life was idealized as the “American dream." An important step in realizing the dream is the elimination of racial discrimination. prejudice. and the physical and emotional threats that Blacks publicly faced. These contextual racial threats consisted of Jim Crow standards in the South, and segregated poverty in the North; and a reduction of these threats implied that racial equality would become the dominant view. However. these results of the movements towards a more psychologically equal racial America have been mixed. Now, more than ever, Whites negative attitudes towards Blacks are more liberalized and egalitarian than in the past (Schuman, Steeh. and Bobo. 1985; Smith and Sheatsley, 1984), and while most White Americans are committed to principles of equality. there is evidence that they are less likely to support race-based policies that attempt to produce equal economic and educational outcomes I‘J (Bobo and Klugel. 1993; Kinder and Sanders. 1996). In addition. a large number of Whites still hold negative stereotypical views of Blacks. For example, research from the early 19905, showed that 59% of Whites believe that "Blacks prefer to live off welfare”. and 54% of Whites believe that “Blacks are Lazy“ (Sigelman and Tuch, 1997). Also. while the opportunities for housing mobility have improved, Blacks are still considerably segregated from Whites. especially in population dense urban environments (Massey and Denton, 1993). Together. these factors provide mixed signals to Black Americans in terms of public commitments to racial equality. and attainment of the American dream (Bobo. 2001; Hochschild, 1995; Sigelman and Welch, 1991). On the one hand. there are increased opportunities for equality. and on the other. there are still social. economic. and most importantly psychological barriers. Blacks’ perceptions of their racial status in American society are potentiallyjust as important as the actual legal, sociological. or economic barriers they confront. Yet, there are few studies focusing on how Blacks . psychologically respond to racially targeted prejudice and discrimination (Swin and Stangor. 1998), and even less is known about how Blacks might reciprocate their prejudice towards other racial groups. especially Whites. Considering the history of race relations in America as contextual, it is reasonable to expect that Blacks hold prejudice attitudes towards Whites. but the question of "how much” has been left unanswered. For example. in their seminal work on Blacks’ racial attitudes. Sigelman and Welch (1991) address many perceptions that Blacks have about their racial environment. However, they don’t empirically focus on the relationships between perceptions of the racial environment and Blacks attitudes towards Whites. The question remains. to what extents do Blacks' perceptions of their racial context affect their negative attitudes towards Whites. The study focuses on Black Americans as a population. and the research problem is specifically directed at the prejudice attitudes of Blacks towards Whites. How prejudiced are Black Americans towards White Americans? This is a straightforward question that has yet to receive a straightforward empirical answer. Those studies that have attempted to assess Blacks‘ prejudice towards Whites. and those that have made the attempt (e.g.. Brigham, 1993; Foley. 1977; Judd et al., 1995; Livingston. 2002; Noel and Pinkney. 1964; Monteith and Spicer, 2001: Paige, 1970: Sniderman and Piazza, 2002) are limited to mostly college students and small or non- representative samples. and/or incorporate unreliable measures of Blacks’ prejudice (Biemat and Crandall, 1993). Previous studies of Blacks racial attitudes have also been hampered by small or unrepresentative samples of Blacks. which have can hamper attempts to conduct rigorous subgroup analysis and make valid generalizations (Schuman et al., 1997; Si gelman and Welch, 1991). Therefore, I focus on Blacks throughout the United States rather than just one particular locality (e.g., Sniderman and Piazza. 2002; Wojniusz, 1979). This is an immense task, because to date, there has not been a single study of Blacks‘ prejudice towards Whites that contains a nationally representative sample of Blacks. This study will facilitate an initial understanding of the current nature of Blacks' prejudice towards Whites, and help direct future research on the study of Blacks' racial attitudes. This research will also shed light on the extent to which Blacks prejudice towards Whites is reactionary. This line of reasoning implies that the content of prejudice for targeted minority groups is different than the prejudice of prejudice providing majority groups. Research has supported this notion. showing that when studied in-depth. Blacks‘ prejudice attitudes contain high levels of resentment and frustration towards Whites. coupled with fears regarding being targeted for racially unfair treatment (Brigham. I993; Monteith and Spicer, 2001 ). Also. while Whites‘ prejudice has been empirically correlated with anti-egalitarian values systems (McConahay, 1986; Monteith and Spicer. 2001)), stereotypes (Hurwitz et al.. 1997; Snidcmtan. 1993). and interracial contact (Sigelman and Welch, 1993). the salient content of Blacks‘ negative attitudes appear to be mistrust (Terrell and Terrell. 1981; Thompson et al.. 1990), resentment over unfair treatment (Sigelman and Welch. 1991), reactions to perceived prejudice and discrimination (Monteith and Spicer. 2001; Thompson et al., 1990), alienation (Bobo and Hutchings, 1996), and strong in-group bias (Judd et al.. 1995; Livingston. 2002). Limited Research on Blacks’ Prejudice Attitudes Unfortunately, a more detailed understanding of Blacks' prejudice requires one to examine research literature across several social science disciplines. such as Sociology. Psychology. Political Science. and Anthropology. In general. across the majority of disciplines racial prejudice has been studied from the perspective of the dominant culture --usually Whites--rather than the targets" perspective-usually Blacks. This gives little attention given to the levels or construct of Black prejudice, or even how Blacks“ prejudice correlates with perceptions of race relations (Livingston, 2002; Monteith and Spicer, 2000; Shelton, 2000). The prejudices of both Blacks and Whites are relevant in order to better understand race relations. and in discussions of potential solutions toward perceived Black-White issues (Kinder and Sanders. 1996; Schuman et al.. 1997; Sigelman and Welch. 1991). There appear to be three main reasons for the paucity of research knowledge on Blacks’ prejudice towards Whites. First. there is uncertainty about how to measure Blacks’ prejudice towards Whites. Most ofthe scales that measure racial prejudice have been developed and initially tested on White college students or White populations (Biemat and Crandall, 1999). and might not be appropriate for an in-depth analysis of Blacks" prejudice. Scales such as the Attitudes Towards Whites (ATW) scale (Brigham, 1993) were developed using references suited for Whites, but switch some of the wording in order to accommodate data collection for Black respondents. This creates a serious potential for measurement error in the study of Blacks‘ prejudice. As previously mentioned. studies suggest that Black and White dimensions (or content) of prejudice are potentially different (Judd et al., 1995; Monteith and Spicer, 2000); yet. whenever there is an attempt to analyze Black prejudice. it is based on a relative measurement to Whites on the same set of items (Foley. 1977; Judd et al.. 2000; Kinder and Sanders, 1996; Krueger, 1997; Monteith and Spicer, 2000). This is primarily because the main focus of study has been White‘s racial attitudes, but the convenience of Black respondents generally necessitates data collection on all variables. When Blacks‘ levels of prejudice are compared to Whites’ levels of prejudice, the assumption is that prejudice can be measured the same way for both groups. This is a potential flaw and a leading indicator of Type I error, potentially representing that differences exist when they do not. These measurement and design issues might be one reason why past studies comparing the levels of prejudice across Blacks and Whites have been inconclusive (Foley. 1977: Monteith and Spicer. 2000: Preston and Robinson. 1974). The next two reasons for the lack of studies on Blacks' prejudice lay in its controversy as a research topic. Since the temis "racism" and "prejudice" are often confused, researchers have cautiously approached studying Black prejudice. usually for fear of misinterpretation. If Blacks are reported to be equally or more prejudiced than Whites, there is a fear that Whites might be absolved of their role in any racial problem. or that such research might be equivalent to blaming the victim (Shelton. 2000). In addition, the notion that Blacks can be racist would surely produce more harm than good by seemingly justifying the negative treatment/attitudes Blacks have received (Jones. 1997). Secondly, any research consisting of valid and reliable measures of prejudice might find it too costly (e.g.. the need for incentives and suitable response rates. and potential item non-response) to collect a large enough Black sample. or even carry out multiple studies over time (Sigelman and Welch. 1991). The final issue centers on a social context that potentially discourages open dialogue about sensitive topics. It is plausible that some respondents will be apprehensive about revealing their actual prejudiced attitude. Although studies have shown that over the past 30 to 40 years. White attitudes have become favorable towards Blacks and policies benefiting Blacks (F irebaugh and Davis. 1984; Gallup Organization. 2001; Schuman, Steeh, and Bobo, 1985; Smith and Sheatsley, 1984), there is still evidence that some of the changes in these attitudes are likely influenced by the race of interviewer (ROI) (Davis. 1997a). As a result the ROI stands as an important source of nonrandom measurement error in the modeling of racial attitudes. Given evidence that both Blacks and Whites exhibit ROI effects (Davis. l997b). there is good reason to believe that any attempt to measure racial attitudes through an interview might be tainted by this effect. Thus, the study of racial attitudes can become complex measurement issue. ROI issues are less a data collection problem than a research approach assessing the importance of the respondent-interviewer interaction. This potential interaction effect is in essence a measure of social context. and can be used as an indication that racial differences underlie any topic that the interviewer source effects (Schuman and Converse. 1971). Nonetheless, the ROI issue is an additional consideration in the development of a consistent body of findings on Blacks racial attitudes. These issues together leave a vacuum in the understanding and study of Blacks' prejudice towards Whites. Thus. there is a definite need for larger empirical studies that incorporate the study of Black racial attitudes towards Whites. Blacks Racially Prejudiced Attitudes Towards Whites The goal of this research is to better understand the nature of Blacks' prejudice towards Whites. I propose that Blacks‘ prejudice towards Whites is a function of Blacks’ perceptions of their negative racial context. In this study. a racially negative context can be thought of as having two related aspects: a perceived negative racial environment, and perceived racial prejudice threat. A negative racial environment implies that Blacks realize and understand the historical impacts of racial prejudice on their group; that their race has, in a large number of cases. been treated unfairly in an informal hierarchy of racial groups. A negative context also implies that any actions or attitudes that might lead to the continued subordination of Blacks can be considered a threat to their race. The threat results from a racially competitive environment where traditionally one racial group has benefited from the subjugation of the other. This theory has traditionally been applied to findings of Whites racial attitudes. but has been recently extended to Blacks under the "sense of group position" model (Bobo. I999; Bobo and Hutchings. 1996). Blacks’ Awareness of Their “Group Position” One reason why Blacks might hold negative attitudes towards Whites, stems from the fact that Blacks are well aware of their negative racial environment. Blacks believe that Whites can be unfairly prejudiced (Grier and C obbs. 1968; Monteith and Spicer. 2000), hold negative stereotypes of Blacks (Sigelman and Tuch, 1997), and racially discriminate (Gallup Organization. 2001; Sigelman and Welch. 1991). Recent research finds nearly half (47%) of Black Americans feel that they received some recent unfair treatment because they are Black (Gallup Organization, 2001). Unfair treatment also extends beyond the individual encounters. In terms of experiences with police and law enforcement, Blacks are much more likely than Whites to perceive that they are being targeted or discriminated against by law enforcement (Gallup Organization, 2001; Tuch and Weitzer, 1997), and 83% of Blacks believe the practice of "racial profiling" is widespread (Gallup Organization, 2001). Even if a Black person has not personally been a target of negativity due to their race, Black Americans still hold a general perception that prejudice, racism, and discrimination still exist within society (F eagin, I991; Monteith and Spicer, 2001; Sigelman and Tuch, 1997; Ruggiero and Taylor, 1997; Sigelman and Welch, 1991). These perceptions, regardless of how real or imagined they are. can have serious psychological consequences, such as lowered self-esteem, stress. frustration, alienation. and fears of stereotype validation (Crocker, Major. and Steele, 1998; Jones, 1997: Steele and Aronson, 1997). Some of the negative perceptions Blacks have about society are due to the ambiguity that occurs through interpersonal feedback. Commonly. when Blacks encounter negative feedback. one of the more available heuristics is "race.” That is. there are relatively high levels of attributional ambiguity. implying that some negative encounter might be due to race (Crocker et al., 1991; Feagin. 1991). Crediting negative treatment to racial discrimination is rational since Blacks are acutely aware their historical mistreatment, almost to the point where the attention has been called a healthy "cultural paranoia" (Grier and C obbs. 1968). In short, there is little doubt that Blacks are aware that race plays an important role in how they are treated in some situations. and given this possibility, they must be on guard for these potential encounters. Blacks’ Responses to Prejudice Blacks are not passive targets of prejudice, they respond to negative treatment both psychologically and behaviorally. The large majority of these responses are viewed as negative because they are brought about by negativity from the out-group, and redirect themselves as group and individual defense mechanisms (C rocker et al., 1998; Swim and Stangor, 1998). Up until the 19905, there was no sound body of research regarding Blacks' perspectives on prejudice (Swim and Stangor, 1998); however. of the research that was conducted, the large majority focused on Blacks’ internalization of being a stigmatized group (Clark and Clark, 1939: Clark and Clark, 1947; Goffman, 1963; Horowitz, 1939: Lewin, 1948). This research centered on the ideas of low self-esteem and self-hate due to 10 identification with a subj ugated racial category. Allport (1954) identified this internal cognitive response, “intropunitive”: the internalization of negativity towards self or the in-group. Inherent in this type of response is that notion that self-evaluations are driven by how individuals perceive others are viewing them. More recent approaches to studying the internal responses to prejudice and racism include social-dominance theory (Sidanius and Pratto. 1999) and system justification theory (Jost and Banaji. 1994). Generally, both models state that in order for hierarchies to survive, subordinate group members must internalize some level of the dominant groups negative beliefs and attitudes about themselves. Another theory. stereotype threat (Crocker, Major, and Steele. 1998; Steele. 1988; Steele. 1997). which is the discomfort targets feel when they are at risk of fulfilling a negative stereotype about their group. has been proposed as an internal response to negative racial attitudes. As an internal response. stereotype threat works to hinder behavioral performance in areas such as academic testing and group participation (Steele, 1997), which can have dramatic long- term effects beyond the individual. In general, there is clear empirical evidence that the internalization of negative attitudes and beliefs is one fomi of response to prejudice. Blacks, and other racial minority groups can also respond to prejudice by redirecting the real or perceived negativity back towards Whites. or any other out-group for that matter. Allport (1954) labeled this external response, “exptropunative”: where the victim of prejudice blames his/her condition on the provider of the prejudice. and reciprocates the prejudice towards the original provider(s). Studies have shown that reactions to perceived negativity form the basis of much of Blacks negative attitudes towards Whites (Livingston. 1991; Monteith and Spicer. 2001: Mphuthing and Duckitt, ll 1998; Thompson et al.. 1995). There are also studies that emphasize the impact that Blacks’ negative experiences with racism and discrimination. and how perceived negative racial stereotypes held by Whites impact Blacks’ racial attitudes (Feagin. 1991; Sigelman and Tuch, 1997; Sigelman and Welch. 1991). I propose that Blacks extropunative responses form the basis of their stated prejudice towards Whites. The systemic racial subjugation of Blacks is primarily a nightmare of the past. however. negative racial experiences do provide a basis for prejudice attitudes. Prior to the 19705. it would have been considered radical to show public resentment towards Whites if they were perceived to be prejudice; however, today it is considered dutiful and necessary. However. knowing that one is a target of prejudice is a threatening notion, characterized by a fear that history may repeat itself. Thus, the extent to which Black resentment towards Whites presents itself is likely a function of the perceived racial environment that exists in America. Prejudice as a Response to Perceptions of a Negative Racial Environment A perceived negative racial environment can be characterized as the perception that a group or individual is treated unfairly on the basis of their race. When members of a disadvantaged group perceive themselves as being treated unfairly in a variety of situations by a dominant group. they are more likely to feel both more hostility toward the out-group. and more identification with their in-group (Branscombe and Ellemers. 1998). This is a key component to understanding the potential reciprocal nature of Blacks’ prejudice towards Whites. The more Blacks perceive their environment to support and maintain unequal standards and treatment. the more Blacks should resent the providers and/or perceived supporters of this treatment. The unfair treatment can be either real or perceived, and the response can come in many forms: cognitive, emotional. or behavioral (Branscombe and Ellemers. 1998). I propose that Black prejudice is a calculated and strategic response to perceived racial threat. Blacks systematically evaluate their landscape. and attempt to evade situations that are physically and psychologically threatening (Davis. 1994; Petti grew. 1964; Majors. 1993). Blacks also seek to avoid individuals and groups who create or promote situations of threat and unfaimess. This avoidance behavior can manifest itself as in-group preference or etlmocentrism. but is generally viewed as a strategic response to protect group social identity (Swim and Stangor, 1998). There is no doubt that Blacks can use prejudice as a self-protective strategy. Research has shown that Blacks refuse to tolerate groups who pose a constant racial threat, such as the Ku Klux Klan (KKK) (Davis, 1994). And if for example. Blacks perceived that all KKK members were Whites. there is little to suggest that an extrapunative response would not extend to Whites as a whole, especially given the cognitive burden of trying to one-by-one evaluate which Whites are discriminating or prejudice versus those that are not. When Blacks encounter negative situations that they perceive as unfair, there is always the possibility that race has played a role in the encounter. For some higher income and educated blacks, this is particularly frustrating given par levels of economic and educational success. Studies and reports have documented Blacks discontentment with their pursuit of the American dream, along with feelings of alienation and mistrust (Bobo. 2001; Cose. 1993; Hochschild. I995; Sigelman and Welch, 1991). These cognitions and attitudes do not exist in a vacuum; they are a function of perceptions of. and experiences with. discrimination and racial prejudice (Sigleman and Welch, 1991). Thus. there is a strong chance that Blacks respond to the contextual constraints and threats of Whites prejudice. with reciprocal levels of prejudice. Perceived Threat and Intolerance C onceptualizing Blacks‘ prejudice towards Whites as a strategic racial reaction to threat is consistent with studies linking intolerance and Black racial attitudes. Much of this literature hypothesizes that Black racial attitudes and behaviors towards Whites are possibly rooted in perceptions of threat, constraining beliefs. or conflict, and are a reaction to White racism or contextual intolerance (Davis, l997a: Davis. 1994: Gibson. 1995; Gibson, 1992). Gibson (1992, 1995) has studied the impact ofcommunity level racism/intolerance on Blacks’ perceptions of their freedoms and ability to express themselves politically. He found the more that anti-Black sentiment exists in a community, the more Blacks feel politically constrained. In addition, Gibson found that Blacks who did not feel free to express themselves were less tolerant of others. and tended associate. and even have spouses who are less tolerant persons. These context effects imply that Blacks are aware of their racially charged environment. and that this environment actually influences individual level beliefs and attitudes. Davis (1994) has also found that Blacks can be highly intolerant of the groups they least like, such as the KKK. Traditionally, the KKK has been a group that provides acumen levels of prejudice towards Blacks and other groups (Davis, 1994). Because of the historic threatening nature of racial prejudice. Blacks tend not to believe that the KKK‘s views operate in a vacuum. When asked in the 19805. how many Americans shared the views of the KKK. Blacks were most likely to report “twenty-five percent or 14 more" (Sigelman and Welch. 1991). Yet. not all Blacks have been exposed to the KKK. or even seen a member ofthe KKK. and they could still consider them threatening. In fact. without even hearing the views of the KKK. most Black Americans would be highly offended by their inclusion at any public or private event. and respond with some level of negativity towards the group. This would be an example or reciprocal intolerance towards a social group. In this instance. it is the perception of physical or psychological threat of the out-group that drives the intolerant attitudes towards the dominant group. While the large majority of Whites are not members of the KKK. this finding might imply that Blacks who view Whites to be threatening. at any level, might be as likely to make a conscious and focused decision to dislike and avoid them as a group. Blacks may feel that it is logical to hold prejudice views against any group that has traditionally held prejudiced views against them. The result could be high levels of in— group favorability coupled with high levels ofout-group dislike. This implies that Black prejudice towards Whites is possibly the result of collective fears and expectations associated with being in a racially threatening environment. This is a highly testable proposition, and one in which I will investigate in this study. The research parallels the political intolerance literature. Instead of contextual intolerance, I am interested in observing the impact that a perceived negative context has on Blacks’ prejudice towards Whites. Both Gibson (1995) and Davis (1994) imply that Blacks perceptions of freedom and their levels of intolerance are reactions to the external constraints and the past behaviors of Whites in America. This study continues along the same lines of reasoning by hypothesizing that these same negative external constraints, and perceptions of threat. work to produce negative prejudice attitudes towards Whites. 15 Focus and Goals of this Research The purpose of this study is to show the empirical relationships that exist at the national level. between Blacks perceptions of their racial context and their negative attitudes towards Whites. Using Gallup Poll Race Relations survey data. collected by the Gallup Organization. I analyze potential correlates of Blacks' prejudice towards Whites. The general hypothesis is that Blacks' prejudice towards Whites is a function of the extent to which Blacks view their environment as negative and racially threatening. Perceptions of a negative context can take the form of perceived unfair group treatment. limits to opportunities for Blacks. and individual experiences with discrimination due to race. Perceptions of how prejudice Whites are in the community provides an indication of how racially threatening Blacks perceive their environment. The Gallup data is well suited for the study of Black prejudice for many reasons. First. each of the data sets - collected independently in 1997. 1998. and 1999 - contain over 950 Black respondents. and two of the three studies contain more than 1,000 Black respondents. All things being equal. this allows for rigorous statistical testing. and more confident results. Second. the data were collected over three separate years using similar sample design and question wording. Three years of data allow for the testing of hypotheses over multiple time periods: hence a more stringent test of the propositions. Third, the measurement of prejudice in the Gallup studies is the same over the three-year period. Specifically. respondents were asked to self-report their levels of prejudice, on an 11-point scale. Finally. all three years of data contain a variable that identifies the race of interviewer. This allows us to partially test the effects of social desirability due to the interviewer‘s race. 16 My primary research question is the following: to what extents do Blacks‘ perceptions of their racial environment affect their individual levels of self-reported prejudice? Taken together. the data. target populations. and research topic make for a unique study of public opinions and attitudes. The ultimate goal of this research is to increase the understanding of Blacks‘ prejudice and its determinants. and provide directions for the future study of Black prejudice. The Concept of Prejudice Before moving forward. it is important to define prejudice, and compare it with other terms commonly associated with race relations. First. prejudice. racism. preference or ethnocentrism, and discrimination are conceptually. and practically different. Broadly speaking, prejudice is an attitude. racism is a belief system of group control based on power and racial superiority, preference is a choice. and discrimination is an actual action or behavior. Of these terms. I am only concerned with prejudice. the attitudinal component of racial negativity. Prejudice has traditionally been examined through two views: a sociological view, and a psychological view (Jones. 1997). The psychological approach focuses on feelings and attitudes. and has traditionally focused on personality or cognition (e.g.. Devine and Elliot. 1995). The sociological approach emphasizes how prejudice groups develop, maintain, and utilize prejudice toward specific target groups. This approach emphasizes that racial and ethnic prejudices develop from a sense of group position (e.g., Blumer. 1958). and are often maintained through stereotypes about the target group (Krueger. 1996; Judd and Park, 1993). Social psychology has helped merge the two approaches. 17 viewing prejudice as a positive or negative attitude. judgment. or feeling that is generalized from beliefs about a target group (Jones. I997). Jones (1997) defines prejudice as a “positive or negative attitude. judgment. or behavior generalized to a particular person that is based on attitudes or beliefs held about the group to which the person belongs (p. 142)." Although. the symmetrical nature of prejudice — positive or negative attitude — is generally accepted. racial prejudice is contextually centered on negative differences based on race. and thus. connotes negativity. In this study. I consider racial prejudice to be. simply. a negative attitude towards a racial group. Yet. a prejudice attitude doesn‘t imply that one is racist. Prejudice and racism are not the same. Racism has many practical definitions (e.g.. aversive. structural. cultural. and ideological). and operates at different levels: individual. institutional. and social structural. Jones (1997). notes that racism has several key elements that make it different from prejudice. First. racism implies a belief in superiority based on ostensible biological and cultural differences. Second, a racist entity has strong in-group preferences and rejects customs and values that differ from their own. Third, racism offers advantages to those with authority. Fourth, racism implies an attempt to justify and validate racial differences with policies and practices. Thus, although racism and prejudice are related, the former involves a much more complex set of ideas and practices (Bobo. 2001; Jones. 1997). Straightforwardly. racism is a belief that . differences in behavior are due to racial inferiority. and a racist practice or policy. is one that is designed to maintain racial superiority for a dominant group. This clarification is important. because a number of research studies of racial attitudes often use the words prejudice and racism interchangeably. For example. 18 concepts such as symbolic racism (Kinder and Sander. 1996). modern racism (McConahay, 1986). and subtle racism (’Pettigrew and Meertens. 1995). have often been called prejudice (e.g.. subtle prejudice or modem prejudice). Additionally. Snidemian and Piazza (2002) in their most recent study of Blacks ”Pride and Prejudice" consistently make the mistake of considering intolerance -- which is essentially defined as a “willingness to put up with“ (Sullivan. Piereson. and Markus, 1982) -- to be the same as prejudice. Therefore. it is important to state the differences at the beginning of this research. The key difference between racism and prejudice is power. Racism implies superiority and a desire to promote one racial group over another with the intention of maintaining power. Since there are relatively few domains where Blacks as a group hold power, many believe that it is not possible for Blacks to be racist. but there is no doubt that Blacks can be prejudice. and express their prejudice through racial discrimination. It is not the intent of this study to show that Blacks and Whites are equally likely to mistreat (i.e.. discriminate against) each other. Nor. is the goal to show that Blacks and Whites are equally likely to undermine the others‘ attempts at a healthy socio- economic lifestyle. The topic of interracial racism is well beyond the scope of this research. The focus of this study is the extent to which Blacks’ reveal their prejudice towards Whites in the face of perceived negative contexts. l9 Chapter 2 THEORY It is negligent to consider the importance ofgroup based racial prejudice, without considering the role of threat. The primary function of racial prejudice is to humble and diminish the relevance of outside groups. In the literature on Whites“ prejudice. racial threat is a continuing theme focusing on economic. social. and political outcomes (Fosset and Kiecolt. 1989; Hertz and Giles. 1996; Key. 1949; Oliver and Mendelberg, 2000; Taylor, 1998). Yet for Blacks. racial threat is almost cursory in its relevance to negative racial attitudes. with researchers directing more attention to personality factors (Brigham, 1993; Foley. 1977; Judd et al.. 1995). mistrust (Taylor. 2000; Thompson et al.. 1990). negative experiences (Grier and C obbs. 1968). social and interracial contact (Segilman and Welch. 1993). racial identity (Snidemian and Piazza. 2002), ethnocentrism (Evans and Giles. 1986). and resentment (C ose, 1993: Monteith and Spicer, 2000). To the extent that Blacks selectively hold prejudice attitudes. their willingness to apply negativity towards some groups and not others likely depends on the extent to which Blacks perceive a particular group as threatening. One of the more consistent findings in the few studies of Black racial attitudes towards Whites is the idea that negative Blacks are responding to some real or perceived threat. One line of research points to economic and social threat across racial lines. Williams (1973) believed that an important source of racial hostility in society is racial group competition for scarce values such as wealth. jobs. power. and prestige. According to Williams, Whites have traditionally had an advantage position in the attainment of these values. and when Blacks perceive that they do not have equal access. they tend to believe it's due to Whites‘ discriminatory and prejudiced practices. Thus. the level of anti-white feeling depends on black perceptions of racial group competition, or real or perceived efforts by Whites to keep Blacks in subordinate positions (Williams. 1973). Similarly, in her study of racial hostility among Blacks in Chicago, Wojniusz (1979) found that Blacks who exhibited a relatively high amount of hostility towards Whites tended to perceive their ability to achieve equal economic and social positions were being threatened by factors such as discrimination and hypocrisy. Bobo and Hutchings (19%) extend a sociological model of prejudice proposed by Blumer (1958) that also emphasizes the role of contextually conditioned threat. The Blumer model, almost entirely applied to White populations. hypothesizes that prejudice is a function of a sense of group position. arising from a competitive context. However. according to Bobo and Hutchings. groups who perceive more alienation and unfair treatment, are also more likely to regard members of other groups as potential threats. The importance this research is the extension of Blumer's model to other minority groups. Specifically. Blacks had the highest reported levels of alienation. and had the highest reported level of perceived economic competition (i.e., in zero-sum terms). The implication from the Bobo and Hutchings study was that since Blacks are very likely to perceive other racial group as competitive threats, they are also likely to respond to these threats with prejudice. Thus, there is a general pattern in the literature on blacks’ racial attitudes that highlights the perspective that prejudice is a response to threat. One of the more recent studies of Blacks’ prej udice. conducted by Sniderman and Piazza (2003), provides an example of how studies of Blacks’ prejudice have missed their mark by omitting threat from their analysis. The authors start with a basic question. asking. "are black pride and black intolerance opposite sides of the same coin (p. 2)." Then they proceed to ignore the key factors in the general literature on intolerance (e.g.. Sullivan. Piereson. and Marcus. 1992). the specific literature on Black intolerance (Davis. 1994; Gibson. 1992; Gibson. 1995). and other research on Blacks' prejudice (e.g.. Livingston. 2001: Monteith and Spicer. 2000: Paige. 1970) by excluding the role of perceived racial threat. Sullivan. Piereson. and Marcus (1992) provide seminal evidence regarding the relationship between perceived threat and political intolerance that has clear implications for studies of prejudice. The authors importantly note at the beginning of their text, that prejudice and intolerance are not the same; however. they suggest that the presence of a negative evaluation. or prejudice. is a key factor in determining intolerance (p. 5). This implies that factors that detemiine intolerance should also correlate with prejudice. Since the strongest predictor of intolerance in the Sullivan. Piereson. and Marcus (1992) study was perceived threat. it follows that those persons who are intolerant of groups they dislike, are more likely to perceive a threat from those disliked groups. Equally important, for the sample studied, the effects of perceive threat held constant across demographic and psychological factors. The findings in this study point to both a need to consider perceived threat in studies of prejudice. and a need to clarify conceptual differences between intolerance and prejudice. Most of the guiding evidence about perceived threat. race. and constraining political attitudes can be found in the intolerance literature. Davis (1995) found that Whites are most intolerant of groups who threaten their belief and value systems. These are least-liked groups such as communists. the Klu Klux Klan (KKK) and other less religious and liberal groups (e.g.. the Black Panthers). With regard to Blacks and intolerance, Davis provides even stronger evidence regarding the importance of threat. He finds that Blacks' intolerance is based on perceived threat from their least liked group. which is overwhelmingly the KKK. Accordingly. Blacks are significantly less tolerance of groups who pose a direct threat to their physical and or social existence. In addition. Gibson (1995) finds that Blacks perceive less freedom in communities and environments that are more racist and more tolerant of racists. The implication from Gibson‘s work is that racially intolerant communities provide a threatening environment that constrain perceived freedoms and potentially lead to fomts of in-group bias. This suggestion is validated in another study by Gibson (1992) which showed that Blacks who feel more constrained are more likely to be intolerant of others, have less heterogeneous peer groups, have less tolerant spouses. and less tolerant communities. The points of emphasis in the literature on Blacks are clear. racial threat via a negative racial context is fundamental to understanding Blacks intolerance. Therefore, the role of perceived racial threat should be considered when assessing both intolerance and prejudice. Additional research points to the importance of Blacks’ reactions to prejudice and racism. There are four altemative, though non-mutually exclusive. theories for why Blacks might hold prejudice views towards Whites: ethnocentrism. the lack of interracial contact, cultural mistrust. and racism reaction. While the large majority of studies on Blacks’ prejudice towards Whites have occurred in the traditional experimental psychology field (Duckitt and Mphuthing, 1998; Judd et al., 1995; Livingston, 2001; Monteith and Spicer. 2000), the common denominator in these studies show that Blacks' prejudice attitudes towards Whites tend to be reactionary to Whites’ attitudes and behaviors towards Blacks. and to other perceptions of social systemic racial unfairness. This is the reverse of the claim by Sniderman and Piazza (2002). as they state that Black‘s perceptions of others‘ prejudices towards them as a racial group, are a function of their ovm prejudice towards others (p. 163). Restated. this means that Blacks perceptions of how they are treated by outside groups are due to their prejudices towards the outside group. Statistically this may be true. but conceptually this ignores quite a bit of research on Blacks’ reciprocal attitudes based on threat. Competing Theories of Blacks’ Prejudice Before going into a more in-depth discussion ofcontextual racial threat, and prejudice. it's important to present the alternative explanations for Blacks’ prejudice that exists in across academic fields such as educational and social psychological, sociology. and political science. Ethnocentrism The ethnocentrism argument states that Black prejudice stems from ethnocentrism: the tendency to view and evaluate one’s own group more favorably than the out-group and to show preference for benefits of belonging to the group. Due to historical factors like racial segregation. Blacks have often been forced to live and socialize within their racial group. regardless of class or social status. The shared perceptions and experiences of large numbers of Blacks establishes a bond leading to cultural favoritism (Judd et al.. 1995). The result is a high valuation of ethnic heritage, and a belief that racial ethnicity matters. The interesting feature of this model is the possibility that the Black prejudice might result from Whites denials of cultural importance. That is, the more Whites (or Blacks) promote the idea ofa melting pot, where there is one culture. the more likely Blacks may view Whites with resentment for not recognizing the importance of racial histories and heritage. What follows is the development of strong preferences for individuals. groups. and entities that recognize the importance of race. Some researchers have presented this dynamic in the context of stereotypes. proposing that even Whites' refusal to hold certain positive stereotypes (e.g., “Black women are stronger women" or "Blacks are better athletes") of Blacks may constitute a “new form of prejudice (Judd et al.. 1995)." The implication is that Blacks might actually value these stereotypes of part of their cultural heritage. and a denial of such images might constitute a perceived dislike of the culture. As a result, Blacks may increase their favorability towards their own group, while decreasing favorability towards Whites. although one does not necessarily lead to the other (Brewer. 1999). Sniderman and Piazza (2002) study the relationships between Blacks and Jews in this very context. They attempt to observe the relationship between the many fomis of psychological black racial identity and anti-Semitism. They hypothesize that as Blacks increase their beliefs in afrocentric ideas and conspiratorial thinking, they are more likely to endorse “negative" stereotypes of Jewish persons. Restated from an ethnocentrism perspective, the more Blacks favor ideas that are pro-Black (e.g.. self-pride. afro- centrism, shared beliefs). the more they should prejudice towards other groups. Essentially, Snidemian and Piazza (2002) find that this is not the case with Blacks. They find no consistent relationships between anti-Semitism and “black pride (p. 162).” fit Interracial Contact The interracial contact hypothesis maintains that Black prejudice stems from the reality that Blacks have less contact with Whites due to racism and segregation (Allport, 1954; Sigelman and Welch. 1993). Similar to the ethnocentrism approach. this notion suggests that Blacks who primarily live around other Blacks. and have low contact with Whites. are more prejudiced towards Whites. or stated differently, biased towards their own racial group. The interracial contact literature dates back to the 19505 and 19605. when it was thought that the desegregation of schools might bring about better race relations (Stevens. 1978). The belief was that Civil Rights legislation, affirmative action, and school and workplace desegregation would result in increased contact, facilitating more positive or at least pragmatic interactions between Black and Whites (Thomas. 1985). In addition, after Allport (1954) suggested that that increased contact might actually lead to a reduction in prejudice attitudes for Whites many researchers decided to test this notion. The consistent findings from the interracial contact studies of prejudice show that Whites are less prejudice when they encountered greater equal-status contact with Blacks; however. increased social contact for Blacks has not shown an appreciable effect on their racial attitudes or levels of prejudice; as Blacks‘ contact increased. their measured prejudice levels remained statistically unchanged (Brown and Albee, 1966: Ford, 1973; Preston and Robinson. 1974; Robinson and Preston. 1976; Sigelman and Welch, 1993; Tsukashima and Montero, 1976). One potential explanation for these findings is that Blacks‘ view racism and prejudice as factors that operate at a macro-social level. For example. evidence shows li that Blacks believe discrimination still exists (Bobo, 2001: Sigelman and Welch. 1991). that it is institutional and systemic rather than episodic (Klugel and Smith, 1986: Bobo. 2001; Bobo et al., 1994; Farley et al.. 1993). and that community level intolerance threatens Blacks' individual feelings of freedom (Gibson. 1995). In addition, Blacks tend to respond to this race based threat in systematic ways that signal to the threat providing group that Blacks are group protective (Davis 1995: Gibson. 1992). Taken together these findings imply that Blacks have less concern over what happens in one on one individual level relations and contact. because the larger political, social, and economic systems are more important to their goals of fair treatment and equality. Social contact is less relevant to Blacks’ prejudice because the contact doesn‘t change the perceived formal or informal prejudiced system. The result is that even if blacks work with, live with, or even marry whites, they might still have high levels of prejudice. because there is a larger understanding that the system is more racist and prejudiced than individual Whites. In general, the social contact hypothesis has been unsuccessful as an explanation for Blacks negative attitudes towards Whites. Cultural Mistrust The cultural mistrust argument states that Blacks‘ prejudice stems from Blacks (and other racial minority groups) being exposed to prejudicial and discriminatory practices by dominant White society (Grier and Cobbs. 1968; Terrell and Terrell. 1981; Thompson et al., 1990). Accordingly, Blacks' mistrust of Whites is a strategy of dealing with the everyday occurrences of prejudice. It serves as a strategic defense mechanism that helps Blacks physically and emotionally avoid negative and derogatory situations including educational institutions (Brazziel. 1974; Russell. 1971). political and legal systems (Abramson. 1983; Kitano. 1974), work and business settings (Baughman. 1970: Rutledge and Gass. 1967). and interpersonal or social contexts (Kitano. 1974). Under the cultural mistrust model. prejudice is the product of mistrust towards Whites and predominantly White power structures. The cultural mistrust model fits into the mold of more traditional psychological theories of prejudice. In general. the theory states that attitudes of mistrust are socialized early and continuously. both in and outside the home. Researchers have suggested that the first signs of mistrust develop in the home. where parents, siblings. other family members. and neighbors provide some the initial insights into the parameters and cautions of being Black in a predominantly White society (Erickson, 1983; Grier and Cobbs. 1968; Taylor, 2000). These attitudes are then reinforced in school. and the external community (Russell. 1971; Thompson et al.. 1990). With regard to Blacks prejudice towards Whites. the development of cultural mistrust serves to maintain a target's group based identification. Group identity is enhanced when group related experiences become personalized, and self-protective. Mistrust can be self-protective because it prepares a vulnerable person to take self- protective reactions or to avoid situations in which vulnerability is increased. The result is an avoidance behavior and lifestyle that leads to in-group preference. and out-group bias. Racism Reactions to Racial Experiences Similar to cultural mistrust, the “racism reaction" perspective claims that Black prejudice stems from actual experiences as the target of racial prejudice. This differs from cultural mistrust in that mistrust is primarily a function of socialization. and there is no indication that anything negative has actually occurred with the individual. there is simply a potential. There are studies that emphasize Blacks’ perceived negative experiences with racism and discrimination (Feagin, 1991; Sigelman and Welch. 1991 ). the perception of what Whites think about them (Si gelman and Tuch. 1997). and the amount of cognitive effort they expend deciding how. and whether or not. to react to White racial prejudice (Crocker et al., 1991; Feagin. 1991). These studies combined with research on “attributional ambiguity" -- how social treatment situations are ambiguous, and the difficulty ruling out race as a factor (C rocker et al., 1991; Swim and Stangory. 1998) -- have led some to conclude that Blacks possess a “healthy cultural paranoia" about race relations (Grier and C obbs. 1968). Blacks‘ awareness of their negative racial history makes them highly sensitive about the potential for racism in conflict situations. There is supporting evidence for the relationship between “racism reaction" and Blacks negative attitudes towards Whites. In a study of Black college students, researchers employed combined methodology of spontaneous open-ended attitude assessment and survey instrument. to understand both the content and correlates of Blacks and Whites attitudes towards each other (Monteith and Spicer, 2000). The results show dramatic differences in the content of prejudice across race. Whites participants claimed that Blacks do not work hard enough. they get more than their fair share, and they cry discrimination when it really does not exists (Monteith and Spicer, 2000). These attitudes where highly correlated with both Modern Racism Scale (McConahay, 1986) scores (positive correlation) and egalitarianism (negative correlation). These findings implied that the negative content of Whites’ prejudice towards Blacks. where in part due 29 to old—fashioned racism, and the perception that work ethic norms are important. Black participants negative attitudes. in part. reflected a concern for these perceptions held by Whites. Black participants' essays suggested that prejudice and discrimination are perceived quite frequently. and provide a good portion of the expressed rationale for disliking Whites. In fact. the content raters in the study found that approximately 50% of the Black participants used at least one of the negative themes, expressing attitudes that were related to participants‘ reactions to White racism (Monteith and Spicer, 2000). This implied that the salient content of Blacks’ prejudice towards Whites where, for the large part. due to their perceptions Whites‘ negative attitudes and actions towards Blacks. Placing Blacks‘ prejudice in the context of reactions to racism, discrimination. and prejudice is efficient especially when Blacks' racial history is considered. There is known potential for individuals and groups to reciprocate the treatment and attitudes they receive. Research in the area of interpersonal relations shows that individuals and groups tend to direct negative evaluations towards those they perceive as evaluating them more negatively (Curry and Emerson. 1970; Curtis and Miller. 1986). This makes the racism reaction hypotheses credible. Allport (1954) proposed that perceived negativity from out-groups might impact the target group‘s affective and behavioral orientations. 1n the context of Blacks’ prejudice towards Whites, Allport‘s research implies was that there is a potential for the target group to respond with hostility toward the provider of prejudice. He characterized this hostile response toward the out-group as "extronpunative‘fi which indicates that members redirect the negativity toward the out-group (Allport, 1954. p. 160). Researchers have also found that extropunative responses tend to lead to both increased self-esteem and in-group bias (Bigham, 1993; Judd et al.. 1995; Livingston. 2001). Thus. the racism reaction literature implies that Blacks negative attitudes towards Whites are a negative reaction to a perceived negative reaction from Whites. Like cultural mistrust. this reaction serves to protect the individual from potentially threatening situations. by protecting in-group identity and group-esteem. Sense of Group Position and Racial Threat The group position model flows from the sociological tradition of studying groups, and their real and perceived conflicting interests in society. This paradigm has many varied fomis and names. such as the power model (Giles and Evans, 1986; Giles and Hertz, 1994). realistic group conflict (Bobo. 1983; Bobo. 1988), and competitive ethnicity (Glenn. 1966), but I will consider the most recent version called the “sense of group position" model (Bobo, 1999). The sense of group position model holds that prejudice between racial groups stems from a sense of perceived competition and racial threat to a dominant group‘s status (Blumer, 1958; Bobo, 1996; Bobo. 1998; Bobo and Hutchings, 1996). More succinctly. racial prejudice is a strategic defensive reaction to challenges to a group’s perceived relative social position. In the context of traditional White antipathy towards Blacks. racial prejudice serves to protect the integrity and position of the dominant group (Blumer, 1958). The model’s central claim is that prejudice involves more that self-interests. negative stereotypes. and negative feelings; prejudice involves positioning for scarce resources among groups in a racialized social order. There are four primary components to this model. First, competing racial group must have a sense of their position in the 31 society in which they live. Second. the sense ofgroup position should include a claim to certain rights, resources. statuses. and/or privileges. The notion of resources in the model is not limited to tangible objects such asjobs. education benefits, housing. and political gains. Resources can also be intangible claims such as positions of prestige. access to certain privileged areas, or possibly even basic civil rights such as equality and fairness (Blumer, 1958; Bobo, 1999). Third. there must be a perception of competition or threat. More recent research has tended to view competition and threat as equivalent concepts in the model (Bobo and Hutchings. 1996: Giles and Hertz. 1994). Fourth. there should be some measure of the threatened in-group‘s response to the out-group. Thus. race relations are a function of the dynamic social. economic. and political competition of racial groups in society. the more negative the context. the more likely relations are to be antagonistic. There are key advantages to incorporating this framework in the study of Blacks‘ racial attitudes. First, the group position approach goes beyond the study of the individual. Psychological models of prejudiced have been criticized as having too narrow a focus on the individual’s racial likes and dislikes (Blumer, 1958). Since. at its very foundation, racial prejudice implies the recognition of racial categories (or groups). and a stated relationship between them, racial-group prejudice is thought to be more functionally dependent on the positional arrangement of racial groups. rather than individual attitudes (Blumer, 1958). The group position approach also allows for the consideration of historically developed group relations and interests that flow from a racially stratified social order (Bobo. 1999). The group position approach views race relations as contextual or environmental. That is. race relations are shaped by socio-historical beliefs and experiences. and those contexts with greater threat are expected to produce more hostility and prejudice. The obvious disadvantage of this perspective is that the overwhelming majority of the empirical research supporting it. as well as the original version of the model. has viewed Whites as the dominant group and Blacks. and others. as the subordinate group (Bobo. 1983: Bobo. 1988; Bobo. 1999: Giles and Hertz. 1994; Quillian. 1995). Since I am concerned with Blacks' prejudice towards Whites, the model must be extended to incorporate this slight deviation. As it turns out. the model is easily extended to incorporate the attitudes of minorities. This was shown in a study of four ethnic groups in the Los Angeles area. designed to test "racial antagonism" (Bobo and Hutchings. 1996). The researchers in the study followed the traditional designs of the group position model by maintaining that perceptions of threat and competition should be grounded in racial beliefs about the social opportunity structure. To consider the attitudes of racial minorities, the study measured the sense of group position in temis of racial alienation. hypothesizing that the more alienated the group. the more competitive threat they would perceive from out- group members.‘ Implicit in the hypothesis was the idea that in a multi-racial setting. higher levels of perceived threat should be related to the extent to which a group considers itself racially disadvantaged. The authors rationalized this relationship by assuming that the sense of alienation should correspond to a racial group‘s historical position in the social structure (Bobo and Hutchings. 1996). As expected, the study revealed that Blacks were both most likely to perceive other groups as competitive 33 threats. and expressed the greatest levels of perceived alienation (Bobo and Huthcings. 1996). Bobo and llutchings’ (1996) study implies that among the predominant racial minority groups in America. Blacks are most likely to perceive threats from other racial groups. and thus should exhibit more hostility towards its competitors due to this threat. The study’s findings also imply that within race. those group members who perceive the greatest threat should exhibit the greatest levels of hostility. The current research extends the Bobo and Hutchings (1996) group position model to incorporate perceived racial threat from Whites. This is a novel approach in the study of Blacks‘ prejudice towards Whites because it explicitly considers Blacks perceptions of threat from Whites, and Blacks self-reported levels of prejudice. The group position model predicts that the more Blacks perceive a racial threat. the more racial hostility they should report. The impact of racial threat on Blacks prejudice attitudes towards Whites has not been considered in the racial prejudice literature; however. there are parallels in the research literature on political tolerance. As previously stated, studies of Blacks‘ political tolerance and intolerance, find that racial attitudes and behaviors towards Whites are possibly rooted in perceived racial threat or conflict. and are possibly a reaction to White racism or intolerance (Davis, 1994; Gibson, 1992; Gibson, 1995). If we accept these implications, then by and large, Black prejudice towards Whites is possibly the result of collective historical experiences with oppressive and intolerant Whites. The Black intolerance literature largely parallels the expectations of the group position model. except that it is the collective perceptions of the racial context that provide the impetus 34 ql for racial prejudice. The context provides cues to Blacks about their position in the social order, which in turn informs Black’s perceived levels of threat. The Plan of the Analysis The goal of this study is to provide insight into the correlates and nature of Blacks’ prejudice towards Whites. These inquiries into Blacks’ prejudice take place over three separate analyses, using three years of data from the Gallup Organization. With regard to Blacks’ prejudice, I am interested in the three basic research questions listed below. 1. Who in the Black population, as defined by demographic characteristics. is most likely to be prejudiced? 2. What are the relationships among perceived racial context, perceived threat, and prejudice within the Black population? 3. How does Blacks’ prejudice vary with racial composition? Together these questions will help create a practical picture of Blacks prejudice towards Whites, and provide a much-needed analysis of Blacks racial attitudes. General Hypotheses The following hypotheses are the expected relationships predicted by the extended sense of group position model. H1: Competitive Social Standing and “Black Frustration ”: Blacks with background characteristics which are more likely to encounter competitive social, political, and economic situations, are more likely to be prejudice towards Whites. H2: Negative Context of Group Unfair Treatment: Individuals who perceive that Blacks are treated differently because of their race are more likely to perceive greater threat. Negative Context of Experiencing Discrimination: Individuals who perceive that they have [individually] been treated differently because of their race are more likely to perceive a threat from Whites. 35 Negative Context of Competition/or Resources: Individuals who perceive that Blacks are limited in their social and economic opportunities relative to Whites, are more likely to perceive a threat from Whites. H3: Perceived Threat: Individuals who perceive a greater level of racial threat are more likely to report they are prejudiced. H4: Perceived Racial Composition and Racial Prejudice: Individuals who perceive that Blacks make up less of the racial population in the United States are more likely to be prejudiced. Also, the larger the proportion of blacks in an area, the lower the likelihood of prejudice. The Data To address the research questions, I employ three years of data collected from the, 1997, 1998, and 1999, Gallup Poll Social Audits on Black/White Relations, also known as the Gallup Race Relations Social Audits - GRRSA (Gallup Organization, 1997; Gallup Organization, 1998; Gallup Organization, 1999). The GRRSAs were systematically designed to update long-terrn Gallup Poll trends on race relations, and use a rigorous methodology with sufficiently large sample sizes for in-depth sub-group analysis. The GRRSA data are used as the basis for discussion and debate, and as an input into policy- making, journalism and scholarship on race-related issues (Gallup Organization, 1997, 1998, 1999). For these surveys, Random Digit Dialing (RDD) telephone interviews were carried out in each of the three years, with each year containing an over-sample of Blacks. While in all three years more than one racial ethnic group was surveyed, the data for this study and analysis contain only Black respondents. Samples of telephone numbers from Survey Sampling Incorporated were used to conduct the survey. The RDD samples used stratified probability designs to represent the adult population living in households with telephones within the continental United 36 of the 3111 hr >111. 163 States. The independently selected directory-listed samples used a stratified disproportionate probability design to represent directory-listed households in the continental United States, and were designed to facilitate the inclusion of an over-sample of Black respondents in 1997, and low-income households in 1998 and 1999. Estimates of both income and race were based on 1990 Census data. In all three years, respondents were randomly selected within each selected household using the “most recent birthday” method. This method involves interviewers asking to speak with the adult in the household with the most recent birthday. In, 1997 the survey employed a calling design that required interviewers to make up to 5 dialing attempts to reach a respondent within each selected household, and up to another 7 dialing attempts to complete an interview with the selected adult within the household. In 1998 and 1999, a 5 (call attempts) plus 5 (completion attempts) design was incorporated. The GRRSA studies have some important features that specifically benefit this study, and inform research on Black Americans’ racial attitudes in general. First, each year asks a self-reported prejudice item. Self-reported prejudice is not commonly found in social surveys (or experimental studies), and more importantly, rarely are Blacks asked to report prejudice towards Whites. Second, each year has stratified sample design, which allowed for an over-sample of the Black population. The rareness of three consecutive years [of a probability sample] of Blacks each with close to, or greater than 1,000 observations, makes this data set very unique for the study of Black American’s attitudes about race relations. 37 Blac pert resp IAl Rel: cal Table 1 provides a summary description of each data set. As shown, in each year, Blacks have a relatively large sample size, with an estimated sampling error of five percentage points. Also, all studies used roughly the same sampling design. The response rates for these studies were moderate, however, once eligible persons in the sampling frame were contacted, they completed the survey at a very high rate. TABLE 1. Data Characteristics for Black Respondents; 1997-I999 Gallup Race Relations Social Audit Sample CASRO Completion Sampling Year Size 3 RR b Rate ‘3 Error Sample Design 1997 1,269 46% 98% :5% RDD Stral‘fied Disproportlonate 1998 996 56% 99% i5% RDD Stratlfied Dlsproportionate 1999 1,001 50% 98% i5% RDD Strallfied Dlsproportionate Note: a N sizes are unweighted: b CASRO response rates = Completes / Eligibles + (Eligibles * Unknown) /(Eligibles * Ineligibles); c Completion Rate = Completes / Eligibles The first course of action is to conduct a descriptive analysis of the Black population. In Chapter 4, I examine which demographic categories for Blacks are most likely to be prejudiced. I take the approach that examining the aggregate level data for each demographic category helps to inform theories of prejudice. The group position model is one such theory, which is typically used to understand Whites’ group based prejudice]. While, the theory has been effectively extended to other minority groups (Bobo and Hutchings, 1996), I believe it can also be especially applied to Blacks. For instance, I explore whether or not those demographic categories that are most prejudiced, are also the most likely to be perceive a negative racial group position or context. An 38 inte 1r} llllt ('0 (D v—q dis 1211 ph interesting pattern emerges that possibly indicates that the groups most likely to be prejudiced are those who are least suspected: the Black upper and middle classes. Chapter 5 assesses the usefulness of perceptions of group position and perceived threat in understanding Blacks’ prejudice. In this section, I test the hypotheses that, in general, Blacks who perceive a threat from Whites are most likely to be prejudiced. I discuss the tenets of the group position model, and highlight the expectations of the model. The main expectation is that prejudice is a contextually condition response to racial threat. 1 operationalize threat as the perception that Whites in the community are prejudiced. The belief in the threat is conditioned by a perception that the threat is meaningful; those who perceive a more negative condition are more likely to perceive a threat. The results of the analysis show that, true to form, threat is a stronger indicator of prejudice than perceptions of the racial environment, and other demographics. Chapter 6 focuses on the impact of Black‘s population size (i.e., racial composition) on attitudes towards Whites. The study is a general replication of many studies observing the effects of context, as measure by out-group population, on Whites” racial prejudice. Research has shown that contexts with larger (or increasing) Black populations tend to show higher levels of prejudice and hostility towards Blacks, or show racially motivated political preferences (Key, 1949; F osset and Kiecolt, 1989; Giles and Hertz, 1996; Hertz and Evans, 1986; Oliver and Mendelberg, 2001; Taylor, 1998; Quillian, 1995). I test whether or not Blacks exhibit the same posture, considering their perceptions of their own population’s size, as well testing the effects of the actual population size. In doing so, I test the power-threat and group position hypotheses, which both emphasize the role of conditioned racial threat. As altemative explanations, I also test the interracial contact and ethnic community hypotheses. The results show the importance of perceived, rather than actual. population size in calculations of perceived threat, and prejudice towards Whites. Blacks’ estimates of their population size in the United States are statistically associated with both threat and prejudice, while the actual proportion of Blacks in the area has no statistical relationship with practically any of the variables. The results also show support for the group position and power-threat models. but no support for the interracial contact and ethnic community models. 40 Chapter 3 KEY MEASURES There are a number of variables that will be used repeatedly in this analysis of Blacks’ prejudice. To expedite the research points of emphasis in the following chapters, I explain the makeup and statistical distribution of those variables here. If there are changes to the variables’ composition in the remaining chapters, they will be noted in the text or table. A complete list of the items from the GRRSA included in this study can be found in Appendix A. Blacks’ Prejudice In all three years of data collection, respondents were asked to self report their level of prejudice towards Whites. Taken as a face valid measure this single item is assumed to measure the extent to which Blacks are prejudice towards Whites. The question reads as follows. “. .. I would like you to place yourself on a zero-to-ten scale, where “0” means that you are NOT prejudiced against Whites at all and where “10” means that you are EXTREMELY prejudiced against Whites. Which number would you choose to describe yourself? You can choose any number between zero and ten.” The distribution of this item was positively skewed in each of the three years. In two years, a slight majority (of respondents chose to identify themselves as “not at all prejudice” (zero). This is a potential sign of social desirability bias: not wanting to appear prejudiced to the interviewer. As a result, a choice was made to After viewing the distributions, a choice was made to recode the data. Respondents who indicated any level of prejudiced (1-10) were coded one, and all other 41 non-missing values were coded zero. Table 3 shows the distribution of prejudice responses based on the dichotomous recoding of the variable. TABLE 2. Distribution of Self-Reported Prejudice Item; 1997-I999 Gallup Race Relations Social Audit 1997 1998 1999 Total N % N % N % N % Not at all Prejudice 571 45.9% 498 51.2% 499 50.4% 1568 48.9% 127 10.2% 98 10.1% 95 9.6% 320 10.0% 117 9.4% 75 7.7% 99 10.0% 291 9.1% 112 9.0% 69 7.1% 77 7.8% 258 8.0% 70 5.6% 38 3.9% 34 3.4% 142 4.4% 146 11.7% 92 9.5% 90 9.1% 328 10.2% 23 1.8% 20 2.1% 23 2.3% 66 2.1% 40 3.2% 22 2.3% 27 2.7% 89 2.8% 21 1.7% 26 2.7% 27 2.7% 74 2.3% 10 .8% 17 1.7% 6 .6% 33 1.0% Extremely Prejudice 8 .6% 17 1.7% 13 1.3% 38 1.2% COOVQM-hWIQ—e Total 1245 100.0% 972 100.0% 990 100.0% 3207 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only TABLE 3. Percentage of Blacks indicating they are Prejudiced; 1997-I999 Gallup Race Relations Social Audits 1997 1998 1999 Total N % N % N % N % Not at All Prejudice (0) 571 45.9% 498 51.2% 499 50.4% 1568 48.9% Prejudice(l-10) 674 54.1% 474 48.8% 491 49.6% 1639 51.1% Total 1245 100.0% 972 100.0% 990 100.0% 3207 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only Table 3 consistently shows that about half of Black respondents are willing to report at least some level of prejudice towards Whites. On the other hand, the table could be interpreted as showing that half of Black respondents report that they are not at all prejudice towards Whites. In general, the table indicates that Blacks are not 42 overwhelmingly prejudice towards Whites, or that they are unwilling to report negative attitudes towards Whites. Since there are virtually no national studies of Blacks prejudice towards Whites, these figures should be taken with some caution. It is possible that Blacks are under or even over reporting their prejudice. This may occur for a number of reasons, one of which might be due to social desirability. In my analysis of Blacks’ prejudice, I include an analysis of the extent to which Blacks might exhibit response differences across the interviewers race. This type of analysis is one method of identifying underlying sensitive issues that exist in a broader social context (Schuman and Converse, 1971). As a univariate representation of prejudice, the self-report measure shows that Blacks in America are equally divided in their prejudice. The pooled data for the three years shows a difference of approximately two percentage points. Viewed as either half- empty or half-full, the data show that are neither void nor full of bigotry towards Whites. Black Preference/Ethnocentrism/In-G roup Bias The 1997 Gallup data asked Blacks three questions about their preferences for working and living alongside Whites, and attitudes towards interracial marriage. The items are listed below. “In living in a neighborhood, if you could find housing you want and like, would you rather live in a neighborhood with Black families. or in a neighborhood that had both Whites and Blacks?” “At work, would you rather work alongside mostly other Blacks, or would you rather work with a mixed group of Whites and Blacks?” “Do you approve or disapprove of marriage between Blacks and Whites?” The underlined responses in the items above indicate the theoretically “ethnocentric” response to the items. Based on the notion that Blacks’ prejudice is a 43 response to Whites’ prejudice, I expect Blacks with more in-group preference to have higher levels of self-reported prejudice. Table 4 shows that relatively few Black Americans provided an ethnocentric response. Less than one in ten Black respondents said they would rather live in a neighborhood with Black families, and only about four out of every hundred respondents said they would rather work alongside mostly other Blacks. TABLE 4. Blacks Ethnocentrism; 1997 Gallup Race Relations Social Audits Measure of Ethnocentrism % N 0 Rather live in a neighborhood with Black Families 83;; 1215 . 3.9% Rather work alongs1de mostly Blacks (32) 852 . . . 19.9% Dlsapprove of marriage between Blacks and Whites (230) 1158 0 Ethnocentric Response to at least 1 item 337? 75 6 1258 No Ethnocentric Response to any of 3 items (9 5'1) Note: Cell values include statistics for 1997 Black respondents for only. Interestingly, about one in five Blacks disapprove of marriage between Blacks and Whites. This is a large proportion relative to the other ethnocentrism measures. Part of the response to these items might be the reality of the choices; Blacks are not very likely to get their choice of residence in terms of racial composition, nor are they likely to get their choice of racial composition in the workplace. With regard to interracial marriage, a negative response (i.e., disapprove) doesn’t necessarily mean one is anti-white. A negative response may indicate an evaluation based on a belief that Whites have historically disagreed with interracial marriage, and 44 therefore, it’s best that interracial marriages don’t occur (Thomas, Leon, and Crester, 1985). Regardless of why Blacks might respond in the negative to any of these items, these numbers speak volumes about how Blacks feel about racial integration; they are highly against racial segregation. The last two rows of Table 4 show that fewer than 1 in 4 black respondents provide an ethnocentric response to any of the three items. Seventy- five percent of black respondents did not provide an ethnocentric response to any of the three items. Thus, Black Americans present no evidence that they are heavily in favor of living in the same neighborhoods with other Blacks, working alongside other Blacks, or against marrying whites. There is also a question of whether there is any difference between Black preference or ethnocentrism, and Black prejudice. Table 5 below provides some indication about the relationship between the two attitudes. This table shows the weak statistical and substantive relationships between Blacks’ self-reported prejudice and their in-group racial preference in terms of their residential neighborhood and workplace compositions. This is a strong indicator that the components of in-group preference, and prejudice towards Whites are likely different. First, only one of the three ethnocentrism measures are significant, and it’s correlation strength is quite low, rtetrachofic f .072. The other two items — “live in neighborhood” and “disapprove of marriage” — both fail to meet statistical importance. The final measure, which identifies those respondents who provided and ethnocentric response to at least one of the items, was significant, but also had a weak correlation value, rtctrachmc = .081. This final measure will be used in the 45 remainder of the analyses, primarily because it has the strongest relationship to prejudice, and because it carries with it the largest valid response size (N=1 23 5). TABLE 5. Correlation between Blacks Preference and Blacks' Prejudice; 1997 Gallup Race Relations Social Audits Blacks’ Prejudice Towards Whites Measure of Ethnocentrism R N Rather live in a neighborhood with Black Families .030 1194 Rather work alongSIde mostly .072 .1. 1223 Blacks Disapprove of marriage 042 1 140 between Blacks and Whites ' Ethnocentric Response .081 .1. 123 5 to at least 1 item Note: * Tetrachoric Correlation is significant at 0.05 level (2-tailed). At a practical level, the distinctions between preference and prejudice are blurry, however, at a conceptual level preference represents a choice, and prejudice represents an attitude. Perceived Racial Threat A racial threat is any signal of impending danger or harm. In this study, a perceived racial threat is a perception that warnings, which signal unequal and unfair treatment, are present. I consider the perceptions that Whites in the community are prejudiced to be an indication of this danger. This is not to state as fact that the perceptions are true, rather they are an omen, spawned in combination with a sense of racial group position. 46 The Gallup surveys asked respondents to indicate the extent to which Whites are prejudice. The exact question wording is below. ‘ . think for a moment about Whites in the area where you live. Use the same zero-to-ten scale, where “0 " means that they are NOT prejudiced against Blacks at all and where “10” means that they are EXT REMEL Y prejudiced against Blacks. Which number would you choose to describe Whites in the area where you live? You can choose any number between zero and ten. ” Table 6 shows the distribution of the perceived threat indicator over the three years of data, and the pooled totals. TABLE 6. Distribution of Perceived Racial Threat Item; 1997-1999 Gallup Race Relations Social Audit 1997 1998 1_9_99 Total N % N % N % N % AtXfiltlffeE‘c’l‘ice 283 24.8% 260 28.0% 257 27.5% 800 26.6% 1 57 5.0% 56 6.0% 64 6.9% 177 5.9% 2 76 6.7% 59 6.4% 61 6.5% 196 6.5% 3 105 9.2% 77 8.3% 65 7.0% 247 8.2% 4 83 7.3% 52 5.6% 52 5.6% 187 6.2% 5 231 20.3% 159 17.1% 189 20.3% 579 19.3% 6 80 7.0% 53 5.7% 56 6.0% 189 6.3% 7 94 8.2% 76 8.2% 58 6.2% 228 7.6% 8 60 5.3% 68 7.3% 66 7.1% 194 6.5% 9 30 2.6% 20 2.2% 18 1.9% 68 2.3% Exuenglifrijudice 3.6% 49 5.3% 47 5.0% 137 4.6% Total 1140 100.0% 929 100.0% 933 100.0% 3002 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents For the analysis, this question was also recoded to represent the presence (a response from 1 to 10), or absence (a response of 0), of perceptions that Whites, in an area, are prejudiced. The rationale for this coding is based on the interest in identifying those who believe that Whites in the area are prejudiced at any level, rather than the extent to which Blacks perceive the existence of prejudice. This helps eliminate any 47 potential for within-group variance in the interpretation of what different index values might mean. For example. a value of three might mean different things to different respondents, or a values that are drastically different, say 4 and 8, may mean the same thing to different respondents. The wording of the question does not explicitly identify Whites as a threat; however, the threat is easily implied if we consider the history of race relations among Blacks and Whites. Similar to previous research regarding feelings of group position (Bobo and Hutchings, 1996), I consider feelings of threat to emerge from historical experiences, feelings of deprivation and alienation, and other beliefs about treatment, conditions, and opportunities (or lack there of) faced by Blacks. Thus. it appears accurate to presume that if Blacks have been unfairly treated in the past due to Whites’ prejudice, then any perceptions that this prejudice exists, still brings about some level of threat. Table 7 below shows the distribution of values for the perceived threat measure. TABLE 7. Percentages of Blacks indicating Whites are Prejudice; 1997-1999 Gallup Race Relations Social Audits Year 1997 1998 1999 Total Perceived Threat N % N % N % N % Whitefimnm 283 24.8% 260 28.0% 257 27.5% 800 26.6% Prejudiced Whites are Prejudiced 857 75.2% 669 72.0% 676 72.5% 2202 73.4% Total 1140 100.0% 929 100.0% 933 100.0% 3002 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only Overall, about 73% of Blacks perceive that Whites in their community are prejudiced. This is a relatively high number considering that slightly over half of Black respondents had personally been treated unfairly due to their race (Gallup Organization, 48 1999). However, it is also an indication of the underlying nature of race relations in America. Blacks are still likely to perceive discrimination and prejudice exist, even if they themselves have not experienced it (Grier and Cobbs, 1968; Sigelman and Welch, 1991; Thompson et al., 1990). This type of perception is thought to be a result of a general mistrust of Whites, developed to serve as a protection for the potential of being singled out for negative treatment by Whites (Jones, 1997). By believing that Whites are prejudiced, Blacks can eliminate a large portion of the distress due to their individual shortcomings, and direct their frustration at Whites. With regard to my own analysis, this item is very important to the study of Blacks’ prejudice towards Whites because it taps how racially threatening Blacks might perceive their living areas. According to the group position model, racial threat should operate in the same direction as Blacks’ self-reported prejudice: the more racially threatening Blacks’ perceive their environment, the more likely they are to report being prejudiced. In addition, negative contextual factors should also positively correlate to perceptions of Whites’ prejudice. For example, the more Blacks perceive unfair racial treatment in the community, the more Blacks should perceive Whites in the community to be prejudice. Further, the more discrimination a Black respondent has encountered, the more prejudice they should perceive in the community. In summary, Table 7 shows that Blacks’ perceptions of Whites’ prejudice are a significant concern, at least in the minds of some Black Americans. Perceived Negative Context In past studies of political or social environments, context is typically characterized by geographic-residential boundaries. such as counties, cities, or local level 49 zip codes (Evans and Giles, I986; Giles and Evans, 1986; Giles and Hertz, 1994; Oliver and Mendelberg, 2000; Quillian, 1995; Taylor, 2000; Tolbert and Grummel, 2003). However, I view context not as an environment defined by spatial boundaries, but as the perceived circumstances, situation, perspective, or framework by which Blacks view their racial situation. This logic parallels the beliefs of the group position model extended by Bobo and others (Bobo, 1999; Bobo and Hutchings, 1996). Accordingly, the context is the perception of where your group stands vis-a-vis another racial group (Bobo. 1999), and as such, it informs the collective sense of group position. I measure the context as a combination of three perceptions about treatment and opportunities. The first, deals with Blacks’ perspectives of how they are treated. The questions ask following: “Just your impression. are Blacks in your community treated less fairly than Whites in the following situations? How about” A. On the job or at work ' On public transportation In neighborhood shops In stores downtown or in the shopping mall. In restaurants, bars, theaters, or other entertainment places In dealing with the police, such as traffic incidents. WWPOF’ These listed items were randomly presented to respondents in an attempt to remove any ordering effects. Any response indicating that Blacks are treated unfairly was coded to indicate the respondent, generally, believes that Blacks are treated unfairly. The goal is to measure whether or not Blacks perceive their racial group is treated unfairly in any given situation. Thus, even if a respondent indicated that they believed Blacks were treated unfair in all the presented situations, they are only coded as agreeing with the general perception of unfair group treatment. Table 8 shows that in each of the three years that Blacks consistently believed that “Blacks” are treated unfairly. The results of the table indicate that Blacks, in general, perceive they are in a negative racial environment. Since the category indicating a perception of unfair treatment represents a number of different situations, there may be considerable variation about the extent to which Blacks are treated unfairly. However, there is very little doubt that regardless of the setting the overwhelming majority of Blacks believe that unfair treatment of their racial group and race are connected. TABLE 8. Negative Context - Percentage of Blacks indicating Group Unfair Treatment; 1997-1999 Gallup Race Relations Social Audits Year 19 7 1998 1999 Total N % N % N % N % Blacks Not Treated 260 20.6% 208 21.0% 197 19.7% 665 20.5% Unfair 0 o BlaCkSTreatedUnfair 1001 79.4% 782 79.0% 802 80.3/o 2585 79.5/o Total 1261 100.0% 990 100.0% 999 100.0% 3250 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only The second perceived negative context is related to Blacks experiences with discrimination. In the survey, Blacks were asked whether or not they had been treated unfairly due to their race. Much has been written about the ambiguity that exists in settings of unfair treatment (Crocker et al., 1991). There is almost always a potential for uncertainty about the nature and rationale for unfair treatment, and thus attributing discrimination to race, is sometimes problematic. 51 Applying this line of thinking, the second contextual measure is not a measure of experienced discrimination per se, rather a measure of individual perceived unfair racial experiences. The specific set of items was wording in the following way: “Can you think of any occasion in the last thirty days when you felt you were treated unfairly in the following places because you were Black? How about:” In a store where you were shopping At your place of work In a restaurant, bar, theater. or other entertainment place While using public transportation In dealings with the police, such as traffic incidents FUDGE”? A response, to any one item indicating unfair treatment was taken to mean the respondent had a general perception that they, themselves, had been treated unfairly because they were black. Similar to the group unfair treatment measure, the goal of this set of items was to tap an overall belief, rather than the number of occasions associated with this belief. Table 9 shows that, on average, Blacks are less likely to experience unfair treatment due to their race. In each of the three years, less than half of Blacks indicated that they had recent experiences with unfair treatment. The results from this table coupled with the results from the previous table (Table 8) appear to indicate that Blacks appear to have a strong sense of collective group ’ identification. That is, while individuals themselves do not appear to have experienced unfair treatment, they tend to feel like Blacks, in general, are experiencing unfair treatment. The relatively low numbers in Table 9 may reflect the idea that Blacks have been unfairly treated due to their race, just not in the recent period, or that the potentially unfair treatment may or may not have been due to race; an idea commonly labeled as attributional ambiguity (Crocker and Major, 1991). Regardless of why the percentages 52 are at their stated level, the measure of individual unfair treatment is an important component to the perception of a negative context, or sense of group position. The third contextual measure attempts to gauge the socio-economic constraints Blacks perceive relative to Whites. The group position model, as well as other models associated with group power (e.g., power theory, realistic group conflict, and realistic group interest), posits that competition over resources provides the setting for racial hostility. If Blacks perceive they are losing out on social and economic opportunities, relative to Whites, they are more likely to perceive these challenges in the context of ' racial unfairness. The Gallup data include three measures associated with social and economic opportunities. The question wordings were as follows. TABLE 9. Negative Context - Percentage of Blacks indicating Individual Unfair Treatment; 1997-1999 Gallup Race Relations Social Audits Year 1 997 1998 1999 Total N % N % N % N % Individual NOT Treated Unfair Ind'v‘dual. 532 41.9% 491 49.3% 477 47.7% 1500 45.9% Treated Unfair Total 1269 100.0% 996 100.0% 1001 100.0% 3266 100.0% Note: N sizes and percentages are based on unweighted counts: Data are for Black respondents only 737 58.1% 505 50.7% 524 52.3% 1766 54.1% “In general, do you think that Blacks have as good a chance as White people in your community to get any kind of job for which they are qualified, or don’t you think they have as good a chance? “ “In general, do you think that Black children have as good a chance as White children in your community to get a good education, or don’t you think they have as good a chance?” “Again, in general, do you think that Blacks have as good a chance as White people in your community to get any housing they can afford, or don’t you think they have as good a chance?” 53 The “education as good as Whites" item, was not included in the 1997 survey, and thus respondents in that year have a slightly lower chance (only two items, instead of three) of being represented as perceiving limits on opportunities. A response to any one of these items affirming a perception that Blacks opportunities are less than equal, indicates a general perception that Blacks have limited opportunities, relative to Whites. Table 10 shows that Blacks consistently perceive they are limited in the stated opportunities relative to Whites. Of all the measures of a negative context, this set of items most closely relates to the notion of competition because it states that the opportunities for Blacks are relative to Whites. Within the item response is an almost zero-sum choice that implies that Blacks’ opportunities are lower because of Whites. This is not to imply that respondents interpreted the item in this manner, yet the finding that Blacks overwhelmingly perceive their opportunities to be less equal, signals not only their sense of group position -- less equal -- but also that the general differences in opportunities are possibly due to race, is substantively important. TABLE 10. Negative Context - Percentage of Blacks indicating Perceived Limits on Social and Economic Opportunities; 1997-1999 Gallup Race Relations Social Audits Year 1997 1998 1999 Total N % N % N % N % EquallyOpportunities 459 36.5% 314 31.6% 278 27.9% 1051 32.3% Unequal Opportunities 800 63.5% 680 68.4% 720 72.1% 2200 67.7% Total 1259 100.0% 994 100.0% 998 100.0% 3251 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only It could also be argued that the way the questions and responses where stated might prime the respondent to agree that the reason for the disparity in opportunities is 54 due to race. Possibly stating the questions without the relative mentioning of Whites might lead some to not consider the competition between the two races. However, the results seem to indicate the salience of race in relation to the opportunities, and that is the relevant characteristic of the items. Overall Negative Context The three separate perceived context items were combined to garner a more parsimonious measure of context. The three contexts reflect Blacks perceptions that they are unfairly treated as a group, that they have negative racial experiences, and that they have limited Opportunities relative to Whites. Together, the three contexts make for a face valid measure of perceived racial position, or context. Based on the group position model, I expect that more negative contexts will lead to a greater potential for perceived threat, resulting in a higher probability of prejudice. Each of the three categories of items was coded one, to indicate a negative response, and zero, to indicate a non-negative response. A negative response is one that signals a negative racial context: experiencing individual unfair treatment, believing Blacks are treated unfairly, and believing that Blacks are limited in their opportunities. The negative context index is calculated as the sum of the three contexts. The index ranges from zero to three, corresponding to the number of instances in which Blacks perceived a negative context. Table 11 shows the distribution of responses for each year and for the pooled data. In each of the three years, over 65% of Blacks reported perceiving two out of three negative contexts. In each of the three years, there are approximately 13% or less of Blacks in the “no negative context” category, and almost 12% total. The outlook of 55 ‘// the table illustrates that Blacks perceive their racial context to be more negative than positive. This data provides evidence of Blacks’ often negative view of their social, economic, and political environment; and is consistent with previous research that states that Blacks are aware of the negative racial environment in which they live (Grier and Cobbs, 1968; Jaynes and Williams, 1989; Pettigrew, 1964; Sigelman and Tuch, 1997; Sigelman and Welch, 1991; Terrell and Terrell, 1981). This is not to say that there are no positive racial contextual factors, blacks do prefer to live and work in integrated areas, and the large majority approve of interracial marriages. However, when offered an opportunity to report perceptions of unfairness, they readily claim that their context is far from racially equal. TABLE 11. Distribution of Responses to the Negative Context Index; 1997-I999 Gallup Race Relations Social Audits 1 97 1998 1999 Total N % N % N % N % NONegftlvecomeX‘ 166 13.1% 103 10.3% 109 10.9% 378 11.6% (No negative responses) Less Negative Context . 20.6% 200 20.1% 164 16.4% 626 19.2% (One negative response) Ix) O\ IN) Negal‘Ye com” 452 35.6% 326 32.7% 349 34.9% 1127 34.5% (Two negative responses) Most Negative Context . 389 30.7% 367 36.8% 379 37.9% 1135 34.8% (All negative responses) Total 1269 100% 996 100% 1001 100% 3266 100% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only The perceived context index will provide the measure of context for the remainder of the analysis. Those Blacks with the most negative context, should perceive the 56 greatest competition for valued resources and status. and thus should be most likely to report their perceptions of threat. Race of Interviewer Whenever public discussions of race occur, there is a potential for some level of emotional discomfort or tension to occurs due to the historic patterns of US race relations. And while most individuals easily take a psychological stance on race or racial issues (Carmines and Stimson, 1980), they often defer their outward expression of opinions to meet the social norms. By not violating social norms, individuals reduce the potential for discomfort, and present themselves in a favorable light (Goffman, 1959). If we consider the survey interview to be similar to a social conversation, we might expect the same social desirability bias to occur in the survey setting. One way to test for this bias is to assess the effects of the interviewers’ race on survey response. The presumption is that observable characteristics of an interviewer might suggest to a respondent, what are the socially acceptable responses. Race of Interviewer (ROI) effects are quite common in both face-to-face and telephone interviews (Anderson, Silver, and Abramson, 1988a; Davis, 1994; Davis, 1997; Finkel, Gutterbock, and Borg, 1991; Schuman and Converse, 1971); especially in studies of race. There are essentially two ROI measures to consider: actual and perceived. The interviewer’s self-reported race is the “actual” race of the interviewer (AROI) as coded in the Gallup interviewer database. In all three years, the interviewer was provided three categories to classify themselves: White, Black, and Other. Those interviewers who 57 chose the “Other” category were not included in this analysis. Table 12 shows the distribution of respondents and their interviewer’s race. The perceived race of interviewer (PROI) effect is also assessed to help account for the more realistic measure of who the respondent perceives they might be talking to, in terms of the interviewer’s race. The PROI item was only included in the 1998 and 1999 surveys, and therefore, when used in the pooled analyses, it reduces the number of valid cases to the two years of data. Nonetheless, it is an important consideration in the measurement of racial attitudes. The PROI measure was gauged by the following question. TABLE 12. Actual Race of Interviewer (AROI): Distribution of Interviews by the Interviewers Race; 1997-I999 Gallup Race Relations Social Audits 1997 199 1999 N — % N — % N % Black interviews 314 26.4% 310 33.5% 290 32.5% White interviews 876 73.6% 614 66.5% 603 67.5% Total 1 190 100% 924 100% 893 100% Note: Totals do not include interviewers who classified themselves as any race other than Black or White, and thus there are actually more respondents in the data file than are reported in the table. “This last question is just for research purposes. You may not have thought about this but I’d like to ask you to guess my race. Would you guess that I am white, black or some other race?“ Interviewers were asked to probe for the respondent’s best guess, and if necessary, remind the respondent that we only want this information for research purposes. Interviewers were allowed to tell the respondent their race after the guess. Table 13 shows the majority of respondents guessed their interviewer was White. The counts and percentages do not include respondents who said “other” or “don’t know”, nor do they include those respondents who chose not to respond. 58 When combined with the previous table that includes the actual number of interviews completed by race, respondents correctly guessed the interviewers race 78.3% of the time in 1998, and 83.7% of the time in 1999. This indicates the high, but not near perfect correlation between AROI and PROI. An initial analysis of PROI effects in the pooled 1998-1999 data shows that Blacks are more likely to say they are prejudiced when they perceive they are talking to a black interviewer, than when they perceive they are talking to a white interviewer. TABLE 13. Perceived Race of Interviewer (PROI): Distribution of Interviews by the Interviewers Race; 1997-1999 Gallup Race Relations Social Audits 199 1999 N % N % Percelved mac" 244 31.0% 264 32.0% Interviewer Perie‘ve‘iwme 542 69.0% 560 68.0% nterVIeW Total 786 100.0% 824 100.0% Note: Data only include respondents who guessed “White” or “Black.” Fifty-six percent of Blacks indicated they were prejudiced when talking to a black interviewer, compared to 45% of Blacks interviewed by Whites. This, roughly, 11% difference was large enough to be statistically significant (t = 3.96, df = 1,584, p < .001), however, the eta square statistic was .01 , indicating that only 1% of the variance in prejudice could be accounted for by the PROI. Because of their high correlation, the remainder of the analyses that consider the interviewer’s race will use the AROI variable. The effects of the PROI are duly noted, and based on this analysis, in future studies of prejudice should attempt to assess the 59 importance of the respondent’s perceptions of the interviewer’s race, in order to better gauge the ROI effect. Demographics The three data sets contain a number of demographics measured at various levels. Many of the demographic variables are recoded to deal with the fact that some categories do not discriminate well from others, and might be better understood when coded together. Such recoding will be noted when relevant; otherwise, all demographic variables should be assumed to be the same as listed in Appendix A. Along with their level of measurement, the included demographics of interest are listed in the table below. TABLE 14. Gallup Race Relations Social Audit Demographics and Measurement Levels by Year; 1997-1999 Gallup Race Relations Social Audits Demographic Level Of Measurement Gender Nominal (2 categories) Age Ratio Age Category Ordinal (5 categories Education " Ordinal (4 categories) Income Ordinal (5 categories) Urbanicity Ordinal (3 categories) Employment Status Nominal (2 categories) Black Density Ordinal (3 categories) Note: 8 Education in 1997 was recoded to match 98 and 99. These demographic items will be used as both predictors and controls in this analysis of Blacks’ prejudice. Researchers have found that socio-economic variables explain a relatively large amount of variance in Blacks’ racial attitudes, especially their attitudes about discrimination and perceptions of equality (Bobo, 2001; Sigelman and 60 Welch, 1991). Therefore I consider the more important demographics to be education and income. I hypothesize that Blacks with more educated and higher income are more perceptive of their racial environment (Bobo, 2001; Bobo and Hutchings, 1996; Sigelman and Welch, 1991), and should perceive this environment to be unfair, especially in the areas education and the workplace. Bobo and Hutchings (1996) find that the most educated Blacks have the highest levels of alienation. Sniderman and Piazza (2002), in there study of Chicago Blacks find that higher educated Blacks are more likely to believe in, what they considered to be, “Afrocentric Ideas” and “Black Autonomy.” Generally, Blacks with more income and education have more social and economic mobility, and invariably have more encounters with Whites. The more encounters they have, the more opportunities they will have to perceive they are being treated differently. These perceptions should lead to a greater likelihood of being prejudice towards Whites. There is also a perception of social entitlement that goes along with having more income and education. Most Blacks who have achieved middle to upper class status are doing so not only because they dislike the thought of poverty, but also because they believe in the American dream: success and attainment through hard work and merit. However, among middle and upper class Blacks there are also relatively high levels of alienation and disappointment (Bobo, 2001; Bobo and Hutchings, 1996; Cose, 1993; Hochschild). Blacks have come to realize that the American dream is not a certain reality. Hard work, education, and income do not automatically translate into mutually beneficial rewards for any race, but when taking into account a negative racial context, Black Americans might perceive that race is a factor in the non-realization of the dream. 61 In essence, most Black Americans have found that race still matters (i.e., has negative consequences), regardless of social status (Bobo, 2001; Hochchild, 1995). Thus, middle and upper class Blacks should be more likely to have higher levels of resentment and anger towards Whites, because they perceive they should be the least likely to have to deal with prejudice and discrimination. Summary There are a number of variables to consider in the study of Blacks’ prejudice. I specifically focus on the perceived racial context of Black Americans. A perceived negative racial environment can have negative consequences for both race relations and American politics. If Blacks perceive their social setting to be racially discriminatory, racially unequal, and limiting in terms of opportunities for their own race, they should be more likely to perceive their environment to be threatening. Blacks respond to this threat not only with frustration about the future or race relations, but also resentment, anger, and prejudice towards the dominant cultural group. Chapter 4 A DEMOGRAPHIC PROFILE OF BLACKS’ PREJUDICE The lack of nationally representative studies of the Blacks population, have left many questions unanswered, especially with regard to sensitive topics such as prejudice. One curiosity centers around “who” is most likely prejudiced in the Black population. The group position model and other models based on inter-group competition, state that individuals who are most likely to be in competition with an out-group, are most likely to develop hostile or prejudiced attitudes towards that out-group (Blumer, 195 8). This chapter addresses the question of which demographic segments of the Black population are most likely to be prejudiced against Whites? There is evidence that suggests that professional middle class Blacks are the most likely candidates because they are a group highly shown to be susceptible to frustration, disillusionment, alienation, and mistrust (Bobo, 2001; Cose, 1993; Feagin and Sikes, 1994; Hochschild, 1995). This segment includes those Blacks with higher incomes, more education, and others who are seeking economic upward mobility. Although, not all research supports this proposition (Sniderman and Piazza, 2002), continued exploration of its validity is necessary. The goal of this chapter is to test the notion, which some have called “Black Rage” in the upper and middle classes of Black America (Cose, 1993), and better identify and understand which demographic characteristics are most associated with Blacks’ prejudice. Demographics and Racial Attitudes The most extensive examination of the impact of personal characteristics on Blacks’ racial attitudes was conducted in the early 1990s. Sigelman and Welch (1991) 63 believed that demographic variables had a strong potential to explain Blacks’ perceptions of racial discrimination. They reasoned that various background and demographic variables helped to shape the perceptions and realities, and opportunities for Blacks. For example, older Blacks that experienced the struggles for civil rights might be most attuned to the fact that prejudice and discrimination exist. Also, Sigelman and Welch suggest that interracial contact might influence Blacks’ attitudes towards Whites. For example, Blacks with higher levels of education are perhaps more likely to face job discrimination or prejudiced attitudes in the workplace, because they are more likely to encounter Whites and other races there. I test this notion in Chapter 6. Sigelman and Welch (1991) acknowledge that there are far too many personal characteristics to fully capture what literally accounts for differences in personal perceptions. experiences and opportunities. However, their point is that persons who are more likely to perceive treatment and opportunity differences between Blacks and Whites, might also be more likely to blame these differences on racial prejudice and discrimination. Further, those who are more likely to attributes differences in opportunities and status are those who are more likely to perceive that the racial environment is negative and potentially threatening. It’s no secret that Blacks are sensitive to the racial perceptions and actions of the Whites majority (Gibson, 1995; Sigelman and Tuch, 1997). These racial signals, also manifested as negative racial climate, or context, are strong indicators of Blacks’ perceptions of threat (Bobo and Hutchings. 1996) and their racial attitudes (Evans and Giles, 1986; Livingston, 2002). 64 The group position model explicitly considers the perceptions that racial groups are treated different, and makes a prediction based on such perceptions (Bobo, 1999). The more Black’s perceive they live in a negative racial context, the more likely they are to perceive a competitive threat (Bobo and Hutchings, 1996). It follows that Blacks, who perceive a more negative racial environment, as measured by perceptions of differences, might also be more likely to report being prejudiced. This descriptive analysis provides an opportunity to address which of Blacks demographic categories are most likely to report being prejudiced. Primarily for descriptive purposes, this analysis only includes demographic variables. In the absence of psychological and behavioral measures, demographics are often the only available information; therefore, this initial analysis considers only the demographic characteristics of Blacks as exogenous factors. This investigation will help provide baseline measures of self-reported prejudice, and to allow for future comparisons across studies. Connecting the tenets of the group position model, I attempt to explain why certain demographic groups are most likely to be prejudiced. I expect that those demographic categories most likely to feel a negative sense of group position (i.e., perceived unfairness due to race) are most likely to report being prejudiced. Studying Blacks’ Racial Attitudes: Demographics and Prejudice A number of studies observe small numbers of Blacks, and their racial attitudes (Brink and Harris, 1964; Marx, 1967; Noel and Pinkney, 1964; Paige, 1970; Sigelman and Tuch, 1997; Sigelman and Welch, 1993; Sigelman and Welch, 1999; Wojniusz, 1979; Works, 1961). In addition, previous studies of the relationship between negative racial attitudes and demographics have either not studied Blacks’ racial prejudices (e.g., 65 Sigelman and Welch, 1991) or found show conflicting conclusions (Noel and Pinkney, 1964; Works, 1961). Sigelman and Welch (1991) study a myriad of demographic variables only to find two that lead to statistically significant relationships. Both of the variables centered on economic situations: home ownership and economic pressure (i.e., difficulty paying bills). None of the other variables -- age, education, income, gender, employment status, social contact with Whites -- showed a significant impact. Other researchers have found little to no relationship between pessimistic racial attitudes and demographics, such as age (Klugel and Smith, 1986; Parent, 1985), gender (Sniderman and Hagen, 1985), socio-economic status (Giles and Evans, 1986; Hamilton, 1976), and social contact (Robinson and Preston, 1976). Finally, Jaynes and Willarns (1989) find that while Blacks perceive a significant amount of unfairness, hostility, and threat from Whites and predominantly White social institutions, there is little evidence that these differences in perceptions transcend class, social status, age, or gender. These findings might lead some to believe that Blacks’ demographics may be most valuable only as statistical controls in multivariate models of negative racial attitudes, or at the very least might follow the same covariance patterns as demographics for Whites. The one variable that receives considerable attention, regardless of its empirical effect, is education. According to Schuman et al. (1997), more educated Blacks are more likely to believe there is discrimination in jobs, housing, and police treatment, and to believe that Whites don’t care about Blacks. This is an important signal to the study of prejudice because it implies that those Blacks with more education are more likely to be aware of negative racial contexts. More education has also been strongly associated with a number of other racial attitudes, including greater perceptions of competitive threat and 66 alienation (Bobo and Hutchings, 1996), political frustration (Dawson, 1993), Black “rage” (Cose, 1993), intolerance (Davis, 1995), and disillusionment (Hochschild, 1995). As Sigelman and Welch (1991) state, “education can be viewed as a source of enlightenment, fostering knowledge about members of difference races, teaching people to recognize prejudice and to understand its dangers (p. 42).” For example, addressing the relationship between education and law enforcement, evidence shows that Blacks with the highest levels of education are most likely to perceive unfair treatment by the police (Schuman et al., 1997). Ninety percent of Blacks with seventeen or more years of education believe that police treat Blacks unfairly. Eighty-eight percent of Blacks with sixteen years of education felt the same way. These figures were compared to 54% for Blacks with zero to eight years of education, 68% of Blacks with nine to eleven years of education, and 73% of Blacks with twelve years of education (Schuman et al., 1997). Researchers have also found connections between Blacks political activities and education. Dawson (1993) and others (Nie and Verba, 1972; Tate, 1993) found that more educated, higher income, older, and employed Blacks are more likely to be involved in political activism and voting. This implies that Blacks with higher socio-economic status and age are those most likely to be politically competitive with other groups, and thus according to Blumer (195 8) and Bobo (1999) may be most likely to be prejudiced. It also reasons that those Blacks in racially mixed areas, such as low black density and suburban or rural areas, are more likely to compete with Whites for opportunities like housing, jobs, and education, and thus also more likely to be prejudiced. Interviews, of primarily Black middle class professionals, conducted by Cose (1993) show Blacks’ perceptions of a “brick wall” in society. The aforementioned wall is 67 characterized by discrimination, racial stereotypes, and other various manifestations of racial hostilities. Also a more recent study by Sigelman and Tuch (1997) shows that Blacks who are more likely to interact with Whites for extended periods are more likely to believe Whites hold negative images of Blacks. Taken together this evidence of perceived negative context and middle class disaffection provide compelling evidence that more educated and higher income Blacks are more likely to hold prejudice attitudes towards Whites. Yet, there are dissenting opinions. Some researchers believe that the Black middle class is not disillusioned. In one of the most recent analyses of Blacks’ prejudice, Sniderman and Piazza (2002) provide evidence that more educated and higher income Blacks are more likely, than their lower income and educated counterparts, to believe that their financial situation will be better off in the future, and there is hope for the future, as measured by anomia scores. Together, these two findings are said to provide evidence that the Black middle class still has belief in the American dream (p. 135-136). Oddly, in an adjacent section of their book, Sndierrnan and Piazza (2002), conduct a “Life Chances” experiment, where they find that Blacks “strongly agree” with the notion that if, generally, people work “hard enough they can make a good life for themselves.” However, they do not provide a statistical or visual breakdown by income or education. Readers of the study are left with a question of whether or not the Blacks middle class actually has different attitudes in this experiment. This basically leaves the general research question unanswered. Is there evidence that the Black middle class disillusioned with the American dream? 68 I attempt to build on this line of reasoning by testing the group position hypothesis, which states that those groups who perceive a more negative context, are more likely to be prejudiced. If categories that represent the Blacks middle class are more likely to perceive a negative context, then they should also be more likely to be prejudice. This finding would support arguments by Cose (1993), Hochschild (1995), Bobo (2001), Jaynes and Williams (1989), and F eagin (1991) that suggest that middle to upper class Blacks believe less and less in the strict ideals of the American dream. In the next sections I lay out the proposed relationships among negative context, and prejudice, and show the relationships between the demographics and Blacks’ prejudice. Perceived Racial Environment as Context Traditionally, Blacks’ perceptions of unfair treatment and discrimination in American have been associated with the perception of alienation (Bobo, 2001; Bobo and Hutchings, 1996; Hochschild, 1995; Jaynes and Williams, 1989; Sears and McConahey, 1973; Sigelman and Welch, 1991). Racial alienation relates to how a group feels about their experiences, considering both the history and implications of racial and ethnic stratification (Schuman and Hatchett, 1974). In the words of Bobo and Hutchings (1996), “feelings of racial alienation reflect the accumulated persona, familial, community, and collective experiences of racial differentiation, inequality, and discrimination.” Group based alienation is undoubtedly a sense of isolation, a perception that there is a separation between two or more groups. Groups who are more likely to feel alienated tend to have a profound sense of group “disenfranchisement” (Bobo, 1999) and grievance. They are also more likely to sense they are subordinate to a dominant group. 69 The dominant group is thus a provider of competitive threat (Bobo and Hutchings, 1996), and alienation acts as a contextual factor in a group’s sense, or perception, of social position. As such, it fits with in the context of the group position theory. The group position model (Blumer, 1958; Bobo, 1999; Bobo and Hutchings, 1996) posits that competition for social and economic resources, gives rise to a perception of threat against the outside competing group. This notion has traditionally been applied to Whites, and their beliefs that Blacks are growing in number, and encroaching on the limited resources, statuses, and privileges afforded in America (Bobo and Hutchings, 1996). The potential loss of these prerogatives leads Whites to develop both defensive and offensive attitudes towards Blacks. The reactions are defensive in the sense that Whites already have the socio-economic statuses and privileges, and must react when they are faced with potentially sharing or loosing them. These reactions are considered negative because prejudice seeks to sustain a framework for a continued stratified racial order. Thus, Whites realize that their racial position in the hierarchy of groups is threatened by other racial-ethnic-religious groups, and seek to diminish the threat both psychologically and materially. Bobo (1999) has been the lead proponent in extending this theory to other minority groups. He incorporates the notion of alienation as a sense of group position. He posits that alienation acts as a context for competition. The greater the sense of alienation for a particular group, the more likely the group is to perceive competition and threat. Bobo and Hutchings (1996) found that Blacks had the deepest levels of alienation and perceived the greatest threat from Whites. The implication is that Blacks might also 70 be most likely to report being prejudiced against Whites. since they are the main threat provider. Middle class Blacks are those who are most likely to have higher expectations of the social — economic system. They have achieved relatively higher levels of education and income, and the result is assumed to be upward mobility, both in the economic and social arena. However, Blacks also have a greater sense that they are in competition for both tangible and intangible benefits afforded most White Americans (Wilson, 1973). This competition is conditioned by a belief that Blacks are a racially subordinate group in American, and the subordination has historically been at the hands of Whites. This belief gives rise to some conspiratorial type statements that put Blacks up against, “the system”, or “the man” (Cose, 1993). It is this racially conditioned environment that is the breeding ground for Blacks’ prejudice towards Whites. Segmenting the Black Population Segmentation analyses are typically used in research to extrapolate those demographics that best relate to attributes or behavior. Segmentation studies are often performed on aggregate level data. Aggregated data are the composite values of different categories, and thus, the aggregated groups exist as the units of analysis. The goal of this segmentation study is to identify which demographics for Blacks are most likely to be prejudiced. This will help provide some initial understanding of underlying roots of group-based prejudice towards Whites. Research suggests that there are three generally important demographic correlates of racial prejudice for Whites: region of the country, age, and education (Maykovich, 1975; Oskamp, 1991). However, for the Black population the reviews are mixed. Studies 71 in the 1950’s and 1960’s, found that Blacks who were most susceptible to authoritarian personalities were more likely to have anti-white attitudes (Paige, 1970). For example, research found anti-white attitudes were more common in Blacks with either lower education, who were women, born in the South, but living in the North, or were younger (Marx, 1967; Noel, 1964; Williams, 1964). Studies from the 19708 hypothesized that hostility and alienation were due to in-group favorability, militancy, and mistrust (Farley, Hatchett, and Schuman, 1979; Schuman and Hatchett, 1974). By the 19805 and early 19908, most of the variance in Blacks’ racial attitudes towards Whites was being attributed to interracial contact (Sigelman and Welch, 1993), mistrust (Taylor and Taylor, 1981), or perceptions of disenfranchisement and competition (Evans and Giles, 1986). My analysis centers around eight demographic variables, education, income, age, gender, urbanicity, Black density, and employment status. Following the propositions of models based on group competition, I expect that any demographic category that has greater feelings of alienation, competition, or threat will show greater likelihood of prejudice. Data Analysis To test the hypothesis that Blacks most associated with alienation (i.e., middle and upper class Blacks) were most likely to report being prejudiced, I employed all three years of the Gallup data. I made the decision to combine the three years of data for primarin three reasons. First, there was no real interest in looking at the demographics across the three separate years. The data were collected in three consecutive years, so it’s reasonable to presume that any effect during any particular year of data should have a similar effect on the preceding, or following years. Second, by combining the three years of data, the segment sizes remain relatively large enough to make numerous comparisons among the demographic categories. Third, combining the data allows for a more succinct and parsimonious analysis. I conducted an aggregate level analysis using Black demographic segments as the unit of observation. The demographics of interest were gender (2 categories), education (4); age (4), income (5), employment status (2), Black population density (3), urbanicity (3). and a combination of high Black density and urbanicity (2). The final analysis consisted of twenty-five demographic categories. Next, I calculated the proportion of prejudiced respondents for each segment, and then used the distribution of values to create three groups based on percentile rank. The cut-off score for the lowest prejudiced segments was 49.5%; any segment reporting this percentage or lower were considered “less prejudiced.” Any segment with a reported proportion between 49.6% and 54.9% was considered “moderately prejudiced.” The final segment consisted of groups with reported prejudice greater than or equal to 55%. I identified these segments as the “most prejudiced” group. With these three basic breakouts, one can discern which demographic segments were most likely to report being prejudiced. For each demographic category, I also report the percentages of respondents who provide a negative racial context response to any three of the sets of context measures: group unfair treatment, individual unfair treatment, or limited opportunities. The specific items included in the context measures can be found in Chapter 3 of this study and Appendix A. 73 Before moving to the more general segmentation of all demographic categories, I explore each of the demographics in detail. I am interested in identifying which within- variable categories are most likely to perceive a negative context, perceive a racial threat, and report being prejudiced towards Whites. In an effort to highlight substantive differences across the groups, rather than statistical differences, I do not show the statistical tests across categories. The primary reason is due to the large sample sizes, which tend to show statistical differences. With a total sample of approximately 3,200 cases, 1% equates to about 32 respondents. As a rule of thumb for this data, if the differences across categories are greater than or equal to 4% (or 128 respondents) points, then they are likely statistically different numbers. Education Bobo and Hutchings (1996) report a positive relationship between education and feelings of alienation for Blacks. This is consistent with their group position hypothesis, which stated that those racial groups with a greater sense of alienation are most likely to perceive a competitive threat, and are thus more likely to exhibit prejudice. I expect that Blacks with higher levels of education will be more likely to report being prejudiced towards Whites. More educated Blacks are in constant social and economic competition with Whites for upward mobility, and have traditionally been found to have higher levels of alienation. They are also the most common group in the Black middle class. Tables 15a and 15b, show the breakdown of negative context, perceived threat, and prejudice. Both tables show that Blacks with higher education tend to perceive the racial environment to be more negative, perceived a greater threat from Whites, and are more likely to be prejudiced, than Blacks with lower education. This finding is consistent 74 with other studies that show more educated Blacks tend to be acutely aware of their racial environment (Cose, 1993; Sigelman and Welch, 1991). TABLE 15a. Education Category by Racial Context Variables; 1997-1999 Gallup Race Relations Data Education Blacks Individual Black have Category Unfairly Treated Unfairlv Treated Unequal Opportunities % N % N % N LT HS 76.1% 744 45.1% 747 61.3% 739 HS Grad 77.9% 1073 43.3% 1079 67.0% 1078 Some College 84.9% 780 50.5% 782 71.1% 781 College Graduate 81.2% 617 45.8% 620 73.6% 617 Total 79.8% 3214 45.9% 3228 67.9% 3215 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only TABLE 15b. Education Category by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data Education Prejudice Category Perceive Threat Towards Whites % N % N LT HS 65.9% 678 45.2% 730 HS Grad 73.7% 987 49.7% 1059 Some College 76.9% 741 55.6% 779 College Graduate 779% 574 56.2% 607 Total 73.5% 2980 51.3% 3175 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only. Another noticeable pattern is that, with the exception of individual unfair treatment, Blacks with less than HS education tend to always have the lowest percentages. They are the least likely to perceive Blacks are unfairly treated, the least likely to perceive that Blacks have limited opportunities, the least likely to believe that Whites are prejudiced, and the least likely to be prejudiced towards Whites. These differences potentially highlight differences in the experiences of Blacks. It’s possible that as Blacks attain more education, they tend to involve themselves in more racially 75 competitive environments, such as the job and housing markets. Again, given the ambiguity of negative situations it’s likely easier for educated Blacks to attribute any perceived mistreatment towards themselves or others to race. Income Similar to education, I expect Blacks with higher incomes to be more likely to be prejudiced. Due to the notion that those with higher incomes are more likely to be members of the middle class, and middle class persons are more likely to perceive alienation and competitive threat. Moderate to high income is a strong indicator of those who are in a competitive economic market. Income implies work, and work signifies performance. Because persons who have higher paying jobs are typically expected to perform at a higher level, I expect they are also more likely to face competitive threats from others who seek both their job and income. These threats can be real (i.e., job evaluations and cuts) or perceived (i.e., training a subordinate for a similar position). The bottom line is that in a free market employment-based economy, one of the most constant threats is to ones economic subsistence. I expect this threat to translate into higher levels of out-group hostility or prejudice. The analysis reveals that higher income blacks are overwhelmingly more likely to perceive a negative context, perceive a racial threat, and report being prejudiced. Table 16a shows that 8 out of 10 Blacks in the top three income categories believe that blacks are unfairly treated. The table also shows that Blacks who earn the highest incomes are the most likely to report that they themselves have been mistreated, at 54.3%. Table 16b show that Blacks, who have household incomes of $55K or more, are clearly at the top of the list of Blacks who perceive a racial threat. Their probability of 76 being prejudiced is 25% points higher than the lowest income category. This is further evidence of that the Black middle and upper classes are most likely to feel disillusioned and alienated in American society. These results coupled with those in Tables 15a and 15b, provide the cornerstone of my argument that Blacks’ prejudice is a response to perceived negative contexts and racial threat. TABLE 16a. Income Category by Racial Context Variables; 1997-1999 Gallup Race Relations Data Income Blacks Individual Black have Category Unfairly Treated UnfairLLIreated Unequal Opportunities % N % N % N LT 25K 77.4% 1287 42.6% 1295 65.4% 1287 25K to LT 35K 81.1% 540 48.5% 540 66.9% 540 35K to LT 45K 86.5% 341 52.0% 342 70.5% 342 45K to LT 55K 86.3% 240 50.8% 240 72.4% 239 55K or more 86.7% 474 54.3% 475 78.1% 475 Total 81.4% 2882 47.4% 2892 69.0% 2883 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only TABLE 16b. Income Category by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data Income Prejudice Category Perceive Threat Towards Whites % N % N LT 25K 64.9% 1190 41.3% 1275 25K to LT 35K 79.1% 507 54.9% 534 35K to LT 45K 78.7% 329 61.5% 340 45K to LT 55K 84.4% 231 65.8% 240 55K or more 86.8% 453 66.5% 472 Total 74.5% 2710 52.5% 2861 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only Employment Status Similar to income, employment status is a primary component of social competition. Any person that has ever been through a job search process can sympathize 77 with the discomfort of being evaluated. Often this feeling can translate to on the job pressures. Being employed might also be thought of as “staying” employed. Higher perceived levels of competition may stem from feeling that the loss of a job stands in the way of economic survival. Thus, Blacks with jobs (i.e., employed) should be more likely to be prejudice, than Blacks without jobs. TABLE 17a. Employment Status by Racial Context Variables; 1997-1999 Gallup Race Relations Data Employment Blacks Individual Black have . . Category Unfalrly Treated Unfalrly Treated Unequal Opportunities % N % N % N Unemployed 72.2% 909 31.2% 921 63.6% 914 Employed Full-Part Time 82.8% 2295 52.1% 2298 69.6% 2291 Total 79.8% 3204 46.1% 3219 67.9% 3205 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only TABLE 17b. Employment Status by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data Employment . Prejudice. Category Perceive Threat Towards Whites % N % N Unemployed 60.9% 798 39.4% 885 Employed Full-Part Time 78.2% 2172 55.7% 2282 Total 73.5% 2970 51 . 1% 3167 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only ' mm. Tables 17a and 17b show that employed Blacks are more likely than unemployed Blacks that perceive a negative context, perceive a racial threat, and be prejudiced. The results from these two tables likely indicate that Blacks who are working experience more negative racial situations, than those Blacks who do not work. This is further evident by the slightly greater than 21% point difference the individual unfair treatment 78 number. In general, employment status shows one of the largest between group variances of all the demographic categories. Age The relationship between age and prejudice is difficult to gauge. In general, older persons should be more likely to believe that Whites are prejudice, however, this doc not automatically translate into a belief that the social and economic differences between Blacks and Whites are due to prejudice. Bobo and Hutchings (1996) find that feelings of alienation tend to decrease with age. If this is true, then younger persons are more likely to report being prejudiced. I believe the age groups most likely to perceive alienation are those who represent the most economically competitive group, the working middle class. Typically ranging between 30 and 50 years of age, these Blacks are the segment most likely to be near the average working age. These are Blacks that have gotten beyond the initial entry into the job market, yet have not reached the age of retirement. They have not only gone through the initial years of employment and education, but are potentially in their stages of child rearing and family development. Thus, they not only have their own experiences to consider, but also the experiences of their wives, children, and extended friends and age cohort. I expect that these middle-aged persons (or non- extreme) are more likely to be prejudiced. The results seen in tables 18a and 18b, present an intriguing picture. It appears that those most likely to perceive a negative context, are those who are most likely to either be in the labor force (i.e., non-retired age) or those who are potentially just starting families. Generally, Blacks between the ages of 18 and 53 are most likely to perceive that Blacks are unfairly treated. Blacks between the ages of 18 and 27 are the most likely 79 to perceive they have been unfairly treated because of their race. Finally, Blacks 38 to 53 are most likely to perceive they have unequal opportunities relative to Whites. Interestingly enough, the youngest age category is the least likely to perceive they have unfair opportunities, possibly indicating that they have yet to experience (e.g., housing and jobs market, and education for their children) the potential tribulations of such opportunities. TABLE 18a. Age Category by Racial Context Variables; 1997-1999 Gallup Race Relations Data Age Blacks Individual Black have Category Unfairlv Treated Unfairly Treated Unequal Opportunities % N % N % N 18-27 84.8% 599 60.2% 600 59.8% 600 28-37 82.8% 681 55.1% 681 68.8% 680 38-53 81.9% 1032 48.5% 1035 72.7% 1032 54+ 71.2% 873 27.0% 884 67.3% 874 Total 79.7% 3185 46.2% 3200 68.0% 3186 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only TABLE 18b. Age Category by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data Age Prejudice Category Perceive Threat Towards Whites % N % N 1827 74.9% 577 54.8% 599 28-37 79.2% 643 55.8% 678 38-53 78.9% 977 55.3% 1024 54+ 60.2% 757 40.4% 850 Total 73.4% 2954 51.3% 3151 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only With regard to perceived threat and prejudice, there is very little variance across age categories, except for those Blacks who are 54 years or more, they are considerably less likely to‘perceive Whites in the areas are prejudiced, and report being prejudiced. I 80 believe this is due to the levels of racial acceptance for older Blacks. Blacks who were 53 or more during the data collection were, between 20 and 25 years younger during the mid-19605; which makes their age during that between 18 and 23 years old. Over the course of those twenty something odd years, they’ve seen the change in America and can likely appreciate the virtual elimination of Jim Crow laws and racial unrest. Because of this appreciation for change they are probably more likely to give Whites the benefit of the doubt and say they are not prejudiced, and claim that they themselves are not prejudiced. Urbanicity Primarily because studies in the late 1960s showed that urban residents are more likely to feel a sense of alienation, and mistrust, I might expect Blacks in urban areas to be more prejudiced than Blacks in non-urban areas (Campbell and Schuman, 1968; Schuman and Hatchett, 1974; Sears and McConahey, 1973). However, most of the studies showing this relationship analyzed data collected in urban areas such as Los Angeles, CA, and Detroit, MI, in the wake of urban riots. If urban residents are indeed more likely to be aware of their racial group position (i.e., context), then it follows that they should be more likely to be prejudiced. However, this might clash with the belief that, on average, urban areas might produce fewer encounters with Whites, and that fewer encounters might lead to a diminished perception of the extent of racial prejudice. I believe that Black residents in suburbia are more likely to encounter Whites, or mixed racial environments, and are thus more likely to perceive greater socio-economic and political competition in those areas. 81 The analysis of urbanicity shows moderate variation across the racial context measures. Tables 19a and 19b show the maximum difference between categories for any of the social context measures is 7% points (seen under “Individual Unfairly Treated”) between those in suburban (47.3%) versus rural (40.7%) areas. It’s sensible to see that urban and suburban blacks are more likely than rural blacks to perceive that blacks as a group are treated unfairly, but the 2% points variance across the urbanicity categories for unequal opportunities is somewhat surprising. I would expect that blacks in suburban areas and rural areas might perceive opportunities to be more constrained than in urban areas. TABLE 19a. Urbanicity by Racial Context Variables; 1997-1999 Gallup Race Relations Data Urbanicity Blacks Individual Black have Category Unfarirly Treated Unfairly Treated Unegual Opportunities % N % N % N Urban 80.6% 1914 46.7% 1923 68.0% 1916 Suburban 80.4% 822 47.3% 827 66.4% 821 Rural 74.3% 514 40.7% 516 68.7% 514 Total 79.5% 3250 45.9% 3266 67.7% 3251 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only TABLE 19b. Urbanicity by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data Urbanicity Prejudice Category Perceive Threat Towards Whites % N % N Urban 71.8% 1747 51.4% 1893 Suburban 78.8% 782 55.2% 812 Rural 70.2% 473 43.4% 502 Total 73.4% 3002 51 . 1% 3207 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only 82 Observing tables 19c and 19d, reveals the findings of an analysis observing the relationship between urbanicity and education. The tables show that college educated (28.1%) and highest ($55K+) income (41.5%) Blacks are more likely than all other education and income categories to live in suburban areas, and may thus account for the threat and prejudice statistics in the suburban category. TABLE 19c. Urbanicity by Education Categories; 1997-1999 Gallup Race Relations Data 13:12:23? Urban Suburban Rural % N % N % N LT HS 58.9% 440 23.0% 172 18.1% 135 HS Graduate 56.4% 609 25.6% 276 18.0% 194 Some College 62.5% 489 25.2% 197 12.3% 96 College Graduate 58.4% 362 28.1% 174 13.5% 84 Total 58.9% 1900 25.4% 819 15.8% 509 Note: Black Respondents only; Percentages are the percentage of the row categories. TABLE 19d. Urbanicity by Income Categories; 1997-1999 Gallup Race Relations Data 13:21:? when M Rural % N % N % N LT 25K 59.5% 770 18.4% 238 22.2% 287 25K to LT 35K 61.5% 332 24.8% 134 13.7% 74 35K to LT 45K 56.4% 193 30.4% 104 13.2% 45 45K to LT 55K 64.6% 155 27.1% 65 8.3% 20 55K or more 53.5% 254 41.5% 197 5.1% 24 Total 58.9% 1704 25.5% 738 15.6% 450 Note: Black Respondents only; Percentages are the percentage of the row categories. While suburban Blacks do not perceive higher levels of negative context, they do highest likelihood of perceived threat, and prejudice. This evidence, combined with the socio-economic status of those living in suburban areas, is further support for the 83 disaffection and “rage” that exists within the Black middle and upper classes, particularly for those living in suburban areas. Black Density Similar to urbanicity, Blacks who live in areas with higher Black density should be more likely to report being prejudiced, than those Blacks who live in moderate to low Black density areas. However, density also has a potentially mixed affect. Group position theory might expect that Blacks that live around more Blacks are more aware of their group position and the differences between Blacks and Whites. On the other hand, Blacks who live in moderate to low density areas might have more opportunities to encounter potential situations that might lead to discrimination or prejudice. In general, this will also make for an interesting finding because it implies that Blacks potentially diminish or enhance their perceived levels of negative context base on where they live. Tables 20a and 20b show that Blacks who live in medium Black density areas are most likely to perceive a negative racial environment. They share this distinction with Blacks in high black density areas for the group unfair treatment and perceived unequal opportunities contextual items. TABLE 20a. Black Density by Racial Context Variables; 1998-1999 Gallup Race Relations Data - Blacks Individual Black have B1 k D 3&6;ng Unfairlv Treated Unfairlv Treated Unequal Opportunities % N % N % N High 79.6% 1665 47.3% 1671 70.3% 1666 Medium 80.5% 282 55.3% 284 71 .1% 284 Low 73.8% 42 47.6% 42 61.9% 42 Total 79.6% 1989 48.5% 1997 70.3% 1992 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only 84 TABLE 20b. Black Density-Urban by Racial Context Variables; 1998-1999 Gallup Race Relations Data Density-Urban Blacks Individual Black have Category Unfairly Treated Unfairly Treated Unequal Opportunities % N % N % N All others 78.6% 896 49.7% 899 69.5% 898 High black density-urban 80.5% 1093 47.4% 1098 70.9% 1094 Total 79.6% 1989 48.5% 1997 70.3% 1992 Note: N sizes and percentages are based on unweighted counts: Data are for Black respondents only However, there are very small percentage differences for the racial context measures when high black density and urban residence are combined and compared to the other black density and urbanicity categories. Yet when it comes to prejudice, Black respondents who live in low black density areas are most likely to perceive a racial threat, and to be prejudiced towards Whites. Table 20c shows that low black density areas are 12% higher in perceived threat and 11% points higher in prejudice, than high black density areas. Some of this effect of low black density areas can be seen in Table 20d, which shows that the combination of high black density and urban perceive less (-7%) threat than the other categories of the two demographic variables, however, they are roughly equal in their percentage of reported prejudice. The analysis of black density reveals that higher black density areas are not, in and of themselves, breeding ground for negative attitudes. This is counter to the ethnic community hypothesis often promoted by political scientists (e.g., Ellison and Gay, 1991; Evans and Giles, 1986; Gutterbock and London. 1983; Olsen, 1970; Patchen, 1982) studying attitudes such as political efficacy and group identity. The ethnic community model states that areas with higher minority populations provide an environment for 85 increased activism, empowerment, and ethnocentrism. It appears that areas with the highest proportions of prejudice blacks are those in low black density, suburban areas. TABLE 20c. Black Density by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data . Prejudice Blg:;::1ty Perceive Threat Towards Whites °/o N % N High 70.8% 1548 48.7% 1639 Medium 78.8% 273 50.5% 281 Low 82.9% 41 59.5% 42 Total 72.2% 1 862 49.2% 1962 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only TABLE 20d. Black Density—Urban by Perceived Threat, and Prejudice; 1998-1999 Gallup Race Relations Data D ’ 't U b Prejudice ensr - r an C atzg 0 ry Perceive Threat Towards Whites % N % N All others 76.0% 858 48.8% 883 H‘gh ”3336“"? 69.0% 1004 49.5% 1079 Total 72.2% 1862 49.2% 1962 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only Gender I expect males are more likely than females to be prejudiced. First, males usually fit into the gender role of primary breadwinner, and are thus more likely to be in competition for work. Secondly, Black males have traditionally been the objects of racial animus (e.g., beatings and lynching), often denied the opportunity to fulfill their role as breadwinner, and thus are more likely to have a sense of their historical position with regard to both race and gender. Although some studies have shown no differences in 86 perceived discrimination and unfair treatment across gender (Sigelman and Welch, 1991; Sniderman and Hagen, 1985), I expect men to have a greater likelihood of being prejudiced than women. The main rationale for this expectation is men’s awareness their racial environment and history. Table 21a, reveals that males have a slightly higher chance of perceiving a negative context in all three of the variables. For the three context factors, the largest percentage difference across gender is 5% (individual experiences with unfair treatment). Otherwise, it is safe to conclude that Black men and women perceive similar racial contexts. TABLE 21a. Gender by Racial Context Variables; 1997-1999 Gallup Race Relations Data Gender Blacks Individual Black have Category Unfairlv Treated Unfairly Treated Unequal Opportunities % N % N % N Male 81.1% 1217 49.2% 1222 69.7% 1217 Female 78.6% 2033 44.0% 2044 66.5% 2034 Total 79.5% 3250 45.9% 3266 67.7% 3251 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only Table 21b shows that males are more likely than females to perceive a racial threat from Whites, and to be more prejudice. This is likely due to the nature of male gender roles in the Black community. However, the level of the perceived threat variable is quite high regardless of the gender category. Research by sociologists such as Janet Mancini-Bilson and clinical psychologist Richard Majors (1993) has explored in detail the trials and experiences of Blacks males in American society, and their need to adopt strategic protective styles. According to their research it is normal for Black males to feel angry and alienated when they are they perceive they’ve been denied opportunities to 87 fulfill traditional gender roles because of experiences with racial prejudice and discrimination. While the statistical findings across gender are not dramatically large. it's likely that whatever differences exist are due to perceive gender role constraints. TABLE 21b. Gender by Perceived Threat, and Prejudice; 1997-1999 Gallup Race Relations Data Gender Prejudice Category Perceive Threat Towards Whites °/o N % N Male 77.6% 1 149 54.4% 1204 Female 70.7% 1853 49.1% 2003 Total 73.4% 3002 51 .1% 3207 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only Gender and Education While gender has about a 5% main effect on prejudice, there is a possibility that because the majority of Blacks in the sample are female, that the effects of education might actually be interacting with gender to produce strong education effects. Of the 3.266 respondents in the pooled data, 63% (N=2,044) were female, and 37% (N=1,222) were male. In fact, the ratio of women to men was no less than 1:6, 1:7, and 1:5 for each of the respective three years of data. Thus, any main effect with prejudice shown in the data may actually covary with gender. " Among the demographic variables, income and education appear to have the strongest and most important relationships with prejudice. Therefore, it is important to assess the extent to which the effects of education and income are confounded by gender. Figures 1 and 2 both show that, on average the relationship between education and prejudice, and income and prejudice are the same for both males and females. 88 85 --_-- ..- -..-. ----- --.-.3..- .33. 3.-- - . ----.--_- - 3 3.3...- -.-.--.- ,3-3.3-.333---------.3.--3 3233.-.... 3, 80 ” +Male i 75 - ' 0 " Female 1 70 d 66'6 64.03 63.7 1 65 ‘ 58.8 fit _. 1 60 ~ -0 1 - - - - - o - - - I a n n 1 55 r _ . -o- - ' 58.06 1 50 .. ‘ , . e ‘ 545 56.53 1 0 5 45 1 47.86 i 40 I I I 5 LT HS HS Grad Some College College Graduate FIGURE 1. Percentage of Prejudice Respondents By Education and Gender; 1997- 1999 Gallup Race Relations Social Audit 85 qM~--omv.~w-vv. .-w4- ‘m—m 7, ---.. .3. ---._,.-.. - . ..- ,-- .7. -S-.- ~\__.- -,- --3 m~-»v~.._. 2-‘ -,__3-3_‘ _.- -.2 Wt..3»..-_u 3“,-.. - »-,,_....-.--.wf..._. 80 '- —O—Male 75 " ' 0 " Female 68 69.3 69.6 70 " 65 1 60 ‘ 55 a 50 '1 45 ‘ o 43.8 40 4 r - , LT $25K $25K - LT $35K - LT $45K - LT $55K + $35K $45K $55K FIGURE 2. Percentage of Prejudice Respondents By Income and Gender; 1997- 1999 Gallup Race Relations Social Audit 89 First, Figure 1 shows that males are more prejudiced than females for all levels of education. This general pattern parallels the data in Table 21b, showing that males are slightly more prejudiced than females. The results show that there were no statistically significant interaction effects of gender and education on self-reported prejudice (F=1.28, df=3, sig. = n.s.). Next, Figure 2 shows that income is also basically free of the confounding effects of gender. The relationship between gender and prejudice was roughly the same for all levels of income. The interaction effect was not statistically significant (F=1.50, df=4, sig.=n.s.), although the largest gender difference in percentage across income was for the lowest income group, at roughly 10%. Demographic Profile of Blacks’ Prejudice The most noteworthy discoveries from the analysis of the background variables were the patterns indicating that higher educated and income Blacks are the groups most likely to be prejudiced. Employed, working age blacks, with higher education and income, and those who live in low black density suburban areas, all individually represent the most prejudiced demographic categories. This is sound support for the hypothesis that the Black middle and upper class may be the Black segment most disillusioned and alienated about the opportunity structure in the US, and the most prejudiced. Just as the group position model predicts, those Black demographics with the highest levels of negative context, also tend to perceive more threat, and self-report more prejudice. Moving forward from the individual demographic analysis, to the aggregate analysis, will show how the categories cluster together and form more familiar patterns. Rather than looking at the numbers en masse, I create distinct breaks based on the 90 percentile distributions to help further discriminate who is “most prejudiced” from who is least. Figure 3 below shows the demographic categories identified as “most prejudiced” based on the proportion of respondents who indicated they were prejudiced. Any demographic categories with 55% or more of their respondents indicating prejudice were placed in this group. The first three bars on the left side of the chart show that the top three categories of income -- $35K-S44K, $45K-$54K, and $55K+ -- have the largest percentages of persons who reported being prejudiced against Whites. These income categories are each above 60%, and exhibit the most variance from the other categories in the “most prejudiced” segment. The next six categories in the most prejudiced segment have approximately 1% of variance separating their percentages. The two categories of education -- college graduate and some college -- represent the upper two education groups for that variable. The two age categories -- 28-37 and 38-53 -- represent the middle two categories the age variable. Finally, those Blacks who are employed and living in suburban areas complete this segment. The most prejudiced segment is characterized by upper income, higher educated, employed, young middle to upper middle aged Blacks, and Blacks who reside in suburban areas. This generally represents what some would call the Black middle class, or more aptly the Black upper middle class. Figure 4 shows the categories of the “middle prejudiced segment.” There is considerably less variance in the percentages for the categories in this segment. The category with the largest percentage of prejudice respondents, Blacks who have an household income of $25-$34K, is separated by only 5.2% from the group with the 91 lowest percentage of prejudiced respondents, the combination of urban and rural residents (or non-suburban residents). '00 . ....-. .3..th ..... ---.3-.-. . .- --. 3... 33.33. 3.-.... .. "3...... .3.- 66.5 65.8 ..... _. 61-5 56.2 1 "1 ugh: '1: . o S k. ' 0 I i V I f U I ' I I $55K + $45K - $35K - College 28-37 yrs Employed Some 38-53 yrs Suburban $54K $44K Grad College Residents FIGURE 3. Most Prejudiced Segment; 1997-1999 Gallup Race Relations Social Audit 1 00 . .3..- 3.... -3.-- .3 -.......... ......... 333...“..- 54-9 54.8 54-4 51.7 51.4 50.7 49.7 49.7 p 1 ‘ ' ' r; fl 3. . . ”‘1; _- ,1 o I I I I I I. l i I $25K - 18-27 yrs Males Low-Med Urban Suburan & HS Grad Urban & $34K Black Residents Rural Rural Density Residents Residents FIGURE 4. Middle Prejudiced Segment; 1997-1999 Gallup Race Relations Social Audit This segment is characterized by lower income, lower educated, younger males. who live in either urban or the combination of suburban and rural areas. The middle prejudice segment is rounded out by those Blacks who live in the low to medium Black density areas. 92 Figure 5 identifies those demographics in the “least prejudice” segment. Blacks who appear to be at the extremes of the demographic categories characterize this segment. For example, older Black females, those with the lowest income and lowest education, unemployed persons, and those who live in areas with larger Black populations, describe the profile of this segment. The variance in percentages for this segment also mirrors the variance in the most prejudiced segment; at least ten percentage points separate the lowest and higher percentage value. 100 333-333-3333. «3.3.3.33--. 49.5 49.1 48.8 48.7 ‘P' . —1 177:1 0 ‘- I i ‘7 i I l I I i i. i I I T T T I | ‘7 I 7 High Black Females High Black High Black LT HS LT $25K 54+ yrs Unemployed Dcnsity- Density-Non- Density Education Urban Urban FIGURE 5. Least Prejudiced Segment; 1997-1999 Gallup Race Relations Social Audit Together, the three bar graphs show that important variation exists in prejudice for the demographic categories. The most prejudice groups appear to be employed, well- educated, higher income Blacks, while the least prejudice groups have lower income and education, and are unemployed. Another interesting trait separating the profiles of the most and least prejudiced groups is the fact that the most prejudice typically live in suburban areas, while the least prejudiced live in high Black density areas (both urban and non-urban). 93 The results provide continued support that statements regarding perceptions of racial alienation by researchers (Hochschild, 1995; Cose, 1993; Bobo, 2001) have some validity. The groups most likely to be in competition with Whites for political and social prerogatives characterize the most prejudiced segment. The desires for upward social and economic mobility have been key goals for Blacks since the civil rights era. In fact, many liberals and conservatives have considered economic equality to be the answer to the many issues related to race relations (Cose, 1993; Hochschild, 1995). The results characterize a potentially grim tale that shows that those most likely to be par with middle to upper class Whites are actually more likely to be prejudiced towards them. Yet, this is consistent with the expectations of the group position model, which states that those most in competition for social and economic resources (e.g., working, middle aged, more educated and middle income persons) are most likely to be prejudiced. Figure 6 provides a final look at the demographic segmentation results. The Role of Negative Context In the group position model, the notion of competition implies a stratified context, a sense that a group is either a winner or loser in the social order. For Blacks, the context is one characterized by perceptions of historical unfair treatment, and discrimination. The context also reminds Blacks that they are potentially still competing for equal status. This is better understood when hearing statements indicating that Blacks have to work “twice as bar ” as Whites in order to make it in society. Blacks understand that race, matters. They understand that although they have education, income, and housing opportunities they might still face racial unfairness at any moment (Feagin and Sikes, 94 1994). This is apparently most frustrating for the more educated Blacks of the middle and upper class. * High Black Density "' $25K to LT $35K * $35K - LT $45K * Females * 18-27 yrs "‘ $45K to LT $55K " 54 + yrs * Males * $55K + “ Less than HS * Medium to Low Black * 28-37 yrs Education Density * 38-53 yrs * Unemployed * Urban resident * College graduate * Less than $25K "' Suburban & Rural * Some college "' High Black density- combined "' Employed Urban * Urban & Rural combined * Suburban residents * HS Grad Least Prejudiced Middle Prejudiced Most Prejudiced (Bottom Third) (Middle Third) (Upper Third) X S 49.5% 49.6 S X S: 54.9 X Z 55% X = Percentage of Blacks Who are Prejudiced FIGURE 6. Demographic Segmentation Results 1997-1999 Gallup Race Relations Social Audit According to Sniderman and Piazza (1993; 2002) more schooling or formal education should breed tolerance. They argue that the more schooling people have, the greater the likelihood they will be exposed to values, such as tolerance and equality. They further reason that more educated individuals have a higher level of political sophistication, and are more likely to have an awareness of cultural norms. More educate persons are essentially in a better position to accept or reject notions that are inconsistent with democratic norms than less educated persons. If this is true then why are the more educated Blacks in the most prejudiced segment? It appears that middle class Blacks are most likely to be prejudiced for the very reasons that Sniderman and Piazza (1993, 2003) state; more educated persons are more aware of their context. More educated Blacks are 95 more perceptive of their negative environment, yet they respond with what appears to be a more defensive, or prejudiced posture. Since education and income are highly correlated, it comes as no surprise that higher income Blacks share the same views as higher educated Blacks. This is similarly true for Blacks who are employed and live in suburban areas. The fact appears to be that the most prejudiced demographic groups for Blacks are also those who are most likely to be perceptive of their racial context. It also appears those Blacks most perceptive of their context, also perceived the most negative context. Tables 22a and 22b show that when the most prejudiced and least prejudiced segments are compared, there are vast differences in their perceptions of their negative racial contexts. The columns represent the three sets of contextual measures: perceptions that Blacks are treated unfairly, perceptions that the respondents themselves have been treated unfairly due to their race (i.e., the respondent has experienced discrimination), and the perceptions that Blacks have limited opportunities in housing, employment, or education. The rows represent the demographic categories that make up the segments. The cell values are percentages of respondents who provide a negative response (e.g., did experience unfair treatment) for each of the listed demographic categories. The values in the next to last row of cells are the mean values for the column. They represent the descriptive values for each negative context in the segment. For example, the percentage of respondents who are prejudiced for the nine demographic categories that make up the most prejudiced segment is 58.6%. The bottom row of the table contains the sample totals for all cases in the three years of data. This row is the same for both tables. Because of item non-response, some of the context measures have 96 more respondents than others, and the sample total of 3,207 represents the number of respondents for the prejudice item, which is the minimum N size for the context items. The most important observation from the tables is that those Black demographic groups in the most prejudice segment perceive a more negative context in all three of the measures. On average, the most prejudiced demographic groups are more likely to perceive Blacks are treated unfairly, they have experienced discrimination, and that Blacks have limited socio—economic opportunities relative to Whites. The differences in the segmentation percentages may not appear very meaningful, until one considers that the analysis is based on over three thousand cases, meaning one percentage point approximates 32 cases. TABLE 22a. Most Prejudiced Segment and Negative Context; 1997-1999 Gallup Race Relations Social Audit % of Blacks % Blacks % of Blacks Perceived % Treated Experienced Limited Demographic Category Prejudiced U nfairly Discrimination Opportunities N $55K or more 66.5 86.7 54.3 78.1 475 $45K to LT $55K 65.8 86.3 50.8 72.4 240 $35K to LT $45K 61.5 86.5 52.0 70.5 342 College Graduate 56.2 81.2 45.8 73.6 620 28-37 yrs 55.8 82.8 55.1 68.8 681 Employed Full or Part Time 55.7 82.8 52.1 69.6 2,298 Some College 55.6 84.9 50.5 71“ .1 782 38-53 yrs 55.3 81.9 48.5 72.7 1,035 Suburban residents 55.2 80.4 47.3 66.4 827 Segment Mean (Average) 58.6 83.7 50.7 71.5 Sample Total 51.1 79.5 45.9 67.6 N 2 3,207 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only It should come as no surprise that persons in the highest income category, represent the most prejudiced category, have the largest percentages of respondents 97 perceiving unfair group treatment, have the second largest (to 28-37 year olds) percentage who have experienced discrimination, and have the largest percentages of Blacks who perceive limits on their housing. employment, and educational opportunities. TABLE 22b. Least Prejudiced Segment and Negative Context; 1997-1999 Gallup Race Relations Social Audit % of Blacks % of Blacks % Blacks Experienced Perceived % Treated Discriminatio Limited Demographic Category Prejudiced Unfairly n Opportunities N High Black Density-Urban 49.5 80.5 47.4 70.9 1,098 Female 49.1 78.6 44.0 66.5 2,044 All Others (non-Urban and 48.8 78.6 49.7 69.5 899 High Black Density) High Black Density Area 48.7 79.6 47.3 70.3 1.671 LT High School education 45.2 76.1 45.1 61.3 747 LT $25K 41.3 77.4 42.6 65.4 1,295 54+ yrs 40.4 71.2 27.0 67.3 884 Unemployed 39.4 72.2 31.2 63.6 921 Segment Mean (Average) 45.7 76.8 41.8 66.9 Sample Total 51.1 79.5 45.9 67.6 N 2 3.207 Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only This finding fits neatly into the logic of the group position model because while this category of Blacks perceives the most negative context they are also represent the group who should, presumably, not have to deal with these perceived inequities. It also highlights discussions by both Bobo (2001) and Hochschild (1995) alerting us that those Blacks attempting to live the American dream may also be disillusioned by it. General Conclusions This analysis is informative in primarily two important ways. First, it is a rare segmentation study of Blacks’ prejudice that actually incorporates both a measure of prejudice, and utilizes a nationally representative sample of Blacks. The three years of 98 data allow for a rigorous test of the segments attitudes, and because the sample represents a cross section of Black Americans, the results have stronger external validity than previous studies. Second, the demographic categories found to be most prejudiced follow a pattern that identified middle and upper classes Blacks as the most prejudice segments in Black America. Generally, those socio-economic categories that were most prejudiced also reported higher levels of perceived negative context. This highlights the utility of studying prejudice from the group position perspective, and supports past research about Black middle class cynicism. Future research can take note that those Blacks in higher income and education categories are most likely to be prejudiced. The next question of interest may be, “why is the same not true for Whites.” Recent studies by researchers, such as Sniderman and Piazza (2003) laud educated Whites for their ability to see through the inefficiency of prejudice, racism, and intolerance, and present more egalitarian attitudes. This does not appear to be true for Blacks. If we combine the findings of the Sniderman and Piazza (2002) study, with my own, we see a pattern that shows that Blacks are able to strategically separate perceptions of hard work and financial optimism from the racially negative environment in which they live. According to Sniderman and Piazza, more educated and higher income Blacks are less anemic and more optimistic about their financial future. Yet, the findings in this chapter show they also perceive a more negative context, a greater racial threat, and are more prejudiced towards Whites. Is it the case that Blacks actually go through a calculus as to the importance of race? I believe so. I believe that Blacks have actually accepted, but not discounted that race continues to be, and will always be a problem, yet realize that 99 it is not a complete barrier to political, social, and economic achievement. In a sentence, Blacks are cautious, and conscious dreamers, they know that the dream is tenuous and can easily turn into a nightmare if prejudice is still a part of their daily lives. 100 Chapter 5 SENSE OF GROUP POSITION, THREAT, AND BLACK PREJUDICE The literature on Whites’ prejudice dominates the racial attitudes area of political psychology. Few studies have concerned themselves with the negative attitudes of Blacks towards Whites. However, the practical need for studies of anti-white prejudice is clear. As American continues to show a divide between Blacks and Whites attitudes about racial relations (Bobo, 2001; Gallup Organization, 1999; Schuman et al., 1997), it becomes increasingly important to understand both groups’ antagonisms. In view of the paucity of studies on Blacks’ racial attitudes, I seek to examine factors associated with Blacks prejudice toward Whites. Research on Whites’ racial attitudes suggests that contexts with greater perceived threat from Blacks, produce higher levels of attitudinal hostility among the competing racial groups (Bobo and Hutchings, 1998; Giles and Evans, 1986; Giles and Hertz, 1996; Quillian, 1995; Taylor, 1998). These findings support the tenets of the power threat- model (Giles and Hertz, 1994; Taylor, 1998) in general, and the racial threat model (Key, 1949; Blalock, 1967; Tolbert and Grummel, 2003) in particular. Both models emphasize the role of competitive threat resulting from a change in macro level context. The macro level threat is typically implied, because the results often lead to a competitive social, political, or economic loss within the context. This research is predominantly characterized by geographic contexts where Whites are said to perceive an implicit level of threat from growing black populations (Evans and Giles, 1986; Giles and Evans, 1986; Giles and Hertz, 1994; Key, 1949; Taylor, 1998). 101 Group position theory provides one explanation for the relationships between context, threat, and prejudice (Blumer, 195 8). As it relates to racial groups, the theory can be summarized by a related set of perceptions about where racial groups ought to stand within a stratified racial order. These perceptions take place within an intergroup context. The context creates and conditions an environment for competition over resources. The potential loss of resources (e.g., social standing, jobs, housing, and privileges) produces feelings of threat and anxiety. The threat itself is subjective rather than objective because the fear is grounded in a perceived racialized social context, rather than real and material conditions (Blumer, 1958; Bobo, 1999). The sense of group position, by name, implies the integration of subjective perceptions of reality on the part of individuals. Rather than the need for perceptions to be accurate. the model simply states that members of one group “feel” that they are losing ground to members of another group (Bobo, 1999). The group position model is similar to power-threat models in that, competitive threat arises from changes in the context, which creates the condition for competition. For the power-threat model this means that out-group population size creates the threat. Researchers have found that large proportions of, or changes in, out-group size can affect political party affiliation (Giles and Hertz, 1994), policy attitudes (Giles and Evans, 1986; Oliver and Mendelberg, 1998; Taylor, 1998), perceptions of racial threat (Fossett and Kiecolt, 1989), and prejudice (Pettigrew, 1959; Quillian, 1996). For the group position model, contextual changes mean a shift in perceptions of the relative positioning of racial out-groups. For example, researchers have shown that Whites support egalitarian principals more than policies designed to create equality because the policies themselves 102 are threatening (Bobo, 1983; Bobo and Klugel. 1993). It is the individual attitudes and reactions about the outcomes of the policy that create the racially contextualized threat. Equal outcomes indicates that some members of the majority groups will obviously loose out to members of the minority group, and thus. the threat is implied from the proposed implementation of the policy. One of the key distinctions between the development of the group position model and other models of prejudice based on threat. is the weighting of the subjective positioning of racial groups in society. Group position proponents view context as more than a matter of sheer group size and numbers (Bobo, 1999). The group position model maintains that context is comprised of racialized perceptions about a society’s opportunity structure (Bobo, 1999; Blumer, 1958). Jobs, housing, education, political party membership, and other rights and privileges are assumed to be guaranteed to dominant group members at a certain level. It is the dominant group’s beliefs about their collective privileges and status that drive the need to be protective of such prerogatives. When dominant group members perceive these entitled rights and privileges to be threatened, they develop “fears, apprehensions, resentrnents, angers, and bitterness, which become fused into a general feeling of prejudice against subordinate racial people” (Bobo, 1999; Blumer, 1958). In the power — threat models, this component is virtually implied, rather than measured. While numerous studies have documented the contextual effects for Whites (e.g., Giles and Hertz, 1994; Key, 1949; Oliver and Mendelberg, 2000; Quillian, 1995; Taylor, 1998), very few studies have examined the relationship between context and Blacks’ perceptions of threat (Bobo and Hutchings, 1996; Livingston, 2002). Research has 103 documented that perceptions of threat are associated with Blacks’ negative attitudes towards Whites. For example, Blacks’, who feel more alienated from white society, are more also likely to perceive social and economic threats from Whites (Bobo and Hutchings, 1996). In the area of political tolerance, research shows that Blacks are willing to express intolerance, limiting the freedoms of groups, such as the Klu Klux Klan, who pose greater threats to Blacks (Davis, 1995). Also, Blacks who live in areas that exhibit less tolerance, and stronger anti-black sentiment, tend to be more intolerant, and associate with others who are less tolerant (Gibson, 1992; 1995). The relevant evidence from each of these findings implies that context is potentially just as important for Blacks, as it is for Whites, and that perceived negative contexts and threat are important indicators of Blacks’ prejudice. Theoretical Perspective: Sense of Group Position and Racial Threat Following the tenets of the group position model, I propose that Blacks’ negative racial attitudes are primarily a function of racial threat, which is in large part, shaped by the perceived racial context of Blacks. The racial context signifies the sense of group position, and creates an environment for competitive threat for Blacks. Figure 7 provides a visual conceptual framework for this proposal. The model shows that prejudice is a function of contextually conditioned threat. In this case, the perception that Whites are also prejudiced is the more important determinant of Blacks’ prejudice. This awareness is primarily driven by a negative racial context that provides cues to Blacks about their position in society. I measure perceived negative context as the aggregate of recent experiences with discrimination, perceptions of unfair group treatment, and perceptions of limits on socio-economic opportunities. 104 Perceived threat is measured by the belief that Whites in the community are prejudiced. Black prejudice is measured by Blacks’ self-reported prejudice “against Whites.” The plus signs above the arrow indicate the hypothesized direction of the relationships. + + Perceived Negative Racial Context Perceived Racial Threat Blacks Prejudice 0 Recent Experiences with Perceptions of Whites’ "’ Blacks Self-Reported Discrimination Prejudice in the Community Prejudice Against 0 Perceptions of Unfair Group Whites Treatment 0 Perceptions of Limited Socio- economic Opportunities FIGURE 7. Conceptual Model of Blacks’ Prejudice towards Whites I incorporated three years of the GRRSA data to examine the relationships among context, threat, and prejudice. I hypothesize that Blacks who perceive a more negative racial context, are more likely to perceive a threat from Whites in the community, and are more likely to be prejudiced. Hypotheses H1: Blacks, who perceive a more negative context, are more likely to perceive a threat from Whites. H2: Blacks, who perceive a threat from Whites, are more likely to be prejudiced. The Relationship between Threat and Context Blacks, who perceive their racial group, or themselves, in a more negative context, should be more likely to feel threatened by traditional providers of racial hostility. A perceived negative context implies an understanding that in the social environment, race matters, and is a divisive factor in inter-group relations (Bobo, 1999). Assuming that being considered racially different and socially less important matter, 105 Blacks are more likely to feel threatened by any group that favorably promotes these ideas, or any attitudes or statements that encourage them. Moreover, the threatening group doesn’t need to show explicit racial animus, it only needs to be perceived from the threatened group (Bobo, 1999). Threatened groups use their perceptions of the racial context to evaluate the threat: the more negative the context. the greater the likelihood of threat. Table 24 provides an illustration of the relationships between racial threat and negative racial context. The correlation statistics show the relationships between each of the negative context traits, and perceptions that local Whites are prejudiced (i.e., perceived threat). The last row of the table shows the relationship between the overall index of negative context and threat. Specific wordings for the items, and the index can be found in Chapter 3 of this study, and Appendix A. The year to year fluctuation in statistics brings to light that there may be factors that occur in each year that differently impact the various context items. For example, in 1997, experiences with discrimination had the strongest relationship with threat, however, by 1999, it showed the weakest. Alternatively, perceived limits on opportunities showed the weakest relationship to threat in 1997, yet by 1999, it had the strongest. In addition to the movement of variable importance, each of the three variables substantially changed its coefficients’ value. Perceptions of group unfair treatment went from .180 to .289, while experiences with discrimination went from .240 to .167. To smooth out these variations, the data were pooled across the three years, and the correlations calculated. The last column shows the pooled data analysis of each context and perceived threat. For all three years, and in the pooled data, the context 106 index shows a stronger relationship with perceptions of Whites prejudice than any of the individual items. About 8% of the variance in perceptions of Whites’ prejudice can be accounted for by the index of negative context. While this is a relatively small amount of variance, it is statistically significant at the 99% confidence level. It is also noteworthy that when context is correlated with the threat measure in its original eleven point scale, the explained variance increases by only 2%. Thus, the relationship is relatively stable even when considering the change in measurement level. The context index will be used for the remained of the analysis, instead of the individual items. TABLE 23. Correlations between Negative Racial Contexts and Perceived Racial Threat; 1997-1999 Gallup Race Relations Social Audit Racial Threat: Perceptions of Whites' Prejudice Towards Blacks in the Community 1997 1998 1999 Pooled Perceive Blacks Treated Unfairly As a Group a .180“ -180** .289“ .214“: Ragzgriseglfiit‘glilma .240” .166“ .167** .191** Perceived that Blacks Have Limited Opportunities a Negative Context Index” .278** .293** .331" .294M .177** .259** .291** .234” a . . . . b . . . . . . Note: Statistics are tetrachorlc correlations; Statistics are eta values; ** Correlation ls srgnlficant at the 0.01 level (2-tailed). Observing only the relationship between the context and threat, Table 24 shows the 40% point difference in prejudiced responses across levels of negative context. The more Blacks perceive they are living in a negative racial environment, the more likely they are to believe Whites in their areas are prejudiced. The results also show that perceiving a negative context matters more towards perceptions of threat, than perceiving no negative context. Note that the differences between perceptions of threat and no 107 threat, for each of the context levels, gets progressively larger as you increase the negativity of the context. Of those Blacks who do not believe that the context is racially negative, 57% do not believe Whites in their areas are prejudice. The differences between perceptions of threat and no threat, for each of the context levels, gets progressively larger as you increase the negativity of the context. TABLE 24. Percentage of Blacks who Perceive Whites as Prejudiced For Levels of Context; 1997-1999 Gallup Race Relations Social Audits Perceive No Perceive Total Threat Threat — N % N % N % No Negative Context 175 53.7% 151 46.3% 326 100.0% Less Negative Context 216 37.9% 354 62.1% 570 100.0% Negative Context 267 26.0% 760 74.0% 1027 100.0% Most Negative Context 142 13.2% 937 86.8% 1079 100.0% Total 800 26.6% 2202 73.4% 3002 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only Of those Blacks who do not believe that the context is racially negative, 53.7% do not believe Whites in their areas are prejudice. This is in comparison to the 86.8% of Blacks in the most negative context, who believe Whites in their area are prejudiced. The implication is that as groups’ own perceptions of negativity increase, so do their perceptions of threat, but at an increasing rate. That is, the threat is becoming more crystallized in the minds of the group as they perceive their racial position to be different. This can be gleaned from the percentage difference in each level of context across threat. Blacks’ Prejudice Blumer (1958) believed prejudice to be a defensive reaction to the challenging of a group’s position in the social-economic racial order. The group position, and competition to maintain it relative to others, set the context for threat and hostility. The 108 greater the perceived threat to the group’s position, the more likely prejudice would be targeted at the threatening group. At the time of his writings, his concern was Whites’ prejudice towards Blacks. In essence, Whites were threatened by the increasing competition with Blacks for socially important prerogatives. Bobo (Bobo, 1999; Bobo and Hutchings, 1996) elaborated and extended on Blumer’s thoughts, studying the competitive threats arising from a perception of negative context. Their research found that the theoretical relationships between group position (context), threat, and prejudiced might be extended to minority groups. In their study, Bobo and Hutchings (1996) viewed context as a measure of alienation ranging from complete enfranchisement and entitlement, to complete disenfranchisement and grievance. They found that context, characterized by alienation, was consistently the most important predictor of perceptions of competitive threat. Among Whites, Blacks, Asians, and Latinos, Blacks reported the highest levels of alienation. Also, Blacks -- closely followed by Latinos -- were most likely to view other ethnic groups as threatening. However, Bobo and Hutchings (1996) stopped short of addressing the impact of this context on prejudice. Their analysis viewed prejudice as one component of perceived threat. To the authors’ credit, their goal was focused more on perceptions of competition and threat rather than prejudice per se. The goal of this analysis is to take the Bobo and Hutchings (1996) extensions an additional step, explicitly studying the effect of threat on Blacks’ prejudice. In this analysis, at the individual level prejudice amounts to a Black respondent self-reporting they are prejudiced, and at the aggregate level, prejudice is the pr0portion of Blacks who indicated that they are prejudice towards Whites. 109 Respondents who indicated any level of prejudiced (1-10) were coded one, and all other non-missing values were coded zero. Table 25 shows the distribution of prejudice responses. TABLE 25. Percentage of Blacks indicating they are Prejudiced; 1997-1999 Gallup Race Relations Social Audits 1997 1998 1999 Total N % N % N % N % Not atAll Prejudice (0) 571 45.9% 498 51.2% 499 50.4% 1568 48.9% Prejudice(l-10) 674 54.1% 474 48.8% 491 49.6% 1639 51.1% Total 1245 100.0% 972 100.0% 990 100.0% 3207 100.0% Note: N sizes and percentages are based on unweighted counts; Data are for Black respondents only In two of three years, a slight majority of respondents indicated that they were not at all prejudiced against Whites. Yet, due mostly in part to higher numbers in 1997, the pooled analysis, shows a slightly greater than 2% point majority indicate they are prejudiced. The table provides a picture that Blacks are almost equally divided in whether or not they are prejudice towards Whites. Since there are virtually no national studies of Blacks prejudice towards Whites, these figures should be taken with some caution. It is possible that Blacks are under or even over reporting their prejudice. This may occur for a number of reasons, one of which might be due to the race of the interviewer (ROI). An analysis of ROI effects on prejudice shows that Blacks are more likely to say they are prejudiced when talking to a black interviewer, than they are when talking to a white interviewer. Fifty-six percent of Blacks indicated they were prejudiced when talking to a Black interviewer, compared to 45% of Blacks interviewed by Whites. This, roughly, 11% difference was large enough to be statistically significant (t = 3.96, df = 1,5 84, p < .001 ), however, the eta square 110 statistic was .01, indicating that only 1% of the variance in prejudice could be accounted for by R01. As a result, race of interviewer will be not be included in further analysis in this Chapter, but it’s potential effects are duly considered. Prejudice and Context Figure 8 shows the bivariate relationship between prejudice and context. Blacks in the most negative context category have the largest percentage of prejudiced persons, and those who indicate no perceived negative context, have the smallest percentage. In terms of an eta squared statistic, 5% of the variance in prejudice can be accounted for by the categories of the context index. The hypothesized relationship holds true, the more negative the racial context, the more likely one is to report being prejudiced. Most Negative Context (N=l,121) More Negative (N=1,109) Less Negative (N=612) No Negative Context (N=365) 0 d‘ / e 0% 00,0- FIGURE 8. Percentage of Blacks Who are Prejudiced By Negative Context; 1997- 1999 Gallup Race Relations Social Audits Prejudice and Threat Figure 9 shows the gulf between those Blacks who perceive the threat of prejudice from those who do not. The approximately 54.2% difference amounts to an eta squared statistic of .230, indicating that 23% of the variance in prejudice can be 111 accounted for by perceptions of threat. This is a substantially stronger relationship than seen with context, and appears to support the expected nature of the relationships among the variables in the group position model. As hypothesize, those Blacks who believe Whites are prejudiced are more likely to report being prejudiced. Believe Whites are Prejudiced (N=2.181) Believe Whites are NOT Prejudiced (N=795) 12.4% 00 0‘0 /0 o a, 0% FIGURE 9 - Percentage of Blacks Who are Prejudiced By Perceived Threat; 1997- 1999 Gallup Race Relations Social Audits The Relationship between Context, Prejudice, and Threat According to the group position model, prejudice is contextually condition by a perception of threat. To test this idea, I held constant the level of context, and observed the relationship between threat and prejudice. The results can be seen below in Figure 10. Within the graphic, a number of interesting patterns emerge. First, across all levels of context, Blacks who perceive that Whites are prejudice are most likely to be prejudiced, while those Blacks who perceive no threat from Whites, are less prejudiced. Looking at the most negative context level, approximately 70% of Blacks who perceive the most negative context, and perceive the threat of prejudice from Whites, are 112 prejudiced. In comparison, Blacks who perceive no negative context, and no threat from Whites, are least likely to be prejudiced, at 10.9%. :| 20.4% (N=142) -t'*‘2::st.':.rirrreernxcft‘tfizlssziw-r.enumerate.-rrvzivrr‘a:14:.‘31munéifien‘vrrt‘t 70.69/13 (N=930) Most Negative Context I 13.3% (N=264) amtewnmvemmmmmummfiw 66 _O% (N = 7 5 5 ) More Negative - 7.5°/ N=2l4 Less Negative ,, __ ).. 60.6% (N=348) ‘ I 10.9% (N=l75) u i k 3‘ ' j‘ ‘ new?! a! .' r we"; .. “- No Negative Context 58.8% (N=148) T I . 0 J‘ / [II No Perceived Threat 0? 0% 003_ 0 I Perceived Threat FIGURE 10. Proportion of Respondents Who Are Prejudiced By Context and Threat; 1997-1999 Gallup Race Relations Social Audits A second interesting relationship gleaned from Figure 8 is that threat and context have no statistical interaction effect on prejudice. The relationship between threat and prejudice is the same for all levels of context. The differences in prejudice between those perceiving threat and those not perceiving threat for the lowest context is approximately 48%, while same groupings have a difference of 50% in the most negative context group. When compared to the bivariate relationships between context and prejudice, and threat and prejudice, this second finding supports the notion that threat helps to moderate the relationship between context and prejudice. A confirmation of this relationship is evident when a full modeled two-way analysis of variance (ANOVA) is run on the data. With context and threat, and their 113 interaction, as the independent variables, and percent prejudice as the dependent variable, the full model shows that context (eta sq = .007, F = 6.91 , df = 3, p S .000) and threat (eta sq = .181, F = 655.86, df = 1, p S .000) have statistically significant main effects, yet the interaction term is not statistically significant (eta sq = .000, F =.327. df = 3, p = n.s.). Between context and threat, threat is clearly the more powerful predictor of prejudice. Further evidence of this finding can be seen when comparing the most negative and no negative context categories. Those Blacks in the no negative context category, who perceive a threat, are more likely to be prejudiced (58.8%), than Blacks who have the most negative context, yet perceive no threat (20.4%). This implies a reciprocal or reactionary form of prejudice, similar to that described in research looking into Blacks racial attitudes (Livingston, 2000; Monteith and Spicer, 2000). Blacks’ prejudice is primarily associated with beliefs that Whites are prejudice, rather than how they perceive they are treated because of their race. Multivariate Models of Prejudice To test the strength of the relationships among context, threat, and prejudice, I ran a series of multivariate logistic regression models observing the impacts of context and threat on prejudice, controlling for demographic factors. The models included a baseline model with just demographics, individual models for context and then threat, each with demographics, and then a full model with all demographics, context, threat, and a context-threat interaction variable. Table 26 shows the results of the multivariate analyses. Each column indicates a separate analysis run for the variables on the row side. The values in the table are 114 adjusted odds ratios. indicating the effect of a one unit increase in the values of each independent variable on prejudice, when controlling for the demographic variables. At the bottom of the table are model summary statistics that indicate the strength and accuracy of the models, as well as the number of cases included in each model. TABLE 26. Logistic Regression Odds Ratios for Regression of Prejudice Response on Selected Demographics, Context, and Perceived Threat; 1997-1999 Gallup Race Relations Social Audits Baseline Model Model Model Model Predictors Model 1 2 3 4 Constant 302*" .148*** .060*** .042*** 054*" Gender (Males= 1) 1.05 .943 .946 .930 .928 Age 895* .91 1 .958 .966 .966 Education 139*“ 1.34*** 1.22M 1.20" 1.20" Household Income 1.14*** l.18*** 1.14“ 1.14" 1.14" Urban (Urban-=1) 1.167 1.16 1.40* 1.379* 1.38* Suburban (Suburban=1) 1.221 1.22 1.30 1.29 1.30 High Black Density (HBD=1) 1.064 1.07 1.14 1.15 1.15 Work Status (Employed=1) 1.032 .987 1.01 .979 .978 Context - 1.498*** - 128*" 1.1 l Threat - - 954*" 854*“ 623*" _‘ Threat x Context - - - - 1.19 Number of Cases 1720 1720 1641 1641 1641 Pseudo R Square .086 .127 .280 .290 .291 Percentage Correctly Classified 60.3% 62.1% 70.4% 70.6% 70.6 Note: The cell values are adjusted odds ratios estimated fi'om logistic regression Analysis; the data are unweighted, and on Black respondents are included. The percentage correctly classified are based on a .50 probability that the Dependent variable = Prejudice. 115 The baseline model shows those younger Blacks with higher incomes, and more education, are most likely to be prejudice. The constant, which represents, rural, unemployed females, was also statistically significant. The most interesting result is that more education tends to lead to a greater likelihood of prejudice. This finding is consistent with earlier findings in Chapter 4, as well as other studies that found higher income Blacks to perceive more threat (Bobo and Hutchings, 1996), or perceive a more negative racial environment (Sigelman and Welch, 1991). However. this is one of the few instances where higher income is empirically linked to more prejudice. For example, as a black person increases their education from a high school diploma, to “some college”, their odds of prejudice are 1.39 times greater. The exact opposite effect has been hypothesized and found for Whites (Bobo and Hutchings, 1996; Kinder and Sanders, 1996; Sniderman and Piazza, 2002). The results of the baseline model indicate that education, income, and age are the primary demographics to consider when viewing the remaining models. Model 1 shows the impact of context, controlling for the demographic variables. Education and income remain statistically significant, while age drops below the significance threshold of 95% confidence. Perceived negative context is statistically significant at the 99.9% confidence level. As a black respondent increases their perception of a negative context, they increase their odds of being prejudiced by 1.498. This effect is greater than all of the other individual demographic variables in the model. The context variable, alone, increases the r-square value by 4%, from the baseline pseudo R-square value. 116 Model 2 includes demographics and the perception of threat variable. By including the threat variable, the pseudo R-square value increases 20% from the baseline model value. In addition, the threat variable dominates all other variables in effect size; if a Black respondent perceives Whites to be prejudiced, their odds of prejudice are 9.54 times greater than Blacks who perceive no threat. This is a substantially large effect, especially when controlling for other demographics. The effects of education and income are still statistically significant, however, being from an urban environment (urbanicity) becomes, not only significant, but the strongest demographic predictor variable. For respondents from urban areas, the odds of prejudice are 1.40 times higher than for respondents from suburban and rural areas. Finally, as a test of model fit, the variables in this model correctly classify over two-thirds of the respondents, which indicate its strength as a predictive model. Model 3 incorporates the effects of both context and threat, along with the demographic items. Both negative context and threat are significant, but threat, although slightly less, once again, dominates the relationships among the predictor variables and prejudice. Urbanicity’s effect size also decreases slightly; however, it is still larger than other demographics factors, and the negative context measure. This might indicate that threat and urbanicity operate together to form an important influence on prejudice, such that those Blacks, who live in urban areas and sense Whites to be prejudice, are also most likely to be prejudiced. However, ad-hoc analysis indicates that this is not the case. When an interaction term for threat and urbanicity (not shown in the table) is added to model 3, both urbanicity and the interaction term are no longer statistically significant. That is, Black respondents living in an urban environment, who perceive Whites to be 117 prejudice, are no different in their reporting of prejudice, than Blacks from suburban and rural environments who perceive no threats. Also, when an interaction term for context and urbanicity (not shown in table) is included in model 3, context and the interaction term, are no longer statistically significant, but urbanicity remains barely significant. This indicates that when threat is considered, context tends to have less an impact on prejudice, when controlling for urbanicity. This is a relevant pattern that will be explored in the discussion that follows. The fourth and final model, considers all demographics, context, threat, and a context—threat interaction term. The goal of this final model is to test the strength of the relationships between context and threat, and prejudice. The model results in the final column show that when controlling for context and threat independently, those who perceive Whites to be prejudiced, regardless of level of negative context, are no more prejudiced than those who do not perceive Whites to be prejudiced. Neither the interaction term, nor the negative context, terms are statistically significant. However, urbanicity, education, and income remain statistically significant, and the threat variable remains the dominant predictor. This final model predicts almost 21% more variance than the baseline model. The stability of the demographic predictors and the threat variable indicate a clearly defined model for understanding Blacks prejudice. It appears, other things being equal, that Blacks, who perceive a threat, have higher income, higher education, or live in urban environments, are more likely to be prejudiced. The strength of the threat variable in the models indicates its importance in understanding the nature of Blacks prejudices. and prejudice in general. 118 General Conclusions The research on Blacks prejudice towards Whites has been handicapped by both a lack of sheer numbers, and low priority (Schuman, Steeh, and Bobo, 1985; Schuman et al., 1997; Shelton, 2001; Sigelman and Welch, 1991). This study attempts to contribute to the understanding of Blacks’ prejudice, by testing the stated outcomes of the group position model (Blumer, 1958; Bobo, 1999). The group position model asserts that prejudice is a defensive reaction to a contextually defined threat. In this study the context was the perception that Blacks were treated more negative, or generally, different than Whites. This context can help explain why it matters to Blacks, that Whites in the area are prejudiced. Whites’ prejudice constitutes a threat to the relative status of Blacks. It signals a desire, on the part of Blacks, to protect against going back to the real and perceived threatening ways of the past. The results show that perceived threat, as conditioned by context, is a significant predictor of Blacks’ prejudice. While both context and threat have significant main effects, there is no interaction effect. Blacks are more likely to be prejudiced, if they perceive Whites, around them, as being prejudiced; this is true regardless of the level of context. This relationship also holds true when controlling for demographic variables. The role of threat is so important that even when Blacks perceive no negative context, the data show that still have a relatively high probability of being prejudiced if they perceive a threat a threat from Whites. Education, income, and urbanicity were also significant predictors of prejudice. More educated and higher income Blacks were more likely to be prejudiced 119 towards Whites. Also, when controlling for demographics, and the density levels of the Black population, Blacks who lived in urban areas, were also more likely to be prejudiced. These relationships might imply that those Blacks who are in the middle, and upper middle class are more likely to be prejudice, because they are more likely to view themselves in competition with Whites. This notion is consistent with the group position model, and other models of context effects, such as the power-threat model (Giles and Hertz, 1994; Taylor, 1998; Tolbert and Grummel, 2003). Similarly, since research has found that Blacks’ who interact with Whites for extended periods are the least likely to believe that whites hold positive images (i.e., stereotypes) of Blacks (Sigelman and Tuch, 1997), they may also be more susceptible to perceiving the potential for a racial threat from Whites. The impact of education on Blacks negative attitudes towards Whites has been a consistently unexpected relationship in the study of Blacks racial attitudes. Considering that the majority of studies on Whites’ racial attitudes show a negative correlation between education and racial hostility (Sigelman and Welch, 1991; Sniderman and Piazza, 2002), it is often surprising to see the opposite effect with Blacks. However, studies have found that Blacks with higher education were more likely to feel higher levels of alienation*(Bobo and Hutchings, 1996), and exhibit more in-group preference (Evans and Giles, 1986). The results from this study support the argument that as Blacks climb the education, and income ladder, they tend to become more aware of the negative racial environment they face (Cose, 1993; Feagin and Sikes, 1994; Sigelman and Welch, 1991), and possibly develop strategies, such as out-group prejudice, to deal with a potentially threatening racial environment (Branscombe and Ellemers, 1998; Feagin, 1991). This study concludes that in order to understand Blacks’ prejudice it is important to study their perceptions of where Blacks stand, politically, socially, and economically, relative to Whites, as well as the extent to which they perceive prejudice or other racial negativity from Whites. The results of this study also imply that models that account for Whites prejudice may also account for Blacks prejudice, although they may be contextually framed differently. Originally developed based on Whites’ prejudice towards Blacks, the group position model now holds potential value in understanding, not only the prejudice of Blacks. but the prejudice of other minority groups in competition with a majority group. 121 Chapter 6 HOW BLACK ATTITUDES TOWARDS WHITES VARY WITH RACIAL COMPOSITION How do Blacks’ population sizes impact their attitudes towards Whites? A number of studies have been conducted that show the impact of minority presence on Whites’ racial attitudes (Giles and Hertz, 1994; Evans and Giles, 1986; Oliver and Mendelberg, 2000; Pettigrew, 1959; Quillian, 1996; Taylor, 1998), but there are very few (c.f., Bobo and Hutchings, 1996; Evans and Giles, 1986), studies of context effects on Black prejudice attitudes. The general consensus from the studies of White samples is that Whites’ racial hostility and prejudice toward out-groups rises in relation to the size of the surrounding Black population. For Black samples, there are more assumptions about context effects on Blacks (Evans and Giles, 1986), than empirical evidence. This research is concerned with the question of whether or not Blacks’ population size is associated with their prejudice towards Whites, and if so, what are the strength, nature, and direction of the relationship. I am also interested in what variables help account for the relationship? Towards these ends, I test four models that make predictions about the relationship between Black population size and prejudice. The notion of competitive threat is the theoretical basis for the majority of these findings. The very fact that the Blacks had a larger proportion of the population, or had increased it’s population, is considered a strong enough factor so as to encourage white majorities to be on guard that Blacks are likely impending on their prerogatives (Evans and Giles, 1986; Giles and Hertz, 1994; Oliver and Mendelberg, 2000; Quillian, 1995; Taylor, 1998). Aside from questions about other possible macro level variables that may 122 also be important, and arguable differences in the importance of geographic level at which data are collected (Oliver and Mendelberg, 2000; Qullian, 1996), theories of social-racial context effects have pretty much reached consensus: “minority visibility worsens Whites’ attitudes (Taylor, 1998).” There are a number of competing theories regarding the impacts of increases in out-group population. Increased contact -- particularly close, equal status and sustained contact -- has been shown to reduce hostility and prejudice, and increase tolerance (Powers and Ellison, 1995). The contact-hypothesis (Allport, 1954; Sigelman and Welch, 1993) has been tested over the past several decades and found that racial antipathy is lower among Whites who maintain closer or more frequent contact with Blacks (Aberbach and Walker, 1973; Meer and Freedman, 1966; Robinson, 1980; Sigelman and Welch, 1993). For Blacks however, the results are mixed. There are instances when tests of the contact hypothesis revealed increased likelihood of prejudice (Ford, 1973; Robinson and Preston, 1976), or showed no statistical effects altogether (Brown and Albee, 1966; Sigelman and Welch, 1993); yet, the contact hypothesis continues to receive attention as a potential factor in racial attitudes (Powers and Ellison, 1995; Sigelman et al., 1996; Taylor, 1998). The contact hypothesis is that areas where the Black population‘is smaller, Blacks should have greater contact with Whites and lower levels of prejudice because they are in more direct contact with Whites, which may increase tolerance and reduce stereotypes. Another competing model is the ethnic community model (Gutterbock and London, 1983; Olsen, 1970; Patchen, 1982). This model suggests that in areas of higher minority concentrations, minority membership generates enhanced cohesion and 123 perceived in-group power. Research shows that an increase in Black population size for defined geographic areas. can serve to generate higher levels of ethnocentrism among in- group members (Ellison and Gay, 1989), leading to a heightened awareness of group differences, and an potential increases in prejudice expressed toward Whites (Davis, 1973). The same finding exists in school settings and contexts. Patchen (1982) found that schools with larger Black populations tended to express greater hostility towards Whites within the same school. Up through the mid-19808, the ethnic community model was one of the few approaches that could be applied to Black samples to explain contextual effects. Because of the lack of other models to describe Blacks political attitudes in the context of population size, and empirical evidence arguing differently, researchers often assumed that increases in the Black population in given contexts, would produce greater levels of ethnocentrism and increased prejudice towards Whites (Evans and Giles, 1986). The theory appears plausible, but has been virtually untested in the study of Blacks’ prejudice towards Whites. As it relates to this study, if accurate, the ethnic community model should show that Blacks in areas with larger proportions of Blacks should be more likely to be both prejudiced and ethnocentric: prefer their own race. The group position and power-threat hypotheses (Blumer, 1958; Bobo, 1999; Giles and Evans, 1984; Oliver and Mendelberg, 2000; Taylor, 1998) predict that higher levels of out-group population, and thus, lower levels of the in-group population, should be associated with more negative attitudes. The rationale for this outcome is that competition is more likely to occur in areas where the out-group is larger; yet the impact of population related contextual effects has rarely been tested on Black samples. Aside 124 from important studies conducted in multi-ethnic settings, comparing multiple ethnic groups (Bobo and Hutchings, 1996; Oliver. 2001; Tolbert and Grummel, 2003), theoretical and empirical relationships between racial contexts and prejudice attitudes, among Blacks, requires further investigation. To review, the contact-hypothesis (Alport, 1954; Sigelman and Welch, 1993) predicts that more interracial contact reduces negative racial attitudes. The contact hypothesis has been applied to Black samples on several occasions and has rarely been found to reduce Blacks negative attitudes (Sigelman and Welch, 1993). If interracial contact matters, then those Blacks with more contact with Whites, when controlling for population size, should be less likely to be prejudiced. The ethnic community model predicts that Blacks in areas with larger Black populations are more likely to be prejudiced. This theoretical outcome differs from the threat and contact models in that larger in-group populations are believe to bring about increased perceptions of in-group power, which result in negative attitudes towards Whites (Evans and Giles, 1986). I will analyze the relationship between context and prejudice with a slight twist. Traditionally, social context is measured as a macro level measure of out-group population size. However, with the current sample of Blacks, I will incorporate a measure of in-group population size. This approach has been attempted before, and no relationship between Black population size and perceptions of relative power were found (Evans and Giles, 1986). However, the study had a relatively small sample of Blacks (N=97) and was conducted using county level census data. This study will incorporate a much larger sample, and deal with multiple measure of context. 125 There has also been an assumption that Whites have both a sense of the impending out-group population size or increase in size, and as a result of the propinquity, had a real sense of threat (Giles and Hertz, 1994). There was always a possibility that the racial out-group size was ambiguous to Whites, and thus, the threat was assumed to be present based solely on out-group population size (Glazer, 2003). While more recent studies have seemingly dealt with the validity of threat (Bobo and Hutchings, 1996), there are still questions about the perceived knowledge of out- group population size, both real and perceived. Studies have shown that both Whites and Blacks are likely to make incorrect guess about the population size of Blacks (Nadeau, Niemi, and Levin, 1993; Sigelman and Niemi, 2001). Whites or Blacks may under or over estimate the population of Blacks in their area. However, if such estimation is not considered, there is uncertainty about a group’s awareness or acceptance of the threat (Bobo, 1999). For these reasons I consider the perceived size of the in-group, as well as the actual size. The association between threat and perceived racial group size has been documented previous research. Studies have found that feelings of racial threat are related to perceived population proportion (Nadeau, Niemi, and Levin, 1993). In general, people who‘feel threatened by minorities are mostly likely to overestimate their numbers (Pettigrew, Allport, and Barnett, 1958). This evidence combined with the other research proposing that prejudice arises from a perception that minority out-groups are encroaching on important values and material resources, makes the study of perceived minority population size critical. 126 This chapter’s analysis attempts to conduct a rough test of the effects of population size, by considering the real and perceived size of the in-group, rather than the out-group. In a traditional example, if Whites perceive Blacks to be a larger proportion of the population, or if Blacks actually comprise a larger proportion, then Whites should be more likely to perceive Blacks as a greater threat, and respond with hostility and prejudice. This is the basic tenet of the power-threat hypothesis (e.g., Giles and Hertz, 1994; Key, 1949; Taylor, 1998; Tolbert and Grummel, 2003). It is possible that in theory, similar to Whites, as Blacks also perceive the White population to be larger in a particular context, they are also more likely to perceive a competitive racial threat from Whites, and according to some theories, exhibit more prejudice towards Whites. Phrased in the context of this study, as Blacks perceive that they make up a smaller proportion of the population, they are more likely to perceive a racial threat from Whites, and thus exhibit a greater likelihood of prejudice. This is simply a restatement of the existing group position and power theories applied to Blacks. I test these potential effects using both actual and perceived in-group size on Blacks’ reported prejudice. Hypotheses HI: Power Threat Model. Blacks, in areas with lower Black populations are more likely to both perceive racial threat from Whites, and be prejudiced. Blacks who perceive the Black population in the United States to be lower, are more likely to be 3“ prejudiced. H2: Group Position Model. Blacks, who are more likely to perceive their racial group position as negative, are also more likely to perceive a threat. Those Blacks who perceive a threat and negative group position are more likely to be prejudiced. H3: Interracial contact Model. Blacks who have increased contact with Whites are less likely to be prejudiced. Ha: Ethnic Community Model. Blacks, in areas with higher Blacks populations, are more likely to show signs of in-group preference, and be prejudiced. 127 Data and Measures To test this proposition, I used one year of data, 1997, from the Gallup Race Relations Social Audit (GRRSA). This single year of data comes from a national over- sample of the Black population, conducted by Random Digit Dialing (RDD). The data are unique because they provide a Black over-sample of 1,243 self-identified Blacks, and their area codes. The data contain proportional Black racial compositions provided by Survey Sampling, Inc., and measure the percentage of Blacks in the respondent’s area code. The data in this year are also unique because they contain measures of Blacks’ self- reported prejudice, a question asking Blacks to estimate the proportion of the Black population, questions referring to interracial contact, measures of perceived group position, and perceptions of Whites’ prejudice. Because of sampling issues mentioned in the 1997 GRRSA methodology report (Gallup Organization, 1997), Gallup recommended that I apply weights to the 1997 data file to bring proportions back in line with target national averages. The sampling error for the Black over-sample is approximately plus or minus five percentage points (Gallup Organization, 1997). Blacks’ prejudice, perceived threat, ethnocentrism, and racial context (i.e., group position) were all measured using the same items and coding as stated in Chapter 3 of this study. Their exact question wording can also be found in Appendix A. Actual Black population size was measured as the percentage of Blacks in an area code. The use of area code as a measure of context must be interpreted cautiously because of the possible variance across such a large geographic boundary. Thus, a second measure of context is used that considers the perceived population size in the United States. 128 Perceived population size, was measured using a question that asked respondents to estimate the proportion of Blacks in the United States. The question was phrased as an open-ended query, and then coded. Thus, the respondent was not primed to select a certain population proportion. Specifically the question asked the following. it just your best guess, what percent of the United States population today would you say is Black? (Open ended and code)» The ordinal-coded categories were, less than 10% (Coded 1), 10% to less than 15%, 15% to less than 20%, 20% to less than 30%, 30% to less than 40%, 40% to less than 50%, and 50% or more (Coded 7). According to the group position and power-threat models, the greater the perceived proportion of Blacks, the lower the likelihood that Blacks should be threatened by Whites. The lower the likelihood of threat, the lower the likelihood that Blacks will report being prejudiced. Conversely, if Blacks perceive they are smaller in proportion, then they should be more likely to report being threatened and prejudiced. To restate the caveat with these contextual variables, actual Black population and perceived Black population are measured at two different levels with the former at the area code level, and the latter at the national level. However, if population size matters (Taylor, 1998), then both actual and perceived in-group size, should indicate a significant effect in the same direction. The primaquifference being the fact that one measures actual [area code] population size and the other measures perceived or estimated racial group size in the US. Interracial contact was measured by a series of questions starting with the following. “Next, here are some questions about the contact you have with Whites. Thinking about your close friends, are any of them White, or not?” 129 If a respondent say “yes” to this item, they were code =1. This item was added to the list of the remaining items, which had response categories of “almost all White”, “mostly White”, “about half White and half Black”, “mostly Black”, and “All Black”. For each item, if a respondent indicated their situational contact with Whites was “about half”, “mostly White”, or “all White”, then they were code 1. “Now, thinking about the CHURCH OR PLACE OF WORSHIP you attend, are the people in your place of worship. . .” “How about the NEIGHBORHOOD or area where you live in? Are the people in your neighborhood...” “How about the place where you WORK? Are the people you work with. . .” “How about any CLUBS OR ORGANIZATIONS you happen to belong to? Are the members...” “How about the school your OLDEST child attends? Are the students there...” The sum of the item codes was taken to create an index of interracial contact ranging from O to 6. The higher the index value, the more interracial contact Blacks had with Whites. In addition, several other independent variables were included in the analysis. Age is measured as the respondent’s age categorized as an ordinal measure with four levels: 18-27 (coded 1), 28-37, 38-53, and 54 (coded 4) or older. Gender was coded as 1 = male, and 0 = female. Education was coded as an ordinal measure: Less Than H: (Coded 1), HS Graduate, Some College, and College Graduate (Coded 4). Labor Force Status was coded as 1 = employed full or part-time, and 0 = not working. Household income was coded in five categories: Less Than $25K (Coded 1), $25K — Less Than $35K, $35K — Less Than $45K, $45K — Less Than $55K, and $55K or more (Coded 5). Based on their area codes, respondents were classified as living in rural, suburban, or 130 urban areas. Two dummy variables were created to measure living in an urban (1 = urban, 0 = all others), and suburban (1 = suburban, 0 = all others) areas. Finally, the race of the interviewer was included as a control for any potential social desirability on the part of the interviewer. White interviewers were coded as 1, and Black interviewers were coded as 0. Analysis Table 27 shows the correlational values between actual and perceived Black population size and other racial items. The table of correlations is not a square matrix, but is intended to show the main relationships expected in the hypotheses. The table shows an interesting pattern, with the actual Black population measure related only to contact with Whites, and perceived group position. There is no statistical relationship between self-reported prejudice and Black population size, nor perceived racial threat and actual Black population size. However, when the perceived size of the Black population is considered, there is a statistical relationship between both threat and prejudice, indicating that possibly perception of group size just as relevant, if not more relevant than actual out-group population size (Nadeau, Niemi, and Levine, 1993). The measure of group position is negatively, as well as statistically, associated with actual black population size, and perceptions of threat, but not with the perceived group size measure, contact, or respondent prejudice. The negative association with actual Black population may imply that in smaller Black populated areas, Blacks are more likely to perceive a negative racial environment due in-part to real or perceived negative experience, or racial competition. This finding actually fits a key component of the group position model in that the sense of group position sets the stage for perceived 131 competitive threat (Bobo and Htuchings, 1996; Blumer, 195 8), which then results in a potential for prejudice. This initial evidence provides some indication of the nature of the relationships between Black population size and racial perceptions and prejudice. A comforting result fro the data shows that in area codes with actual larger Black populations, Blacks have relatively less contact with Whites. TABLE 27. Correlations between Racial Population Context and Individual Level Racial Perceptions; 1997 Gallup Race Relations Social Audit Actual Perceived Contact Black Black Group with Population Population Position Whites Perceived Threat a -.050 -.126* .317** .057 Respondent is _ _ H Preju dice d a .018 .200 .079 .006 Ethnocentrism a .083 -.100 .076 -.208** Group Position b -.137* .034 - -.003 Contact with Whites b -234“ .048 -.003 - . . b Note: * p< .05, ** p< .01; a Cell values are Eta Statistics; Cell values are Pearson Correlation Coefficients However, interracial contact’s relationship with prejudice fails to meet the basic bivariate test of importance. In fact, interracial contact is not related to either perceived threat or Blacks self-Feponed prejudice, as evident by the extremely weak and non- statistically significant correlations among contact, threat and prejudice. Even when considering a lower level measure of contact, comparing Blacks who reported “about half”, “mostly white”, or “all white” in any of the settings, to those Blacks who reported otherwise, there were no statistical differences in likelihood of reporting being prejudiced (Mean proportional diff = .06, t=.567, df = 321, sig.=n.s.). Basically, there are no 132 differences in likelihood of prejudice across levels of interracial contact, indicating little support for the interracial contact model. Perceived Black population size is negatively associated with both threat and prejudice, indicating that Blacks, who perceive they make up a smaller proportion of the population, are more likely to perceive Whites are prejudice, and report being prejudiced themselves. This statistical significance and directionality supports the predictions of the power and group position hypotheses. The ethnic community model fails on the count of directionality, and association. The negative relationship between real or perceived Black population size and prejudice counters the expected outcomes of the ethnic community hypothesis. Both real and perceived Black populations are not statistically associated with ethnocentrism. Blacks who live in areas with both actual and perceived larger Black populations are no more likely to be prejudice or prefer working, living near, or marrying other Blacks. Thus, there is little support for the ethnic community model. In summary, the group position model and the power threat model are supported by the simple bivariate correlations, while the ethnic community and interracial contact models are not. The initial findings direct attention to the importance of in-group estimates of the Black population, and their relationship to self-reported prejudice. -.. The correlation between perceived racial population and prejudice indicates that about four percent of the variance in Blacks’ prejudice can be accounted for by their estimate of Black population size in the United States. A simple bar chart illustrates the almost negative linear relationship between the racial prejudice and perceived context. 133 Figure 11 shows that as the perception of Black’s population size increases, the proportion of prejudiced Blacks decreases. While compelling, it is very likely that there are some confounding effects in this relationship. For example, previous research shows that educated Blacks are more likely to perceive a competitive threat from both Whites and other racial groups (Bobo and Hutchings, 1996). It is equally possible that educated Blacks also have a better estimate of the true Black population, which borders around 14% - 15% percent of the US. population. Therefore, those Blacks at the lower end of the perceived Black population size measure may also be the most educated. 0/ l00.o '1 84.8% 0 50% .. 46.6 /o 0% FIGURE 11. Proportion of Prejudice Respondents by Perceived Group Size; 1997 Gallup Race Relations Social Audit To test the rigor of the initial findings, as well as assess any relationships that might confound the effects of actual and perceived Black population size, I analyzed the relationships between categorical background variables and population size. I also carried out tests to view the relationships between the background variables, and the predictor variables related to the various explanations of context effects. Table 28 provides the results of the analysis. The rows contain the categorical background variables, and the columns contain the prejudice variable, perceived threat, 134 interracial contact, and actual and perceived Black population. Significant relationships are indicated by boxes around the set of group measures of central tendency (means or proportions). The asterisks indicate which relationships are statistically significant. There are a number of key patterns that can be gleaned from the table. First, age, income, urbanicity, and work status are all associated with interracial contact. Based on the analysis, it appears that persons in the middle age group, those with household income over $25K, those who live in urban settings, and those who are unemployed are each more likely to have increased contacts with Whites. Three of the four background variables associated with contact are also statistically associated with prejudice. This provides some optimism for the interracial contact thesis. Controlling for the effects of other variables, it is possible that interracial contact may arise as a significant predictor of prejudice. Surprisingly, education, which is normally highly correlated with income, is not statistically associated with interracial contact, or any other of the key variables in the column portion of the table. This is particularly interesting because past studies have shown education to be correlated with racial alienation (Bobo and Hutchings, 1996) and political distrust of white institutions (Schuman and Hatchett, 1974). An independent analysis of the relationship between education and income in the data reveals a statistically significant chi-square association ()8 = 47.2, df=12, p<.001), but a non- statistical Gamma (G=-.091, t=-1.32, sig.=n.s.) and Somer’s D (D=-.062, t=-1.32, sig.=n.s.) statistics . This indicates that the categorical relationship may be mainly due to relatively large sample size. 135 TABLE 28. Central Tendency Measures for Key Race Items; 1997 Gallup Race Relations Social Audits Mean Mean % % Mean Actual Black Perceived Background Variable Prejudice Threat Contact Pop Black Pop Gender Male 61.5% 74.7% 2.62 16.1% 4.58 * Female 56.3% 81.7% 2.60 15.0% 5.46 Age 18-27 66.3% ** 77.1% **" 2.49 *** 15.0% 5.59 * 28-37 71.1% 89.9% 3.13 16.34% 5.07 38-53 55.2% 83.7% 2.83 13.8% 4.90 54+ 44.0% 57.2% 1.96 17. % 4.58 Education LT High School 58.3% 79.2% 2.60 16.1% 5.40 High School 61.4% 79.3% 2.79 15.8% 4.89 Some College 63.4% 79.9% 2.42 14.7% 4.57 College Graduate 40.5% 60.3% 2.24 12.7% 5.1 1 Household Income LT 25K 50.8% "‘ 69.4% ** 2.26 *** 15.6% 5.27 ** 25K to LT 35K 60.0% 86.5% 2.84 15.9% 5.17 35K to LT 45K 79.0% 83.9% 2.95 16.5% 5.31 45K to LT 55K 76.4% 88.7% 2.90 13.1% 4.85 55K or more 67.7% 89.7% 3.10 13.7% 4.04 Urbanicity Urban 58.9% 79.5% ** 2.40 ** 16.3% ** 4.99 Suburban 63.3% 81.7% 2.94 12.7% 5.02 Rural 47.1% 62.2% 2.73 17.9% 5.48 Work Status wig?” °’ pa” ' 62.0% * 63.1% *** 1.95 *** 17.0% 5.07 Unemployed 47.4% 82.2% 2.83 14.9% 5.08“" Perceived Group Position Low Negative Context 51.2% 56.8% *** 2.66 19.1% 5.03 Some Negative Context 50.1% 62.8% 2.50 15.7% 4.76 Negative Context 65.5% 78.9% 2.72 15.6% 5.27 Most Negative Context 59.6% 93.3% 2.56 13.9% 5.05 Note: Boxes indicate significant difference in mean scores for each variable within racial group categories; * pg .05, ** pg .01, 1'" pg .001. 136 Also, when controlling for a background variable such as urbanicity, the relationship disappears. There is a statistical association for suburban Blacks (x2 = 33.8, df=12, 155.001), but not for rural (x2 = 8.27, df=12, p= n.s.) or urban Blacks (x2 = 15.5. df=12, p= n.s.). It is possible that education simply has no relationship with the items or income, but its consistent non-relationship indicates a potential bias possibly existing in the data. Another interesting finding from the analysis centers on the differences between actual and perceived Black population size. The variables appear to be truly measuring different underlying group sizes (i.e., real versus perceived). The Pearson’s correlation between actual and perceived group size is -.021 (p. = n.s.). Additionally, while Blacks’ estimates of their population’s size are statistically associated with gender, age, and income, the actual Black population size is statistically associated with a single variable, urbanicity. Since both Black population size and urbanicity are based on the sample area, this relationship is highly plausible. It’s possible that given the wide populations covered by an area code, that the actual measure is a less valid indicator of racial context. Understanding a context’s boundaries is key to understanding how it affects individual level attitudes, and the many challenges to garnering the ideal measure have been noted in the literature on contextual effects (Giles and Hertz, 1994; Glazer, 2003; Oliver and Mendelberg, 2000; Taylor, 1998; Tolbert and Grummel, 2003). One implication from previous research on estimated innumeracy (Sigelman and Niemi, 1991), which applies to this study, is that it may be erroneous to rule out the presence of population based on threat established solely due to actual population size. At least, this appears true when studying the effects of Black population sizes. The 137 differences between actual and perceived Black population size may be grounded in the individual experiences of Blacks. Sigelman and Niemi (2001) found that Blacks who had more interaction with other Blacks were more likely to overestimate the population of Blacks. They also found the same to be true for Whites. Thus, it’s possible that higher contact with Whites influences the relationship between real and perceived Black population estimates and prejudice. The Gallup data show that when controlling for interracial contact, perceived Black population size is still statistically associated with prejudice (Rpan=-.l98, sig.=.001), while actual size is not (Rpan=-.083, sig.=n.s.). Therefore, controlling for contact with Whites, Blacks who perceive a higher Black population size, are less likely to be prejudice. This evidence points away from the ethnic community model. Income and Age are both related to four of the five key variables. Blacks with higher incomes are more likely to be prejudiced, more likely to perceive Whites are prejudiced, more likely to have higher incomes, and more likely to believe Blacks make up a smaller proportion of the US. population. For age, it appears that Blacks between the ages of 28 and 37 are the most likely drivers of Blacks perceptions of racial context. This age segment is most likely to be prejudiced and perceive Whites to be prejudiced. They are also the age segment with the highest level of interracial contact, and tend to perceive the Black population as higher than it is. Age and income represent the key demographic variables in this data set, because they help consistently identify those Blacks most likely to come into contact with Whites, perceive racial threat, report being prejudiced, and potentially overestimate the Black population. 138 The group position (i.e., negative racial context) measure is highly correlated with perceptions of racial threat, and nothing else. The measure is a composite of perceive unfairness for Blacks in general and the individual in particular, and also contains a component dealing with limited opportunities. The table shows that as perceptions of negative context increase, perceptions that Whites are prejudice (i.e., threat) tend to also increase. However, group position is unrelated to prejudice, contact, and population size. This may imply that Blacks, who perceive a more negative racial environment, do so across the board. Regardless of the level of interracial contact, or the real or perceived size of the Black population, Blacks appear to perceive their contexts equally. The final variable gender is only related to the perceived Black population size. Females’ estimates of the Black population tend to be larger than males’ estimates, although they are no more likely than males to come into contact with Whites. In summary, support for the actual Black population measure is extremely weak. It is only associated with urbanicity, a variable similarly related to Black population density. Age, income, and work status are all associated with both threat and prejudice, while education is not statistically associated with any of the key race items. Perceived Black population is strongly associated with income, and moderately associated with age and gender. Urbanicity is related to threat, interracial contact, and actual Black "~- population size. Those Blacks living in the suburbs are mostly likely to be prejudiced, perceive a threat, and have the highest levels of contact with Whites. Finally, perceived group position is only statistically associated with perceived threat, whereas those who perceive the most negative context are most likely to report that Whites are prejudiced. The results of the multiple bivariate analyses further illustrate the need to analyze the data in a multivariate setting. Multivariate Analysis Table 29 provides estimates from a logistic regression analysis. The cell values are adjusted odds ratios, which indicate the effect size of each independent variable. The baseline model contains only demographic variables, and each additional model incorporates additional independent variables. The results from the multivariate analysis show that Blacks’ prejudice towards Whites appears to be a function of age, perceived Black population size, and perceived threat. Also, race of interviewer is a consistently significant factor is Blacks reporting their prejudice. Perceived racial group size is significant, when controlling for only demographics, and when controlling for perceived threat along with the demographics. The results shown in Table 29 thus provide some support for the group position model, and the power threat model. However, since the power threat model is primarily based on actual population size, rather than perceived population size, the results clearly show that among the competing theories, the sense of group position accounts for the most variance. The data do not support the ethnic community, or the interracial contact model. The ethnic community model predicted that real or perceived increases in Black population size, should lead to a greater likelihood of prejudice. If this were the case odds ratios in the table would be greater than one, indicating a positive relationship. The actual racial group size variable is positive, but not statistically significant, while the perceived racial group size measure is statistically significant, and negative. Similar to 140 the bivariate analysis, interracial contact was not a statistically significant predictor of prejudice. In fact when added to the baseline model, interracial contact only added minimally to the explained variance. In summary, there is no support for the ethnic community or the interracial contact model. TABLE 29. Logistic Regression Odds Ratios for Regression of Prejudice Response; 1997 Gallup Race Relations Social Audits Baseline Model Model Model Model Model Model Predictors Model 1 2 3 4 5 6 Constant 2.24 2.82 2.54“ 2.21 .905 1.57 8.293 Gender (Males= l) 1.30 1.29 .971 1.30 1.16 1.15 .868 Age 589*" .583*** .532*** 589*" .610“ .624" 519*" Education 141* 1.42”“ 1.39 1.41“ 1.35 1.34 1.35 Household Income 1.21* 1.23* 1.09 1.21 * 1.13 1.16 1.05 Urban (Urban=l) 1.20 1.13 1.13 1.21 .876 .985 .758 Suburban (Suburban=l) 1.52 1.52 1.68 1.52 1.13 1.40 1.09 Work Status (Employed=1) 1.26 1.39 1.43 1.26 1.03 1.13 1.35 White Interviewer .412" .430M .356” .412" .516* .497* .429"‘ Interracial contact - .886 - - - - - Perceived Racial Group Size - - 729*" - - - .731" Actual Racial Group Size - - - 1,03 - - - Threat - - - - 605*" - 8.38“” Threat x Group Position - - - - - 1.45" .907 Number Of C3565 1,033 1,033 874 1,033 952 952 824 PseudoRSquare (Negelkerke) .140 .146 .203 .140 .254 .176 .322 Percentage Correctly Classified 63.9 66.1 66.1 63.6 72.1 68.5 72.1 Note: * p_<_ .05, ** p5 .01, **"‘ p5 .001 141 The final two models test a variable that is designed to observe the effects of group position if you are threatened or not. The measure is a product of the perceptions of whites prejudice and the four levels of the group position variable. Therefore, in the data, data points receive a value ranging from 0 (no threat, any all levels of perceived group position) to 4 (the perception that whites are prejudiced and the most negative level of perceived group position). The results show that the effects of threat and group position are statistically important when controlling only for demographics (model 5), but insignificantly when controlling for demographics and threat (model 6). The interpretation is that threat has a greater effect on prejudice, than perceived group position, when controlling for demographics. The results also show that older Blacks are less likely to be prejudice. As can be seen in comparison of mean levels of prejudice in Table 28, younger Blacks are more likely than older Blacks to be prejudiced. When estimating the probability of being prejudiced, controlling for other demographics, threat, and perceived racial group size, the probabilities are virtually linear for the age categories. The mean predicted probabilities were respectively PI‘Obprcjudicc = .73 for 18-27 year olds, PfObprejudicc = .69 for 28-38 year olds, Probpmjud,cc = .60 for 39-52 year olds, and Probpmjudicc = .44 for 53 years or greater. "-— The effect of race of interviewer is negative, indicating that Blacks interviewed by Whites are less likely to report being prejudice. The race of interviewer served as a check for the social desirability of asking about prejudice and other sensitive items. The statistical significance indicates that prejudice is still, at least, a sensitive topic indicative of the current racial climate. 142 The variables that are not statistically significant are also of note. The fact that work urbanicity, and work status are not significant implies that, contrary to previous findings related to individual self-interest (e.g., Bobo, 1983), in some cases individual level economic variables may be less important for Blacks, than for Whites. For example, based on previous research, I would expect neighborhood composition, as measured by urbanicity, to impact Blacks racial attitudes. Additionally, those Blacks who are unemployed might perceive they are in greater competition for jobs, and thus exact their frustration on Whites; yet, neither shows evidence of importance. In general, perceived threat is the dominant variable in the analysis of Blacks’ prejudice. When controlling for other background factors, if a Black person perceived that Whites in their community were prejudiced, their odds of prejudice increased 6.05 times. And, when controlling for background variables, perceived group size, and perceived negative group position, the effect of threat increased by a factor of 8.38. Discussion and Conclusions In order to gain a more thorough understanding of how prejudice operates across racial groups, it is important to study the opinions of both minority groups and Whites. This statement has been made numerous times (Bobo and Hutchings, 1996; Kinder and Sanders, 1996; Monteith and Spicer, 2001; Sigelman and Welch, 1991), yet the rate of quality studies of the Black population continues at a negligible pace. One reason for the lack of quality studies of Blacks racial attitudes is due to the rare nationally representative data sets containing large numbers of Black respondents (Schuman, Steeh, and Bobo, 1985; Schuman, et al., 1997; Sigelman and Welch, 1991). The Gallup Organization has attempted to answer the call for national studies of the 143 Black population. Beginning in 1997, and continuing through the present, they collected data for the Gallup Race Relations Social Audit. The original goal of the surveys was to provide a nationally representative baseline for attitudes and opinions regarding race relations, especially those of Blacks Americans. Fortunately, in the first year of the study, Gallup collected general data, that when carefully considered, could be used to test existing theories of prejudice. This goal of this study was to examine the relationships between racial composition and Blacks’ racial attitudes. Because studies of contextual effects have typically assumed that individuals know the out-group population size (or change in size) (e.g., Evans and Giles, 1986; Giles and Hertz, 1994; Key, 1949; Taylor, 1998), I was also interested in whether or not perceived size had a different effect than actual size. The results show that perceived in-group size has a statistically significant relationship with Blacks prejudice, while the actual proportion of Blacks does not. Thus, the general finding is that racial composition matters, but in this case, it’s the perceived in-group size rather than the actual population size that matters to Blacks. The primary limitation of this finding is that the actual and perceived populations are not directly comparable. The actual population is measured at the area code level, while the perceived population is measured at the national level. Future studies may attempt to measure estimates of both the local and national proportions of in-group or out-group size, to determine whether or not there are any differences in their effects on racial attitudes. The second goal of this study was to provide a potential explanation for these contextual effects, while simultaneously testing for alternative hypotheses. After testing the expected outcomes of four theories of racial attitudes, the results of this study show 144 support for the group position and power-threat models. There was no evidence to support the interracial contact or ethnic community models. When controlling for demographic variables, perceived in-group size and racial threat were consistently significant predictors of Blacks’ self-reported prejudice. Interracial contact was not statistically associated with prej udicc. nor was it associated with perceived racial threat. In testing the ethnic community model, increasing levels of real or perceived levels in- group population size, should have been positively associate with Blacks’ likelihood of reporting prejudice; however, I found no evidence that supported the expected directionality. Lower levels of estimated in-group population were associated with higher levels of prejudice, rather than the opposite. The results show evidence that the power-threat and group position models help account for Blacks prejudice attitudes. Particularly, the group position model, with it’s emphasis on subjective perceptions of where racial groups ought to stand, helps to explain the relationship between context and prejudice, for both Whites and Blacks This implies that perceived context (i.e., population size) may be just as, if not more, important as the actual context. If there is an assumption that out-group population, itself, is enough to bring about a threat, the impact of context should likely be framed as a perception, rather than an unobserved fact. Also, the results imply that Blacks prejudice is shaped by factors that are relatively uncontrollable by Whites. The perception that Whites are prejudice is simply a perception, and may not be ground in reality, but rather a sense of real or experienced frustration, attributional ambiguity, powerlessness, and potentially paranoia (Mirowsky and Ross, 1983). This notion may frustrate Whites, but it 145 seems to be a valid part of the equation in understanding Blacks racial attitudes towards Whites. 146 Chapter 7 DISCUSSION AND CONCLUSION There is no way to vanquish the uncomfortable history of race relations in the United States. It is a disturbing and unfortunate fact that Black Americans have been subjected to violence, hatred, discrimination, prejudice, and forced segregation, simply because of their skin color. Many Blacks died wondering if their race would ever be considered equal, if not even human, and in February of each year, Americans celebrate the culture, history, and heroic legacy of African-American life. There is no doubt that racial prejudice towards Blacks has been, and will always be a negative fact of life. But, given the cumulative acknowledgement of the past wrongs targeted at Black Americans, what should be made of the fact that Blacks are also prejudiced towards Whites? Surely, proposing that Blacks are prejudice towards Whites would fuel a number of political debates. On the one hand, emphasizing the prejudice of Blacks can be seen as providing evidence of the need to consider conservative political ideas, such as reverse discrimination, with more seriousness. Or as Shelton (2001) notes, emphasizing Blacks’ prejudice towards Whites may help absolve Whites of any responsibility for helping to eliminate racism, and may provide fuel for justifying Whites negative racial attitudes or behaviors. On the other hand, Black political leaders and activists may consider the reporting of Blacks’ racial attitudes towards Whites, akin to publicly airing “dirty laundry.” They may argue that exposing Blacks as prejudiced makes their job exposing racial prejudice and discrimination much more difficult. In the end, both liberals and conservatives, and Blacks and Whites would seemingly wind up defending their own 147 viewpoints of race relations. Yet, a very important distinction made in this research is that prejudice is an attitude, not necessarily indicative of racism, or discrimination. Blacks’ can be prejudiced, and can discriminate, but there is no evidence that they have an expressed desire to keep Whites “in their place.” Whites’ and Blacks’ prejudice are both potentially negative forces, than can lead to racial discrimination, however, the two races differ when it comes to racism. I believe that Blacks can be racist, in essence promoting their race as superior, and attempting to maintain any hold on power that they may have. However, any researcher would be hard pressed to show that, for example, Blacks believe Whites are unintelligent, lazy, happy go luck, or incompetent solely because they are Whites. Nor would we expect Blacks to implement public policy designed to keep Whites from moving into their neighborhoods, patronizing their stores, or eating at their restaurants. This is the distinction between Whites’ and Blacks’ racial attitudes, Whites have a history in America of providing prejudice, and Blacks’ have a history of receiving prejudice. Yet, this doesn’t mean that Blacks’ would not have extrapunative responses to prejudice. Analytic findings in the preceding Chapters show that Blacks are indeed prejudiced, and what some will find most distressing is that the Black middle class are the primary culprits of this prejudice. The segmentation study in Chapter 4 shows that r“: Blacks with the highest incomes and education are most likely to be prejudice. They are also the segment most likely to perceive that Whites are prejudiced, and that the racial environment is negative. And as the data in Chapter 5 show, this is true even when controlling for other demographics such as age, gender, urbanicity, employment status, and Black population density in residential areas. Simply stated, the Black middle class 148 has an eye on race relations in America, and they are making it known that thing are not okay. These findings will come as no surprise to researchers such as Elis Cose, Lawrence Bobo, and Jennifer Hochschild. These three have already put forth similar proposals about the growing disaffection of the Black middle class. The data in Chapters 4 and 5 both support the empirical research of Hochschild (1995) and Bobo (2001), and provide more ammunition to the qualitative arguments of Cose (1993). If there is one broad implication to be drawn from the combination of my own research along with the work of Cose, Bobo, and Hochschild, it’s that America may be going in the wrong direction in terms of solving race relations. The Civil Rights Act of 1964 had two main purposes of note to this research. First, the Act was designed to bring about equal opportunities for Blacks, and other underrepresented minorities, including women through Affirmative Action programs. Second, there was a hope that the Act and its supporting programs would help to eliminate racial prejudice. While there is no evidence that the Act’s outcomes have directly contributed to Blacks’ prejudice, there is evidence that some, who were the intended target of the Act, are actually most likely to be prejudiced. For example, additional reports from the data show that more educated Blacks have the highest lif -- job, safety, finances, standard of living, and way of life -- satisfaction. Yet, more educated blacks were also most likely to report experiencing discrimination, to believe that race relations “have gotten worse”, that race relations will “always be a problem”, and that racial profiling is “widespread (Gallup Organization, 1997; 1998; 1999).” For Blacks, it appears that attaining more education helps to shed light on the perceived 149 nature and political consequences of a history of negative race relations, rather than explicitly reduce prejudice through knowledge. What should also be made of evidence that Blacks who are benefiting most in the American economic system are also the most likely to report they are prejudiced towards Whites? As the data in Tables 16a and 16b show, those Blacks with the highest incomes are the ones most likely to perceive a negative context, perceive that Whites are prejudiced, and report being prejudiced. They are most likely to report experiencing unfair treatment due to race, and most likely to perceive limited social and economic opportunities. Is it the case that these high income Blacks are simply having to deal with more prejudice? Or maybe, they simply encounter more ambiguous situations that create a perception that prejudice is common. Regardless of why, higher income, and more educated Blacks, are the most prejudice demographic categories, and their status as most prejudiced signals a subtle differences in prejudice for Blacks and Whites. Allport’s (1954) model of prejudice emphasizes the irrational component of out- group hostility. One of the main components of this irrationality is ignorance about members of an out-group. Any factor that might enhance understanding, tolerance, and information about an out-group should reduce levels of prejudice and hostility (Bobo and Hutchings, 1996). One such factor is clearly education. Prejudice scholars believe that education is a lens by which Whites increase their tolerance and understanding of racial out-groups (Sniderman and Piazza, 1993). However, this view is clearly not the case with Black Americans. While educated Blacks are more likely to be prejudice, they should also be less ignorant and irrational about race relations. This is a slight paradox of the relationship between education and prejudice across race. When controlling for race, 150 it appears that education is a lens by which Blacks calculate the necessary threat of racism and prejudice, and respond with reciprocal attitudes, and for Whites education is a lens by which they learn tolerance, understanding, and the benefits of diversity. Thus, while highly educated Blacks are most likely to be prejudiced, they are also most likely to perceive they have a good reason to be prejudiced. As stated in Chapter 2, this is not the first time reciprocal prejudice has been proposed. Duckitt and Mphuthing (1998) provide empirical evidence showing that South African Blacks exhibit significantly higher levels of prejudice against Afrikaans Whites than against either English Whites, or Whites in general. The discrepancy is presumably due to the fact that Afrikaans Whites are perceived as being more racist and antagonistic than either English Whites or White sin general. Using content controlled essays, Monteith and Spicer (2000) show that the content of Black college students’ negative attitudes towards Whites are described in terms of reactions of perceived racism. Black participants essays suggested that prejudice and discrimination are perceived quite frequently and provide a good portion of the rationale for negative attitudes towards Whites. Further, these negative attitudes towards Whites were unrelated to traditional egalitarianism and Protestant ethic (Monteith and Spicer, 2000). Terrell and Terrell (1981) promoted the notion of cultural mistrust developed in everyday dealings of ° --, perceived prejudice. Judd etal., (1995) even propose that Whites’ promotion of a color- blind society is a potential threat to Blacks because it denies the importance of individual cultural differences. Finally, Davis (1994) hypothesizes that Blacks use attitudes such as political intolerance as an “emancipation strategy”. Accordingly, Blacks are intolerant -- selectively deny democratic freedoms -- towards groups they find threatening. Davis’ 151 Black intolerance study, presents one of the strongest arguments for the contextual racial threat model, because it shows that Blacks can respondent to threat with negative out- group attitudes. Indeed, Blacks appear to use their anti-out-group attitudes as a protective strategy to reduce the chance of returning to a history of targeted violence, intimidation, and racial bigotry. Unfortunately, the by-product of having a group-protective strategy is that those who are at the forefront of the racially competitive arena are middle and upper class Blacks. This line of thinking is in direct contrast with the beliefs and proposals of recent research on Blacks’ prejudiced, presented by Sniderman and Piazza (2002). Sniderman and Piazza (2002) find that higher income and educated Blacks are more likely to be optimistic about their financial future, and less anemic about life opportunities. They argue that such findings are support that Blacks have not become disaffected with the American dream. I would argue that financial optimism and race relations operate on completely different dimensions for Blacks. Blacks are ambitious about their financial security, while at the same time they are not na'ive about the nature of race relations in America. Black Americans likely believe, if not promote, the idea that it is possible to achieve economic success even at the hands of racial animus. Thus, contrary to the beliefs of Sniderman and Piazza (2002) it is very possible for higher SES Blacks to believe that race relations will always be a problem in American, and still feel positive about their economic futures. In Chapter 6, I address the relationship between racial composition and prejudice. The impetus was to discover the extent to which encountering persons of an opposite race might influence prejudice attitudes. Up to now, research on racial composition and 152 prejudice has focused its efforts on explaining the attitudes of White Americans (Giles and Hertz, 1994; Key, 1949; Quillian, 1996; Taylor, 1998;), and the efforts have shown that in areas where Blacks constitute a larger proportion of the population, Whites are more hostile to their political interests (Glaser, 203). As it turns out, Blacks also respond to out-group population size, although the data show a slightly different relationship than seen with previous studies. Chapter 6 shows that that Blacks, who perceive they make up a smaller proportion of the US. population, are more likely to be prejudiced. This finding fits within the study of contextual effects in two important ways. First, perceptions about the population composition appear to be more, if not as. important as fact. This falls in line with statements from Glaser (2003) regarding the contextual effects. Glaser, states that most research on context effects begins with the troublesome assumption that researchers can specify a relevant context for individuals. That is, when arbitrary geographic boundaries are used to capture context, there is a greater chance that population size is less relevant to individual attitudes. This is backed by studies of innumeracy that show that American have highly inaccurate ideas about racial group composition in the US. (Nadeau et al., 1993; Sigelman and Niemi, 2001). Second, the data show no support for the interracial contact hypothesis: as contact increase, respondent prejudice remains statistically unchanged. This finding is important because it shows that Blacks who have more interracial contact with Whites are no more likely to be prejudiced, than those Blacks who have less contact. According to the data, the key to understanding Blacks’ prejudice is the perception of threat. If Blacks perceive they exist in a racially divisive world, where 153 some groups are afforded more opportunities than others, then they are more likely to assign fear to the perception that Whites in their community are prejudiced. The fear of racially relevant past inequities and potential discrimination produces a negative perception of Whites, and thus prejudices are developed towards the majority group. However, Blacks’ prejudice is not reduced by a change in context -- perceived fair treatment, and more opportunities -- alone. As shown in Chapter 5, Blacks who perceive no negative context, but perceive a threat are more likely to be prejudice, than those Black who perceive the most negative context, but no threat. If the perception that Whites are prejudiced could somehow be reduced, it’s likely that Blacks prejudice towards Whites might also be reduced. Yet, given the constant reminders of American history, this is an unlikely scenario. Unfortunately, I cannot present a panacea for racial prejudice and bigotry; however, I believe that there are important points about studying race to underscore from this research. First, blacks must be assured that racial prejudice and discrimination are a thing of the past. It is very likely that whenever Blacks see less budgetary and political support for public policies and programs, such as Affirmative Action, the Equal Employment and Opportunity Commission (EEOC), and public education, they perceive a threat to their race’s ability to gain equal opportunities, but not necessarily their own. This threat may be especially pronounced for higher income and educated Blacks. This is not to suggest that America keep policies that it may not feel are necessary, however, there are consequences that result the motivations of political elites. In general, the when Blacks’ perceive their status is threatened, they are more likely to respond with negative attitudes. 154 The second point from this research centers on the need to continuously study the racial, political, and social attitudes of racial minorities. Over the past seven years, the Gallup Organization has collected public opinion data on national over-samples of racial minorities. They are one of only a few survey organizations even attempt this endeavor. However, with the shifts in budgetary priorities in academia and market research, across the country, original data collection on racial minorities may take a back seat to more general studies. The short—term costs of this direction will be the lack of understanding of racial differences in public opinions and attitudes about various features of American society. The long-terrn costs will be the lack of sound theory to help account for these differences. The Gallup data have afforded an opportunity to understand Blacks’ prejudice attitudes, but much more research needs to develop. The insights from these data are invaluable, and deserve continued exploration. This study will hopefully serve as a springboard for more comparative racial research. Limitations There are a number of limitations to be mentioned with this study. First, the data were collected via a telephone methodology, which makes more lengthy surveys of psychological measures and social attitudes more difficult. Most of the items asked in the survey were not designed for the purpose that I consider in this analysis. This makes inference slightly more difficult, and immediately brings about issues of validity. To the extent possible, each construct was considered a content valid measure. That is, considering the domain of topics and research literature associated with the dependent measure, these items where subjectively considered relevant. Another name for this process is establishing face validity: based on a qualitative evaluation of the measures, 155 did items measure, on the surface, the intended characteristic (Allen and Yen, 1979). I both assume that the measures are face valid, and duly acknowledge their limitations on my findings. In addition, the majority of analyses were conducted on a pooled data set of three separate years. In most cases, there were no differences across the years; however, it is possible that the respondents in one year differ on important attributes, than the data in another year. As a result, a number of the models and analysis may have some level of bias. This tradeoff between a parsimonious presentation of multiple data files, and a rigorous cross-sectional time series analysis is sometimes a costly one, and the results do not in any way imply that there are zero differences across the three years of data. 156 APPENDIX A Items from the 1997 — 1999 Gallup Race Relations Social Audits Blacks’ Self-Reported Prejudice I would like you to place yourself on a zero-to-ten scale, where “0” means that you are NOT prejudiced against Whites at all and where “10” means that you are EXTREMELY prejudiced against Whites. Which number would you choose to describe yourself? You can choose any number between zero and ten. Perceived Racial Threat Finally, think for a moment about Whites in the area where you live. Use the same zero-to- ten scale, where “0” means that they are NOT prejudiced against Blacks at all and where “10” means that they are EXTREMELY prejudiced against Blacks. Which number would you choose to describe Whites in the area where you live? You can choose any number between zero and ten. Racial Context — Limited/Unequal Opportunities In general, do you think that Blacks have as good a chance as White people in your community to get any kind of job for which they are qualified, or don’t you think they have as good a chance? (Have Chance, Don’t Have Chance, DK) Again, in general, do you think that Blacks have as good a chance as White people in your community to get any housing they can afford, or don’t you think they have as good a chance? (Have Chance. Don’t Have Chance, DK) In general, do you think that Black children have as good a chance as White children in your community to get a good education, or don’t you think they have as good a chance? (Have as good a chance, Don’t have as good a chance) Racial Context — Group Unfair Treatment Just your impression, are Blacks in your community treated less fairly than Whites in the following situations? (Yes, No, DK) How about: A. On the job or at work B. On public transportation C. In neighborhood shops D. In stores downtown or in the shopping mall E. In restaurants, bars, theaters, or other entertainment places Racial Context — Individual Unfair Treatment (Experiences with Discrimination) 157 Can you think of any occasion in the last thirty days when you felt you were treated unfairly in the following places because you were Black? (Yes, No, DK) How about : A. In a store where you were shopping B. At your place of work (If employed) C. In a restaurant, bar, theater, or other entertainment place D. While using public transportation E. In dealings with the police, such as traffic incidents F. Any other situations? What other situation? (Open ended) Ethnocentrism Do you approve or disapprove of marriage between Blacks and Whites? (Approve, Disapprove, DK) In living in a neighborhood, if you could find housing you want and like, would you rather live in a neighborhood with Black families, or in a neighborhood that had both Whites and Blacks? (Blacks, Whites and Blacks, Other, DK) At work, would you rather work alongside mostly other Blacks, or would you rather work with a mixed group of Whites and Blacks? (Mostly other Blacks, Mixed Group, Other, Not sure, DK) Interracial Contact Now, thinking about the CHURCH OR PLACE OF WORSHIP you attend, are the people in your place of worship (All White, Mostly White, About half White and half Black, Mostly Black, All Black, Or, some other, DK). How about the NEIGHBORHOOD or area where you live in? Are the people in your neighborhood? worship (All White, Mostly White, About half White and half Black, Mostly Black, All Black, Or, some other, DK). How about the place where you WORK? Are the people you work with (All White, Mostly White, About half White and half Black, Mostly Black, All Black, Or, some other, DK). How about any CLUBS OR ORGANIZATIONS you happen to belong to? (All White, Mostly White, About half White and half Black, Mostly Black, All Black, Or, some other, DK). How about the school your OLDEST child attends? Are the students there (All White, Mostly White, About half White and half Black, Mostly Black, All Black, Or, some other, DK). 158 Demographics Employment Status: Are you currently working full-time, part-time, or not employed? (1=Full or Part-time, 0=Not Employed) Gender: 1=Male; 0=Female Age: Please tell me your age. Education: What is the highest level of education you have completed? (Open ended and code) (1=LT HS, =HS Graduate, 3=Some College, 4=College Graduate) Income: Is your total annual household income, before taxes, over or under $25,000? (1=Less Than $25K, $25K — Less Than $35K, $35K - Less Than $45K, $45K — Less Than $55K, 5=$55K or more) Party Identification: In politics, as of today, do you consider yourself a Republican, a Democrat or Independent? (Other Party, DK) Urbanicity (coded from fone file): 1=Urban, 2=Suburban, 3=Rural Black Density (coded from fone file): 1=High, 2=Medium, 3=Low Race of Interviewer Actual Race of Interviewer (coded from fone file): 1=White, 0=Black Perceived Race of Interviewer: This last question is just for research purposes. You may not have thought about this but I’d like to ask you to guess my race. Would you guess that I am white, black or some other race? Perceived Black Racial Composition Just your best guess, what percent of the United States population today would you say is Black? (Open ended and code) (1 = Less than 10%, 10% to less than 15%, 15% to less than 20%, 20% to less than 30%, 30% to less than 40%, 40% to less than 50%, 7=50% or more, DK). 159 REFERENCES Aberbach, Joel and Jack L. Walker (1973) Race in the City. Little, Brown. Abramson, Paul (1983) Political Attitudes in America: Formation and Change. San Francisco, CA: Freeman and Co. Allen. Mary J. and Wend M. Yen (1979) Introduction to Measurement Theory. Monterey, CA: Books/Cole. Allen, Richard L., Michael C. Dawson, and Ronald E. Brown (1989) “A Hueuristic and Schema Approach to Modeling an African American Racial Belief System.” American Political Science Review. 83:421-441. Allport, Gordon W. (1954) The Nature of Prejudice. Reading, MA: Addison-Wesley. Altemeyer, Bob. (1988) Enemies of Freedom: Understanding Right Wind Authoritarianism. San Francisco. Jossey-Bass. Anderson, Barbara, Brian Silver, and Paul Abramson (1988a) “The Effects of the Race of the Interviewer on Measures of Electoral Participation by Blacks in SRC national Election Studies.” Public Opinion Quarterly. 53:53-83. Anderson, Barbara, Brian Silver, and Paul Abramson (1988b) “The Effects of the Race of the Interviewer on Race Related Attitudes of Black Respondents in SRC/CPS National Election Studies.” Public Opinion Quarterly. 53:53-83. Asante, Moliefi (1987) The Afroncentric Idea. Philadelphia: Temple University Press. Baldwin, Joseph A. and Yvonne R. Bell (1985) “The African Self-Consciousness Scale: an Afrocentric Personality Questionnaire.” The Western Journal of Black Studies, 9: 61- 68. Baughman, Emmett.E. (1971) Black Americans. NY: Academic Press. Bay, Christian (1981) Strategies of Political Emancipation. South Bend, IN: University of Notre Dame. Biemat, Monica and Christian S. Crandall (1999) “Racial Attitudes”, In John P. Robinson, Phillip R. Shaver, and Lawrence S. Wrightman (eds.) Measures of Political Attitudes. V0. 2, (pp. 297-411) San Diego, CA: Academic Press. Blalock, Hubert (1965) Toward a Theory of Minority Group Relations. NY: Capriocom Books. 160 Blumer, Herbert (1958) “Race Prejudice as a Sense of Group Position.” Pacific Sociological Review. 123-7. Bobo. Lawrence D.(1983) “Whites’ Opposition to Busing: Symbolic Racism or Realistic Group Conflict?” Journal of Personality and Social Psychology. 45: 1 196-1210. Bobo, Lawrence D. (1988) “Group Conflict, Prejducie, and the Paradox of Contemporary Racial Attitudes.” In P.A. Katz and DA. Taylor (Eds.) Eliminating Racism: Profiles in Controversy. NY: Plenum. Bobo, Lawrence D. (1999) “Prejudice as Group Position: Microfoundations of a Sociological Approach to Racism and Race Relations.” Journal of Social Issues. 55: 445- 473. Bobo, Lawrence D. (2001) “Racial Attitudes and Relations at the Close of the Twentieth Century.” In Smelser, Neil, William J. Wilson, and Faith Mitchell (Eds.), American Becoming: Racial Trend and their Consequences. Washington, DC: National Academy Press. Bobo, Lawrence D.., & Hutchings, Vincent. L. (1994) “Race, Immigration, and the Sense of Group Position.” Paper presented at the 1994 Annual meeting of the American Political Science Association, New York, NY. Bobo, Lawrence D., and Vincent L. Hutchings (1996) “Perceptions of Racial Group Competition: Extending Blumer’s Theory of Group Position to a Multiracial Social Context.” American Sociological Review. 61 : 951-973. Bobo, Lawrence D. and James Kluegel (1993) “Whites’ Opposition to Busing: Symbolic Racism or Realistic Group Conflict?” Journal of Personality and Social Psychology. 70:1142-1163. Bobo, Lawrence D. and James Kluegel (1997) “The Color Line, the Dilemma, and the Dream: Racial Attitudes and Relations in American at the Close of the Twentieth Century.” In J. ngam (ed.) Civil Rights and Social Wrong: Black- White Relations Since World War II. University Park: Pennsylvania State University. Bobo, Lawrence D., Charles Zubrinsky, James Johnson, and Melvin Oliver (1994) “Pbulci Opinion Before and After a Spring of Discontent.” In M. Baldassare (ed.) The Los A ngeles Riots: Lessons for the Urban Future. Boulder, CO: Westview Press. Blalock, Hubert (1967) Toward a Theory of Minority-Group Relations. NY: Wiley. Branscombe, Nyla R. and Naomi Ellemers (1998) “Coping with Group-Based Discrimination: Individualistic versus Group-Level Strategies,”. In Janet K. Swim and Charles Stangor (eds.) Prejudice: The Target '5 Perspective. San Diego, CA: Academic Press. 161 Brazziel, William F. (1974) “Career Education and Black Americans.” In R. Johnson Black Agenda for Career Education. Columbus, OH: ECCA Publications. Brewer, Marilynn B. (1999) “The Psychology of Prejudice: lngroup Love or Outgroup Hate?” Journal of Social Issues. 55: 429—444. Brigham, John C. (1993) “College Students’ Racial Attitudes. " Journal of Applied Social Psychology. 23: 1933-1967. Brink, William, and Louis Harris, (1964) The Negro Revolution in America. NY: Simon and Schuster. Brown, Barry and George Albee (1966) “the Effect of Integrated Hospital Experiences on Racial Attitudes.” Social Problems. 13:324-333. Bullock, Charles and Harrell R. Rodgers, Jr. (1977) Black Political Attitudes: Implications for Political Support. Chicago: Markam. Campbell, Angus and Howard Schuman (1968) “Racial Attitudes in Fifteen American Cities.” In Supplemental Studies for the National Advisory Commission on Civil Disorders. Washington, DC: US. Government Printing Office. Campbell, Bruce A. (1981) “Race-of-Interviewer Effects Among Southern Adolescents.” Public Opinion Quarterly. 45:231-234. Charmichael, Stokeley and Charles V. Hamilton (1967) Black Power. NY: Vintage. Carmines, Edward G. and James A. Stimson (1980) “The Two Faces of Issue Voting.” American Political Science Review. 74: 78-91. Cose, Eloise (1993) The Rage of a Privileged Class. NY: Harper Collins. Cotter, Patrick, Jeffrey Cohen, and Philip Coulter (1982) “Race-of-Interviewer Effects in Telephone Interviews.” PubliC'Opinion Quarterly. 46:278-284. Crocker, Jennifer, and Brenda Major (1989) “Social Stigma and Self-Esteem: The Self- Protective Properties of Stigma.” Psychological Review, 96: 608-630. Crocker, Jennifer, Voelkl, K. Testa, M., and Brenda Major (1991) “Social Stigma: The Affective Consequences of Attributional Ambiguity.” Journal of Personality and Social Psychology. 60: 218-228. Crocker, Jennifer, Brenda Major, and Claude Steele (1998) “Social Stigma.” In Daniel J. Gilber, Susan T. F iske, and Gardner Lindzey (eds.) Handbook of Social Psychology (4’h ed.). NY: McGraw-Hill. Cross, William .E. (1995) “The Psychology of Ni gresence: Revising the Cross Model.” In Joseph G. Ponterotto, J. Manuel Casas. Lisa A. Suzuki, and Kathleen M. Alexander (eds.) Handbook of Multicultural Counseling. Thousand Oaks, CA: Sage Publications. Cross, William E. (1991) Shades of Black: Diversity in African-American Identity. Philadelphia: Temple University Press. Curry, Timothy J. and Richard M. Emerson (1970) “Balance Theory: A Theory of Interpersonal Attraction.” Sociometry. 33:216-238. Curtis, Rebbeca C., and Kim Miller (1986) “Believing Another Likes or Dislikes You: Behaviors Making the Beliefs Come True.” Journal of Personality and Social Psychology. 51: 284-290. D’Souza, Dinesh (I 995) The End of Racism.“ Principles or a Multiracial Society. NY: The Free Press. Davis, Darren W. (1997a) “The Direction of Race of Interviewer effects among Afi’ican Americans: Donning the Black Mask.” American Journal of Political Science. 41: 309- 330. Davis, Darren W. (1997b) “Nonrandom Measurement Error and Race of Interviewer Effects among African Americans.” Public Opinion Quarterly. 61 :183-207. Davis, Darren W. (1995) "Exploring Black Political Intolerance." Political Behavior 17: l-22. Davis, James A. (1973) “Busing.” In R.L. Crain (ed.) Southern Schools. Chicago: National Opinion Research Center. Davis, James A. and Tom W. Smith (1992) The NORC General Social Survey: A User '5 Guide. Newbury Park, London: Sage. Dawson, Michael (2002) Black Visions : The Roots of Contemporary A frican-American Political Ideologies. Chicago: University of Chicago Press. Dawson, Michael (1994) Behind the Mule: Race and Class in African-American Politics. Princeton, NJ: Princeton University Press. Decker, Scott (1981) “Citizen Attitudes Toward the Police: A Review of Past Findings and Suggestions for Future Policy.” Journal of Police Science and Administration. 9: 80- 87. Deutsch, Morton and Mary Collins (1951) Interracial Housing. University of Minnesota Press. 163 Devine, Patricia G. and Andrew J. Elliot (1995) “Are Racial Stereotypes Really Fading? The Princeton Trilogy Revisited.” Personality and Social Psychology Bulletin, 21: 1139- 1150. DeVelis, Robert F. (1991) Scale Development: T heary and Applications. Newbury Park, London: Sage Publications. Dovidio. John F., Kerry Kawakami and Samuel L. Gaertner (2001) “Implict and Explicit Attitudes: Examination of the Relationship Between Measures of Intergroup Bias. In R. Brown and Samuel L. Gaertner (eds.) Blackwell Handbook of Social Psychology (vol. 4: Intergroup Relations) Oxfourd, UK: Blackwell. Duckitt, John and Thobi Mphuthing (1998) “Group identification and Inter-group Attitudes: A Longitudinal Analysis in South African. Journal of Personality and Social Psychologi. 74:80-85. Dyson, Michael E. (1996) Race Rules: Navigating the Color Line. NY: Vintage Books. Ellison, Christopher G., and David A. Gay (1989) "Black Political Participation Revisited: A Test of Compensatory, Ethnic Community, and Public Arena Models." Social Science Quarterly. 70: 1 01-1 19. Erickson, Erik H. (1983) Childhood and Society. NY: Norton. Evans, Arthur S. Jr., and Michael W. Giles (1986) “Effects of Percent Black on Black’ Perceptions of Relative Power and Social Distance.” Journal of Black Studies. 17:3-14. Farley, Reynolds (1993) “Continued Racial Residential Segregation Detroit: Chocolate City , Vanilla Suburbs revisited.” Journal of Housing Research, 4: 1-3 8. Farley, Reynolds, Shirley Hatchett, and Howard Schuman (1979) “A Note on Changes in Black Racial Attitudes in Detroit.” Social Indicators Research. 6:439-443. F eagin, Joe R. (1991) “The Continuing-Significance or Race: AntiBlack Discrimination in Public Places. American Sociological Review, 56: 101-116. Feagin, Joe R. (1997) “Fighting White Racism: The Future of Equal Rights in the United States.” In S. Myers (ed.) Civil Rights and Race Relations in the Reagan-Bush Era. New York: Praeger. Feagin, Joe R. and Melvin P. Sikes (1994) Living with Racism: The Black Middle Class Experience. Boston, MA: Beacon Press. 164 Finkel, Steven E., Thomas M. Guterboxk, and Martin J. Borg (1991) “Race-of- lnterviewer Effects in a Presidential Poll: Virginia 1989.” Public Opinion Quarterly. 55:313-330. Firebaugh, Glenn and K. Davis (1988) “Trends in Anti-Black Prejudice, 1972-1984: Region and Cohort Effects.” American Journal of Sociology. 94: 251-272. Foley, Linda A. (1977) “Personality Characteristics and Interracial Contact as Determinants of Black Prejudice Toward Whites. ”Human Relations. 30:709-720. Ford, W. Scott ( I 973) “Interracial Public Housing in a Border city: another Look at the Contact Hypothesis.” American Journal of Sociology, 78: 1426-1477. Fosset, Mark A. and K. Jill Kiecolt (1989) “The Relative Size of Minority Populations and White Racial Attitudes.” Social Science Quarterly. 70:820-835. Gallup Organization (1997) “The Gallup Poll Social Audit on Black/White Relations in the United States”. Data set and Documentation, Gallup Organization, Princeton, NJ. Gallup Organization (1998) “The Gallup Poll Social Audit on Black/White Relations in the United States”. Data set and Documentation, Gallup Organization, Princeton, NJ. Gallup Organization (1999) “The Gallup Poll Social Audit on Black/White Relations in the United States”. Data set and Documentation, Gallup Organization, Princeton, NJ. Gibson, James L. (1995) “The Political Freedom of African-Americans: a Contextual Analysis of Racial Attitudes, Political Tolerance, and Individual Liberty.” Political Georgraphy, 14: 571-599. Gibson, James L. (1992) The Political Consequences of Intolerance: Cultural Conformity and Political Freedom.” American Political Science Review, 86: 338-356. Giles, Micheal W. and Kaenan Hertz (1994) “Racial Threat and Partisan-Identification.” American Political Science Review. 88: 317-326. Glaser, James M. (2003) “Social Context and Inter-Group Political Attitudes: Experiments in Group Conflict Theory.” British Journal of Political Science. 33:607-620. Glenn, Norval D. (1966) “Occupational Benefits to Whites from the Subordination of Negroes.” American Sociological Review. 28: 443-448. Goffman, Edward (1959) The Presentation of Self in Everyday Life. Garden City, NY: Doubleday. Grier, William H. and Price M. Cobbs (1968) Black Rage. NY: Basic Books. 165 Groves, Robert M. (1989) Surveys Errors and Survey Costs. NY: Wiley. Gutterbock, Thomas M. and Brian London (1983) “Race, Political Orientation. and Participation: An Empirical Test of Four Competing Theories.” American Sociological Review. 48:439-453. Gurin, Patricia, Geral Gurin, R.C. Lao, and M. Beattie (1969) “Intemal-Extemal control in the Motivational Dynamics of Negro Youth.” Journal of Social Issues, 25: 29-53. Gurin, Patricia, Shirley Hatchett. and James S. Jackson (1989) Hope and Independence: Blacks ’ Response to Electoral and Party Politics. NY: Russell Sage. Hacker, Andrew (1992) Two Nations: Black and White, Separate, Hostile, Unequal. NY: Scribners. Harding, John, Bernard Kutner, Harold Proshansky and Isidor Chein (1969) “Prejudice and Ethnic Relations.” In G. Lindzey and E. Aronson (Eds.) Handbook of Social Psychology (2nd ed., Vol. 5). Reading, MA: Addison-Wesley. Hatchett, Shirley and Howard Schuman (1975-76) “White Respondents and Race-of- Interviewer Effects.” Public Opinion Quarterly. 39: 523-528. Hochschild, Jennifer (1995) Facing Up to the American Dream: Race, Class, and the Soul of the Nation. Princeton, NJ : Princeton University Press. Jackson, James 8., Ron E. Brown, Michael Dawson, Katherine Tate and Shirley J. Hatchett (1984-1988) The 1984-1988 National Black Election Panel Study: A Sourcebook. Ann Arbor, MI: Institute for Survey Research. Jackson, James 8., TB. Brown, D.R. Williams, M.E. Torres, S.L. Sellers, and KB. Brown. (1996) “Racism and the Physical and Mental Health Status of African Americans: A 13-year National Panel Study.” Ethnicity and Disease. 6: 1 32-147. Jones, James M. (1997) Prejudice and Racism (2nd Ed.). NY: McGraw-Hill. Judd, Charles M. and Bernadette Park (1993) “Definition and Assessment of Accuracy in Social Stereotypes.” Psychological Review. 100:109-128. Judd, Charles M., Bernadette Park, Carey S. Ryan, Markus Brauer and Susan Kraus (1995) “Stereotypes and Ethnocentrism: Diverging Interethnic Perceptions of African American and White American Youth ” Journal of Personality and Social Psychologv. 69: 460-481. Katz, Daniel and Kenneth Braly (1933) “Racial Stereotypes of One Hundred College Students.” Journal of Abnormal and Social Psychology. 28:280-290. 166 Katz, Irwin and R.Glen Hass (1988) “Racial Ambivalence and American Value Conflict: Correlational and Priming Studies f Dual Cognitive Structures.” Journal of Personality and Social Psychology. 55:893-905. Katz, Phyllis A. and Dalmas A. Taylor (1988) Eliminating Racism: Profiles in Courage. (eds.) NY: Plenam. Key, Valdimer 0., Jr. (1949) Southern Politics — In State and Nation. Knoxville, TN: University of Tennessee Press. Kinder, Donald and Lynn Sanders (1996) Divided by Color: Racial Politics and Democratic Ideals. Chicago, IL: Univ. of Chicago Press. Kitano, Harry H. (1974) Race Relations. Englewood Cliffs, NJ: Prentice-Hall. Klockars, Alan J. and G. Sax (1986) Multiple Comparisons. SAGE Quantitative Applications Series, Thousand Oaks, CA: Sage. Klugel, James R. (1990) “Trends in Whites’ Explanations of the Black-White Gap in Socioeconomic Status, 1977-89.” American Sociological Review. 55:512-525. Klugel, James (1986) Beliefs about Inequality: A mericans ’ views of What is and What Ought to Be. NY: Aldine de Gruyter. Krosnick, Jon and Duane Alwin (1989) “Aging and Susceptibility to Attitude Change.” Journal of Personality and Social Psychology. 57:416-425. Krosnick, Jon A. (1999) “Maximizing Questionnaire Quality”, In John P. Robinson, Phillip R. Shaver, and Lawrence S. Wrightman (eds.) Measures of Political Attitudes. V0. 2, (pp. 297-411) San Diego, CA: Academic Press. Lamouse-Smith, W. Bediako (1973) “The Myth of Black Racism in the US.” Transition. 0:26-34. Lenski, Gerhard E., and John C. Leggett(l960) “Caste, Class, and Deference in the Research Interview.” American Sociological Review. 65:463-467. Lewin, Kurt (1948) Resolving Social Conflicts. NY: Harper and Brothers. Livingston, Robert W. (2002) “The Role of Perceived Negativity in the Moderation of African Americans’ Implicit and Explicit Racial Attitudes.” Journal of Experimental Social Psychology. 3 8: 405-41 3. Majors, Richard and Janet Mancini-Billson (1992) Cool Pose: Dilemmas of Black Manhood in America. NY: Lexington Books. 167 Martin, Elizabeth (1983) “Surveys as Social Indicatory: Problems in Monitoring Trends.” In Peter H. Rossi, James D. Wright, and Andy B. Anderson (Eds.), Handbook of Survey Research. San Diego, CA: Academic Press. Marx, Gary T. (1967) Protest and Prejudice: A Study ofBelief in the Black Community. NY: Harper ad Row. NY: Harper and Row. Massey, Douglass S. and Nancy A. Denton (1993) American Apartheid: Segregation and the Making of the Underclass. Cambridge: Harvard University Press. McConahay, John G. (1986) “Modern Racism, Ambivalence, and the Modern Racism Scale.” In John F. Dovidio and Samuel L. Gaertner (Eds.), Prejudice, Discrimination, and Racism. NY: Academic Press. McConahay, John B. and Joseph C. Hough, Jr. (1976) “Symbolic Racism.” Journal of Social Issues, 32: 23-39. Meer, Bernard and Edward Freedman (1966) “The Impact of Negro Neighbors on White Homeowners.” Social Forces. 45:1 1-19. Mills, Jon K., Donna McGrath, Patti Sobkoviak, Susan Stupec, and Sharlee Welch (1995) “Differences in Expressed Racial Prejudice and Acceptance of Others.” Journal of Psychology. 129: 357-359. Mirowsky, John and Catherine E. Ross (1983) “Paranoia and the Structure of Powerlessness.” American Sociological Review. 48:228-239. Monteith, M.J. and Spicer, CV. (2000) “Contents and Correlates of Whites’ and Blacks’ Racial Attitudes.” Journal of Experimental Social Psychology. 36: 1 25-154. Moore, William J ., William E. Hauck and Thomas C. Denne (1984) “Racial Prejudice, Interracial Contact, and Personality Variables.” Journal of Experimental Education. 52:168-173. Myrdal, Gunnar (1944) The American Dilemma. NY: Harper & Row. Nadeau, Richard, Richard G. Niemi, and Jeffrey Levin (1993) “Innumeracy about Minority Populations.” Public Opinion Quarterly. 57: 332-347. National Advisory Commission on Civil Disorders (1968) Report of the National Advisory Commission on Civil Disorders. NY: Bantam. Nickerson, Kim .J., Janet E. Helms, and Francis Terrell (1994) “Cultural Mistrust, Opinions About Mental Illness and Black Students’ Attitudes Toward Seeking Psychological Help From White counselors.” Journal of Counseling Psychology, 41 : 378-385. 168 Noel, Donald L. (1964) “Group Identification among Negroes: An Empirical Analysis.” Journal of Social Issues. 20:71-84. Noel, Donald L., and Alphonso Pinkney (1964) “Correlates of Prejudice: Some Racial Differences and Similarities.” American Journal of Sociology, 69: 609-622. Oliver, J. Eric and Tali Mendelberg (2000) “Reconsidering the Environmental Determinants of White Racial Attitudes.” American Journal of Political Science. 44:574- 589. Olsen, Mancur E. (1970) “Social and Political Participation of Blacks.” American Sociological Review. 35: 682-697. Paige, Jeffrey (1970) “Changing Patterns of Anti-White Attitudes Among Blacks.” Journal of Social Issues. 26: 69-86. Parsons, C. (1972) America ’s Uncounted People: Report of the Advisory Committee on Problems of Census Enumeration. Washington, DC: National Academy of Sciences. Patchen, M. (I 982) Black- White Contact in Schools: Its Social and Academic Effects. West Lafayette. IN: Purdue University Press. Peffley, Mark, Jon Hurwitz, and Paul M. Sniderman (1997) “Racial Stereotypes and Whites’ Political Views of Blacks in the Context of Welfare and Crime,” American Journal of Political Science. 41: 30-60. Perkins, Eugene (1975) Home is a Dirty Street: The Social Oppression of Black Children. Chicago: Third World Press. Pettigrew, Thomas F. (1959) “Regional Differences in Anti-Negro Prejudice.” Journal of Abnormal and Social Psychology. 59:28-36. Pettigrew, Thomas F. (1964) A Profile of the Negro American . NY: Van Nostrand. Pettigrew, Thomas F., Gordon W. Allport, and Eric 0. Barnett (1958) “Binocular Resolution and Perception of Race in South Africa.” British Journal of Psychology. 49:265-278. Pettigrew, Thomas F. and Roel W. Mertens (1995) “Subtle and Blatant Prejudice in Western Europe.” European Journal of Social Psychology. 25:57-75. Phinney, Jean S. (1992) “The Multigroup Ethnic Identity Measure: A New Scale for Use with Diverse Groups.” Journal of Adolescent Research, 7: 156-176. 169 Powers, Daniel A. and Christopher G. Ellison (1995) “Interracial Contact and Black Racial Attitudes: The Contact Hypothesis and Selectivity Bias.” Social Forces. 74:205- 226. Preston, James D. and Jerry W. Robinson, Jr. (1974) “On Modification of Interracial Interaction.” American Sociological Review, 39: 283-285. Quillian, Lincoln (1995) “Prejudice as a Response to Perceived Group Threat: Population Composition and Anti-Immigrant and Racial Prejudice in Europe.” American Sociological Review. 60:586-611. Ridley, Charles R. (1984) “Clinical Treatment of Non-disclosing Black Clients: A Therapeutic Paradox.” American Psychologist, 39: 1234-1244. Robinson, James Lee, Jr. (1980) “Physical Distance and Racial Attitudes: A Further Examination of the Contact Hypothesis.” Phylon. 41:325-332. Robinson, Jerry W. Jr., and James D. Preston (1976) “Equal-Status Contact and Modification of Racial Prejudice: A Reexamination of the Contact Hypothesis.” Social Forces, 54: 91 1-924. Russell, Marcus (1971) “Erased, Debased, and Encased: The Dynamics of African Educational Colonization in America.” In John W. Blassingame (Ed.), New Perspectives on Blacks Studies. Chicago: Univeristy of Illinois Press. Rokeach, Milton (1973) The Nature of Human Values. NY: Free Press. Rutledge, Aaron L. and Gertrude Z. Gass (1967) Nineteen Negro Men. NY: Jossey-Bass. Schaeffer, Nora C. (1980) “Evaluating Race-of-Interviewer Effects in a National Survey,” Sociological Methods and Research. 8:400-419. Schuman, Howard and Lawrence Bobo (1988) “Survey Based Experiments of White Racial Attitudes Toward Residential Integration.” American Journal of Sociology. 94: 273-299. “‘ Schuman, Howard, and Jean M. Converse (1971) “The Effects of Black and White Interviewers on Black Responses in 1968.” Public Opinion Quarterly. 35:44-68. Schuman, Howard and Shirley Hatchett (1974) Black Racial Attitudes: Tends and Complexities. Ann Arbor: University of Michigan, Institute for Social Research. Schuman, Howard, Charlotte Steeh, and Lawrence Bobo (1985) Racial Attitudes in American: Trends and Interpretations. Cambridge, MA: Harvard. 170 Schuman, Howard , Charlotte Steeh, Lawrence Bobo and Maria Krysan (1997) Racial Attitudes in American: Trends and Interpretations. Revised Edition Cambridge, MA: Harvard. Sears, David O. (1969) “Black Attitudes Toward the Political System in the Aftermath of the Watts Insurrection.” Midwest Journal of Political Science. 12: 515-544. Sears, David O. (1986) “College Sophomores in the Laboratory: Influences of a Narrow Data Base on Psychology’s View of Human Nature.” Journal of Personality and Social Psychology. 51 :515-530. Sears, David O. and Donald R. Kinder (1971) “Racial Tensions and Voting in Los Angeles.” In W.Z. Hirsch, (ed.) Los Angeles: Viability and Prospects for Metropolitan Leadership. NY: Praeger. Sears, David O. and John B. McConahay (1973) The Politics of Violence: The New Urban Blacks and the Watts Riot. Boston: Houghton Mifflin. Sears, David 0., Jim Sidanius, and Larence Bobo (eds.) (2000) Racialized Politics: The Debate About Racism in America. Chicago: Univ. of Chicago Press. Sears, David O., Colette Van Laar, Mary Carrillo.'and Rick Kosterman (1997) “Is it Really Racims: The Origins of White American’ Opposition to Race-Targeted Policies.” Public Opinion Quarterly, 61: 16-53. Sheatsley, Paul B. (1966) “White Attitudes Toward the Negro.” Daedalus. 95: 217-238. Shelton, J. Nicole (2000) “A Reconceptualization of How We Study Issues of Racial Prejudice.” Personality and Social Psychology Review. 4:374-3 90. Sidanius, James, Felicia Pratto, and Lawrence Bobo (1996) Racism, Conservatism, Affirmative Action, and Intellectual Sophistication: A Matter of Principled conservatism or Group Dominance?” Journal of Personality and Social Psychology, 70: 1-15. Sigelman, Lee, Timothy Bledsoe, Susan Welch, and Michael W. Combs (1996) “Making Contact? Black-White Social Interaction in an Urban Setting.” American Journal of Sociology. 101:1306-1332. Sigelman, Lee and Stephen Tuch (1997) “Metastereotypes: Blacks’ Perceptions of Whites’ Stereotypes of Blacks.” Public Opinion Quarterly, 61: 87-101. Sigelman, Lee and Richard G. Niemi (2001) “Innumeracy about Minority Populations.” Public Opinion Quarterly. 65:86-94. Sigelman, Lee and Susan Welch (1993) “The Contact Hypothesis Revisited: Black-White Interaction and Positive Racial Attitudes.” Social Forces, 71: 781-795. 171 Sigelman, Lee and Susan Welch (1991) Black Americans ’ Views of Racial Inequality: The Dream Deferred. Cambridge University Press. Smith, Tom W. and Paul B. Sheatsley (1984) “American Attitudes Toward Race Relations.” Public Opinion Quarterly, 14-15:50-53. Sniderman, Paul M. and Thomas Piazza (1993) The Scar of Race. Cambridge: Harvard University Press. Sniderman, Paul M. and Thomas Piazza (2002) Black Pride and Black Prejudice. Princeton, NJ: Princeton University Press. Sniderman, Paul M., Thomas Piazza, Philip E. Tetlock, Ann Kendrick. 1991. "The New Racism," American Journal of Political Science 35: 423-48. Sniderman, Paul M. and Michael G. Hagen (1985) Race and Inequality: A Study in American Values. Chatham, NJ: Chatham House. Steele, Claude M. and Joshua Aronson (1995) “Stereotype Threat and Intellectual Test Performance of African Americans.” Journal of Personality and Social Psychology. 69:97-11]. Steele, Shelby (I990) The Content of Our Character. NY: St. Martin’s Press. Stephan, Walter G. (1978) “School Desegregation: An Evaluation of Prediction Made in Brown v. Board of Education.” Psychological Bulletin. 85: 217-238. Stouffer, Samuel A. (1955) Communism, conformity, and Civil Liberties. NY: Doubleday. Sullivan, John L., James E. Piereson, and George E. Marcus (1982) Political Tolerance and American Democracy. Chicago: University of Chicago Press. Swim, Janet K., and Deborah Miller (1999) “White Guilt: Its Antecedents and consequences for Attitudes Toward Affirmative Action.” Personality and Social Psychology. 46: 1287-1302. Swim, Janet K. and Charles Stangor ( Eds.) (1998) Prejudice: The Target ’s Perspective. San Diego, CA: Academic Press. Tate, Katherine (1993) From Protest to Politics: The New Black Voters in American Elections. NY: Russell Sage Foundation. Taylor, Dorothy (2000) “Cultural Mistrust and Racial Divides.” Social Pathology: A Journal of Reviews. 6:199-212. Taylor, Marylee C. (1998) “How White Attitudes Vary with the Racial Composition of Local Populations: Numbers Count.” American Sociological Review. 63:512-535. Terrell, Francis, and Sandra L. Terrell (1981) “An Inventory to Measure Cultural Mistrust Among Blacks.” Western Journal of Black Studies, 5: 180-184. Terrell, Francis. Sandra L. Terrell, and Fayneese Miller (1993) “Level of Mistrust as a Function of Education and Occupational Expectations Among Black Students.” Adolescence, 28: 573-578. Thomas, Felicia F ., Joseph J. Leon and Gary A. Cretscr(1985) “Racial Attitudes Toward Miscegenation Laws, 1974 and 1982: A Hypothesis Test and Analysis of Selected Sociodemogaphic Variables.” Western Journal of Black Studies. 9: 32-39. Thompson, Chalmer E., H. Neville, P.L. Weathers, W.C. Poston and DR. Atkinson (1990) “Cultural Mistrust and Racism Reaction Among African-American Students. Journal of College Student Development. 31:162-168. Tolbert, Caroline J. and John A. Grummel (2003) “Revisiting the Racial Threat Hypothesis: White Voter Support for California’s Proposition 209.” State Politics and Policy Quarterly. 3:1 83-202. Tsukashima, Ronald and Darrel Montero (1976) “The Contact Hypothesis: Social and Economic Contact and Generational Changes in the Study of Black Anti-Semitism." Social Forces, 55: 149-165. Tuch, Steven A. and Ronald Weitzer (1997) “Racial Differences in Attitudes Toward the Police.” Public Opinion Quarterly. 61: 642-663. Tucker, Clyde (1983) “Interviewer Effects in Telephone Surveys.” Public Opinion Quarterly. 47:84-95. Watkins, C. Edward, Francis Terrell, Fayneese S. Miller and Sandra L. Terrell (1989) “Cultural Mistrust and Its Effects on Expectational Variables in Black Client-White Counselor Relationships.” Journal of Counseling Psychology, 36: 447-450. Welsing, Frances C. (1991) The Isis Papers: the Keys to the Colors. Chicago: Third World Press. Westie, Frank R. (1953) “A Technique for the Measurement of Race Attitudes.” American Sociological Review, 18: 73-78. Westie, Frank R. (1952) “Negro-White Status Differentials and Social Distance.” American Sociological Review, 17: 550-558. 173 Wilbanks, William (1987) The Myth of a Racist Criminal Justice System. Monterey, CA: Brooks/Cole Publishing Company. Wilkinson, Doris Y. and Ronald L. Taylor (1977) The Black Male in America. Chicago: Nelson-Hall. l Williams, Reginald M. (1964) Stranger Next Door. Englewood Cliffs: Prentice-Hall. Wilson, William J. (1973) Power, Racism, and Privilege. NY: Macmillan. Wojniusz, Helen K. (1979) “Racial Hostility Among Blacks in Chicago.” Journal of Black Studies. 10:40-59. Works, Ernest (1961) “The Prejudice Interaction Hypothesis from the Point of View of the Negro Minority Group.” American Journal of Sociology. 67:47-52. Woodmansee, John and Stuart W. Cook (1967) “Dimensions of Racial Attitudes: Their Identification and Measurement.” Journal of Personality and Social Psychology, 7: 240- 250. Young, Andrew (1996) An Easy Burden: The Civil Rights Movement and the Transformation of America. NY: Harper-Collins. 174 G lllllllllllllllllll