Donate
Independent, objective, nonpartisan research

R 800BCR

Authors

R 800BCR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3742 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(12) "R_800BCR.pdf" ["wpmf_size"]=> string(6) "174271" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(173785) "Ethnic Context, Race Relations, and California Politics ••• Bruce Cain Jack Citrin Cara Wong 2000 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Cain, Bruce E. Ethnic context, race relations and California politics / Bruce Cain, Jack Citrin, and Cara Wong. p. cm. Includes bibliographical references. ISBN 1-58213-051-5 1. California—Ethnic relations—Political aspects. 2. California—Race relations—Political aspects. 3. California— Politics and government—1951. 4. Minorities—California— Political activity. 5. Neighborhood—California. I. Citrin, Jack. II. Wong, Cara, 1970– III. Title. F870.A1 .C33 2000 305.8'009794—dc21 00-040312 Copyright © 2000 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword At some time in the early part of this century, no racial or ethnic group will make up a majority of California’s population. Demographers describe this state of affairs with all the wonderment of astronomers exploring the outer reaches of the universe. Other social scientists, politicians, and planners are more sobered by it. What does this mean for California, the Golden Dream, and life at the neighborhood level? What do we know about how people will get along? Will there be harmony, or will racial or ethnic tensions threaten California’s governance, let alone the smooth running of day-to-day community life? However we imagine this majority minority status and the changes it may bring, California has embarked on a journey that is unprecedented in U.S. history, with all the risk and excitement that such a journey entails. To help map this journey, PPIC has begun a series of research projects to describe this new demographic reality and to study the consequences of ethnic and racial diversity for the formation of public policy. As one of the first contributions to that series, Ethnic Context, Race Relations, and California Politics takes a careful look at ethnic group relations at the neighborhood level. After reviewing precinct-level data on racial attitudes and voting behavior, Bruce Cain, Jack Citrin, and Cara Wong find “no evidence that ethnic group relations are particularly troubled in highly diverse areas or that these areas will serve as cauldrons for future problems.” Even more encouraging is their finding that “where there were differences between whites and nonwhites, they often were in the direction of greater tolerance and agreement between whites and nonwhites in mixed areas than in homogenous ones.” At the same time, the authors found significant ethnic group differences in racial attitudes and voting behavior. They also found that most residents prefer to live in neighborhoods in which members of their own ethnic group constitute a majority. (They note, however, that this iii preference does not necessarily contradict an acceptance of diversity in the workplace, in higher education, or in public life generally.) Finally, the authors predict that “California’s political climate on racial and cultural issues will depend in large part on the perceptions of the more numerous whites who live in majority white areas.” What shapes these perceptions and voting preferences is unclear, but they do not seem to arise from daily interactions with neighbors of different ethnic groups. This study and its companion, How Different Ethnic Groups React to Legal Authority, by Yuen Huo and Tom Tyler, have been supported by PPIC with two objectives in mind. The first is to describe and understand California’s changing demography to anticipate the challenges we will inevitably face. Demographers are fond of saying that demography is destiny, and in California’s case, there is a lot to be learned before we fulfill our destiny. The second objective is to remind Californians that public policy is fundamentally tied to politics at the local level, where tensions between the old and the new are most likely to be felt. These tensions may resolve themselves gradually, smoothly, even indiscernibly. Or the demographic fault lines could shift precariously, perhaps through the workings of either representative or direct democracy. The prudent course is for California’s leaders to be ready for either outcome, and PPIC intends to map that course with reliable research. David W. Lyon President and CEO Public Policy Institute of California iv Summary Immigration has transformed California’s ethnic landscape. Since 1965, the Latino and Asian populations have grown rapidly, and demographers project that by 2030, no ethnic group will constitute more than 50 percent of the state’s population. This transformation has important implications for government policy and electoral politics. Given California’s history, it also raises questions about how much and what kind of ethnic group competition and conflict is likely to occur over the next decades. This increased diversity of California’s population has complicated its ethnic politics. In many cases, racial and ethnic issues cannot be reduced to disagreements between a white majority and a coalition of ethnic minorities. Blacks, Latinos, and Asian Americans may share an interest in anti-discrimination policies, but they compete for political offices and jobs, especially at the local level. Likewise, the main immigrant groups have shared concerns, such as bilingual education, which may not be a priority for blacks. In this context, competition among minority groups over the allocation of public benefits is commonplace, and the explosive growth of the Latino and Asian communities has generated a larger and more complex debate over ethnicity and public policy. In this study, we explore how reactions to ethnic diversity relate to important policy questions. These questions include, but are not limited to, whether government should use ethnicity as a criterion for distributing public benefits. Analyzing a survey of California voters conducted just before the vote on Proposition 209, which forbids the use of such preferences by public agencies in California, we investigate how personal experiences with and attitudes toward other ethnic groups might influence attitudes and electoral choices about racial policies. v Research Design and Methodology The study assumes that one’s ethnic context—which we define as the ethnic composition of one’s neighborhood—affects the frequency and character of one’s interactions with members of different groups. Our main concern is to identify the influence, if any, of such contexts on racial attitudes and behavior. In doing so, we distinguish between two hypotheses. The compositional hypothesis recognizes that people living in ethnically mixed neighborhoods may differ systematically from those who do not. For example, whites who live in such neighborhoods may be younger or more economically vulnerable than those who live in predominantly white neighborhoods. If so, these characteristics could shape their attitudes and behavior more than the ethnic composition of their neighborhoods as such. Thus, the compositional hypothesis is alert to the ways in which ethnic contexts might reflect, rather than shape, attitudes and personal characteristics. In contrast, the contextual hypothesis focuses on the extent to which ethnic context actually shapes attitudes and behavior. Although these hypotheses can be distinguished analytically, in practice they are difficult to disentangle. The fact that people can and do choose their residences—that is, people are not assigned to neighborhoods randomly—creates a causal ambiguity that cannot be ignored or eliminated in any study of this type. Even so, this study differs from previous attempts to understand these contextual effects in several ways. First, the unit of analysis is the consolidated precinct, where some degree of face-to-face contact can be assumed. Second, this study examines the relationships between ethnic context, racial attitudes, and voting patterns on a specific measure, namely, Proposition 209. Third, most reports focus on contact between two ethnic groups, but we consider white reactions to several minority populations. Finally, we analyze the influence of ethnic context on minority group members as well as on whites. The data come from a unique survey of the California electorate conducted just before the vote on Proposition 209 in 1996. The survey asked respondents about Proposition 209, neighborhood diversity, racial attitudes, racial identity, political ideology, personal finances, and education. To determine whether genuine contextual effects are present, vi information about the respondents’ social backgrounds and attitudes are linked to data about their localities. The statistical analyses thus estimate the effect of ethnic context by controlling for other individual-level variables. We deliberately oversampled areas with high concentrations of ethnic minorities to facilitate meaningful comparisons across ethnic groups. We also limited the sample to registered voters to focus more precisely on the political effects of ethnic contexts. Major Findings and Implications The study revealed substantial differences across ethnic groups. Compared to the other ethnic groups, whites were less likely to say that their race or ethnicity was an important part of their political identity. They were also more likely to support Propositions 209 and 187 and to regard affirmative action programs as unnecessary and unfair. These differences cannot be explained by either ethnic context or the other factors we considered, such as party, ideology, or social background. At the same time, there was consensus among the four ethnic groups on a number of questions related to the state’s ethnic composition. A large majority in every ethnic group was either neutral or positive about the effect of people of a different ethnic or cultural group moving into their neighborhood. They also were united in expressing more negative views about the effect of illegal immigrants than any other group. Despite differing attitudes about the need for affirmative action, respondents from all four ethnic groups tended to agree on the egalitarian principles of merit and nondiscrimination as the primary criteria for job promotion and educational opportunity. A majority in every ethnic group seemed to favor the effect of its own group on the neighborhood and to agree that people tended to be happier living and socializing with others of the same background. Whites seemed most worried about the effects of a heavy influx of black residents on property values; blacks were most concerned about the prevalence of hostile attitudes in a largely white community. It would be highly misleading, however, to conclude that Californians favor residential or social segregation. As other studies have shown, people of all ethnic groups indicate that they prefer to live in neighborhoods where members of their own group make up the majority, but not all, of the vii neighborhood’s households. In addition, realism rather than prejudice may underlie the belief that it is easier to socialize with people of similar backgrounds. This “separatist” opinion does not imply that people do not value friendships with members of other ethnic groups. Indeed, the rising rates of intermarriage in California between whites and Latinos and between whites and Asians belie such a claim. Although respondents expressed considerable doubt about the feasibility of achieving a “colorblind society,” the data do not contradict public acceptance of diversity in the workplace, in higher education, or in public life generally. The second major conclusion of this study is that multiethnic neighborhoods are not the primary locus of ethnic political divisions in California. Previous studies have argued for the so-called “threat hypothesis,” which predicts that racial tension will be highest where different groups interact the most. In general, however, the data indicate that perceptions and attitudes varied little by ethnic context. Indeed, where there were differences between whites and nonwhites, they often were in the direction of greater tolerance and agreement between whites and nonwhites in mixed areas than in homogeneous ones. These findings suggest that California’s political climate on racial and cultural issues will depend in large part on the perceptions of the more numerous whites who live in majority white areas. We found that living in an ethnically mixed neighborhood boosted feelings of ethnic and racial identification for all four groups. How such identifications affect political conduct is a more complicated question. From the perspective of minority groups, a stronger sense of racial and ethnic identity may fuel efforts to advance collective goals and redress inequality. Racial and ethnic identifications among minorities were stronger in areas with higher densities of minorities, a finding that suggests that efforts at political organization may be more successful in such neighborhoods. However, foregrounding ethnic identity also may foster backlash and intensify group conflict. The third major conclusion is that attitudes on racial and ethnic issues do not seem to be formed at the local level. The few observed neighborhood variations seem to be compositional rather than contextual: That is, these variations were best explained by differences in the ideologies, party affiliations, and background characteristics of viii respondents. This pattern indicates that racial attitudes are largely acquired through experiences that cut across localities. To use the example of Proposition 209, voting intentions were not influenced by what residents perceived in their local areas so much as by their general political orientations and what they learned from the respective pro- and anti-209 campaigns. This finding suggests that mass media, for example, may be more critical to racial and ethnic attitudes than are local contexts. Our final major conclusion concerns the relatively negative feelings many Californians have about undocumented immigrants. Respondents from all four ethnic groups had distinctively more positive views about legal immigrants than illegal ones. One implication of this finding is that policymakers should be careful not to lump legal and illegal immigrants together in assessing the public support for programs that affect these groups. Policymakers might also emphasize the fact that the legal Latino population is much larger than the undocumented group. In addition to reducing resentment about immigration, this measure might increase the public’s willingness to support policies that aid the larger Latino community. In sum, there is no evidence that ethnic group relations are particularly troubled in highly diverse areas or that these areas will serve as cauldrons for future problems. Racial attitudes seem to be influenced more by one’s political and social background than by one’s local context. Resistance to ethnic diversity is not more likely among those who directly experience it in their neighborhoods. We conclude that ethnic tensions are not preordained to increase as Californians continue to cope with the state’s shifting ethnic landscape. ix Contents Foreword..................................... iii Summary..................................... v Tables ....................................... xiii 1. THE POLITICS OF ETHNICITY IN A CHANGING CALIFORNIA .............................. Purposes of This Study ......................... Residential Patterns ........................... The Motivational Bases of Contextual Effects ........... Methodology ............................... Implications for Public Policy ..................... 1 3 5 7 10 11 2. NEIGHBORHOOD CONTEXT AND REACTIONS TO ETHNIC DIVERSITY ......................... Group Perceptions at the Neighborhood Level .......... Group Perceptions Across Ethnic Contexts ............ The Influence of Subjective Contexts ................ Multilevel Analysis of Context Effects ................ Summary of Findings .......................... 13 14 18 21 23 25 3. RACIAL ATTITUDES AND ETHNIC CONTEXT IN CALIFORNIA .............................. Racial and Ethnic Identification ................... Images of Race Relations ........................ Social Distance .............................. Affirmative Action ............................ Explaining Ethnic Identification and Attitudes Toward Race Relations ............................... Results for Ethnic Identification ................... Results for Beliefs About Race Relations .............. Results for Beliefs About Affirmative Action ........... Summary of Findings .......................... 27 28 31 34 36 39 40 42 43 44 xi 4. ETHNIC CONFLICT, ETHNIC CONTEXT, AND VOTING ON PROPOSITION 209 ................ Racial Threat as an Explanation ................... Accounting for the Vote on Proposition 209 ........... Reviewing the Role of Context .................... 47 48 53 57 5. CONCLUSION ............................. Appendix A. Survey Sample and Data Sources ................... B. Multivariate Analyses from Chapters 2 and 3 ........... C. Survey Question Wording ....................... Bibliography .................................. About the Authors ............................... 61 67 69 77 83 87 xii Tables 2.1. Neighborhood Effect of Different Groups, by Race/Ethnicity of Respondents ................. 2.2. Neighborhood Effect of Immigrants, by Race/Ethnicity of Respondents ........................... 2.3. Neighborhood Effect of Different Groups, by Race/Ethnicity and Context ................... 2.4. Neighborhood Effect of Immigrants, by Race/Ethnicity and Context ............................. 2.5. Neighborhood Diversity, by Neighborhood Effect of Different Groups .......................... 2.6. Neighborhood Diversity, by Neighborhood Effect of Immigrants ............................. 3.1. Racial and Ethnic Identification, by Race/Ethnicity of Respondents ............................. 3.2. Racial and Ethnic Identification, by Race/Ethnicity and Context ................................ 3.3. “Color-Blind Society,” by Race/Ethnicity of Respondents ............................. 3.4. “Color-Blind Society,” by Race/Ethnicity and Context ................................ 3.5. “Minorities Have Less Opportunity,” by Race/Ethnicity of Respondents ........................... 3.6. “Minorities Have Less Opportunity,” by Race/Ethnicity and Context ............................. 3.7. “People Happier with Others of Same Background,” by Race/Ethnicity of Respondents ................. 3.8. “People Happier with Others of Same Background,” by Race/Ethnicity and Context ................... 3.9. “Minorities Gain at Expense of Others,” by Race/ Ethnicity of Respondents ..................... 3.10. “Minorities Gain at Expense of Others,” by Race/ Ethnicity and Context ....................... 16 17 19 20 22 23 29 30 31 32 33 33 35 35 36 37 xiii 3.11. “Minorities Get More Than They Deserve,” by Race/Ethnicity of Respondents ................. 3.12. “Minorities Get More Than They Deserve,” by Race/Ethnicity and Context ................... 3.13. Multivariate Analyses of Racial Attitudes ........... 4.1. Percentage Voting “Yes” on “Ethnic” Initiatives, by Race/Ethnicity ............................ 4.2. Objective Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity .............. 4.3. Perceived Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity .............. 4.4. Percentage Answering “Yes” to Affirmative Action Questions, by Race/Ethnicity and Ethnic Context ..... 4.5. Effects of Race/Ethnicity, Racial Context, and Racial and Ethnic Identification on Proposition 209 Vote ....... 4.6. Explaining Voting Intentions on Proposition 209 ..... B.1. Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 ................ B.2. Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 ................ B.3. Racial Identification and Racial Attitudes—Black Respondents, from Chapter 3 .................. B.4. Racial Identification and Racial Attitudes—Latino Respondents, from Chapter 3 .................. B.5. Racial Identification and Racial Attitudes—Asian Respondents, from Chapter 3 .................. B.6. Racial Identification and Racial Attitudes—White Respondents, from Chapter 3 .................. 37 38 41 48 49 51 52 54 55 70 71 72 73 74 75 xiv 1. The Politics of Ethnicity in a Changing California Immigration has transformed the ethnic composition of Californian society. Since 1965, the Latino and Asian segments of the population have grown rapidly, and demographers project that by 2030, California will be a “majority minority” state.1 The regional concentration and age distribution of California’s population also are changing because of the residential concentration of recent immigrants and because they tend to be younger and tend to have larger families than the mainly white or black native residents.2 These demographic trends have important implications for both government policy and electoral politics. An obvious example is the effect of the influx of non-English speakers on the state’s public schools. The changing ethnic composition of California’s schoolchildren has fueled controversy over bilingual education programs and reshaped affirmative action programs in higher education. On the political front, the growing number of ethnic minorities and their geographic location has an immediate effect on redistricting and reapportionment, with consequences for minority representation in the state legislature and Congress. More generally, ethnic differences in voter turnout and party identification have a direct effect on electoral outcomes. Historically, rapid change in the ethnic composition of California society has engendered group competition and conflict. Beginning with the Gold Rush, the pull of labor demand and the push of poverty led to ____________ 1The term “majority minority” is used to refer to a situation where blacks, Latinos, and/or Asian Americans constitute a majority of the population in an area (and where non-Hispanic whites, therefore, make up less than 50 percent of the population) according to the 1990 Census. 2 In this monograph, we will be referring to the many different racial and ethnic groups in California. We want to clarify that when we use the term “whites,” we are referring to non-Hispanic whites. 1 successive waves of immigration to California from China, Japan, the Philippines, and Mexico. When the influx of new arrivals was particularly heavy and economic times turned bad, communal violence erupted and legislation was passed restricting immigration (Olzak, 1992). Thus, experience suggests that the current demographic changes might create new tensions in ethnic group relations. The very diversity of California’s population, however, complicates the nature of ethnic politics in the state. The assumption that disagreement over public policy generally reflects conflict between a white majority and a coalition of ethnic minorities seems simplistic. The civil rights movement centered on black-white relations and the need to improve conditions for the former group, but the explosive growth of population in Latino and Asian communities has created an expanded and more complex policy agenda. In this context, competition among minority groups over the allocation of public benefits is commonplace. For example, blacks, Latinos, and Asian Americans may share an interest in anti-discrimination policies, but they contend for scarce political offices and jobs, particularly at the local level. Affirmative action in college admissions benefits blacks and Latinos more than Asians. On the other hand, the main immigrant groups have concerns—in the area of bilingual education and the access of newcomers to government services, for example—that may not be a priority for blacks. A significant contrast between the Watts riot of 1965 and the Rodney King riot of 1992 illustrates the growing complexity of ethnic relations in California. The earlier outbreak of violence largely expressed the hostility of black residents toward the predominantly white Los Angeles police force; the more recent eruption revealed antagonism between blacks and Koreans as well as between blacks and whites. The climate of opinion facing recent Americans differs from that facing the earlier waves of newcomers from Europe in a way that may affect ethnic group relations. In the early 20th century, belief in assimilation and Americanization was pervasive among political elites. Today, a “multiculturalist” perspective stressing themes of ethnic identification and group rights challenges the traditional individualist ethos; the question of how government should use ethnicity as a criterion for distributing public benefits is highly controversial. 2 Public opinion influences policy formation in a democratic society, if only indirectly by limiting the range of choices elected officials can safely contemplate. In California, though, the increasing use of direct democracy for deciding important issues magnifies the effect of the views of ordinary citizens. Since the passage of Proposition 13 in 1978, voting on initiatives has determined state policy on taxing and spending, environmental protection, gay rights, gun control, auto insurance, language rights, illegal immigration, and affirmative action. Clearly, one stimulus for direct legislation is an impasse between the governor and the legislators; another source of the recent trend toward government by initiative is the interplay between institutional structures and current demographic developments. The ideological fulcrum in California differs in legislative and initiative politics. Legislative districts are drawn on the basis of population, not citizenship or even legal residency. Thus, immigrants who cannot vote provide additional legislative seats for certain geographic areas. Since these immigrants often cluster in ethnic “enclaves,” representation in the California legislature is weighted toward Democrats and minorities with liberal policy attitudes. On the other hand, the statewide electorate, which decides the gubernatorial race and ballot initiatives, is whiter, older, and wealthier than the population as a whole. Accordingly, outcomes in these contests tend to be more heavily influenced by Republican and conservative voters. This contrast in constituencies helps explain the passage of recent initiatives on illegal immigration, affirmative action, and bilingual education, when voters reversed policies with entrenched legislative support, and it is highly probable that the electoral arena will see continued battles on ethnic issues. Purposes of This Study This study describes public attitudes toward the changing ethnic composition of California and explores how these reactions to ethnic diversity are related to people’s preferences on important policy questions. The vote on Proposition 209, the initiative forbidding the use of racial or gender preferences by public agencies in California, is a useful case study for assessing the effect of the demographic, ideological, and 3 institutional factors briefly described above. Affirmative action uses an applicant’s race or ethnicity as one criterion for allocating places in college, government jobs, and public contracts. A frequent justification for this policy is that fairness requires that the distribution of such benefits should mirror the state’s ethnic diversity. This monograph analyzes a survey of California voters conducted just before the vote on Proposition 209 to investigate how personal experiences with and attitudes toward other ethnic groups influenced both electoral choices and opinions about racial policies. The focus of the empirical inquiry is the influence of one’s personal context—defined as the ethnic composition of one’s neighborhood—on the outlook of each of California’s four main racial and ethnic groups. More specifically: 1. How do subjective experiences of diversity differ across ethnic groups? By comparing people grouped by both ethnicity and residential context, Chapter 2 describes whether group conflict is more or less widespread in ethnically diverse locales. 2. What are Californians’ hopes and expectations concerning race relations? Do they anticipate progress toward a color-blind society with minimal discrimination? How strong is their sense of racial and ethnic group identity, and how does this influence the policy preferences of different groups? Chapter 3 assesses the strength of ethnic group identifications and the influence of these feelings on policy positions of Californians living in different residential contexts. 3. Does a strong sense of racial or ethnic identity have electoral consequences, boosting support for Proposition 209 among whites while enhancing opposition among minority groups? Chapter 4 addresses this issue and investigates whether contact with other ethnic groups, as indexed by one’s residential context, influenced voting on the ballot initiative and whether this effect was similar across ethnic groups. 4 Residential Patterns In a state as large and varied as California, there is no one typical residential mix. The state as a whole has a high level of ethnic diversity, but regional and local profiles vary widely. Some counties reflect the diversity of the state as a whole; others are virtually all white. Similarly, residential segregation persists within diverse areas; within Los Angeles County, for example, the Palos Verdes Peninsula, the Santa Monica Mountains, and the newer northern cities such as Santa Clarita are relatively homogeneous. In California, as nationally, ethnic heterogeneity at the aggregate level tends to mask uniformity at the level of tracts or blocks (Clark, 1996). Using precinct-level information derived from the University of California, Berkeley’s, Statewide Database, we determined that 86 percent of whites in California live in neighborhoods where more than half of the residents share their racial background (majority white areas). By contrast, 29 percent of the state’s blacks live in areas with a majority of blacks, and 44 percent of its Latinos live in majority Latino areas. To the extent that people live in neighborhoods populated by a majority of their own racial or ethnic group, more diversity at the state level may not mean a high level of intergroup contact in one’s daily life. Ethnic homogeneity in residential neighborhoods is not a matter of law but of personal choice and financial ability. In statistical terms, whites are the ethnic group in California most likely to live among people of the same background. Studies show that both whites and blacks would prefer not to live as an isolated minority in their neighborhoods, and that living with a majority of one’s own race is the dominant preference in both groups (Farley et al., 1994). Among whites, this preference rests partly on stereotypes of blacks that engender the belief that as the ratio of blacks to whites increases, property values decline. Among blacks, a reluctance to move into mainly white neighborhoods partly reflects fear of a hostile reception. In other words, attitudes toward one’s own and other ethnic groups influence the composition of neighborhoods by shaping individual choices to move in or out. Therefore, how people react to their neighborhood’s ethnic mix is an important element in the changing pattern of California politics. 5 In this study, the term “context” refers to the composition—ethnic or economic—of a particular area. Depending on one’s analytic purpose, the contextual unit could be the precinct, census tract, county, state, or even nation. The focus here is on the ethnic context of the neighborhood, measured by the relative proportions of the four main ethnic groups in California. Context can thus refer to the overall proportion of all minority groups, to the proportion of a particular group such as blacks or Latinos, or simply to whether or not a locality is a “majority minority” area. This study assumes that ethnic contexts affect the frequency and character of interactions between members of different groups. Its main concern is to identify the effect of the makeup of one’s locality on racial attitudes and behavior. In a statistical analysis delineating these effects, if any exist, on the outlook of individuals of a particular race or ethnicity, it is necessary to distinguish between compositional and contextual hypotheses. The compositional interpretation recognizes that whites living in majority minority locales may differ systematically from those who live in mainly white areas in ways that influence racial attitudes. They may be younger, for example, and therefore perhaps more liberal, or they may be economically vulnerable, and therefore more likely to scapegoat minority group members. More generally, prior attitudes and stereotypes can influence one’s residential choices, so that racially tolerant whites are willing to remain in an area experiencing an influx of minorities while their more prejudiced counterparts take flight. To the extent that people vote with their feet in this way, differences across ethnic contexts might simply reflect variation in the demographic and attitudinal characteristics of the particular ethnic group members living there. The alternative contextual interpretation searches for effects that go beyond the aggregation of individual-level differences. This hypothesis asserts that the residential setting itself influences the outlook of those who experience it, whatever their prior predispositions or background. The notion of a contextual effect thus emphasizes that individuals are not just autonomous units but rather are shaped by social interaction. Ethnic context is an aggregate-level concept that depends on the characteristics of individuals in a particular area, but its influence on behavior is 6 separate from the effects of these background factors on these individual actors. The power of cultural norms is one mechanism for the influence of the local community. Social interactions condition the flow of information, reinforcing some beliefs and undermining others through exposure and conformity pressures. For example, a Democrat living in a community composed mainly of partisans will tend to find her views shared by her neighbors, whereas a Republican in the same area may modify his views to fit in better with the majority. To the extent that one self-selects friends and social contacts, the psychological tendency to prefer similar and reinforcing views will dominate one’s interactions, but the opportunities for such self-selection are limited to some degree by the externally imposed context. Finding a fellow-conservative in North Berkeley, for example, is not easy. The Motivational Bases of Contextual Effects Prior research on the political effect of ethnic context has centered on white responses to the size of the black population in their community. Following V. O. Key’s famous study of the ‘black belt” in the South, a conventional hypothesis is that the larger the proportion of blacks, the more hostile or prejudiced the political outlook of white residents (Key, 1949; Blalock, 1967; Glaser, 1994). Large concentrations of blacks are said to represent a threat to whites because of ethnic competition for economic, political, and cultural power. Numbers matter because voting and market power help determine the allocation of resources across ethnic groups. Accordingly, large concentrations of blacks make the economic and political threat more acute and visible, heightening feelings of vulnerability among whites and increasing their antipathy to minority political demands. In other words, as minority populations in their immediate context grow in size, the resulting diverse ethnic context increases the likelihood of defensive and prejudiced reactions among whites. Transporting Key’s “threat” hypothesis from the South to California involves broadening the definition of ethnic context to refer to the concentrations of Latino and Asian as well as black residents. However, theoretical discussions (Taylor, 1998) generally suggest that 7 concentrations of other minorities will yield the same kind of antipathy among whites as would a larger share of blacks. This means that the more diverse contexts within California should be characterized by a heightened sense of ethnic consciousness among all groups and by political polarization along ethnic lines. Clearly, the “threat” hypothesis implies pessimism about the possibility of achieving harmony in a multiethnic society. An alternative hypothesis with some support in social psychology (Allport, 1954; Prentice and Miller, 1999) is the more optimistic “contact” theory, which proposes that direct experience with members of other groups tends to ease conflict and promote mutual tolerance and understanding. This occurs partly through the erosion of prior stereotypes and partly because the necessities of peaceful coexistence help develop norms of compromise and power-sharing where there are large numbers of minority group residents. If the processes identified by the “contact” theory prevail, then ethnically diverse contexts should be characterized by more positive images of race relations and greater support among whites for policies designed to assist minorities, such as affirmative action. Research indicates that contact between racial groups reduces conflict and prejudice only under special conditions (Stephan, 1987). Interdependence, common goals, equal status, and encouragement by authorities are necessary for intergroup contact to promote tolerance and goodwill. The mere presence of a large number of minority residents in a locality does not guarantee these conditions. Nor does it assure intimate contact with someone of a different background, particularly when the contextual unit is large. Previous studies regarding the relationship between the numerical size of the black community and racial hostility among whites also give mixed support for the threat hypothesis. For example, Voss (1996) concludes that the size of the black population in one’s locality did not boost white voting for David Duke in Louisiana. By contrast, analysis of the 1990 national General Social Survey (Taylor, 1998) found that traditional prejudice and opposition to race-targeted policies among whites in metropolitan areas swell as the local black share of the population expands. However, concentrations of local Asian American 8 and Latino populations did not engender white antipathy toward these groups (Taylor, 1998; Hood and Morris, 1997). These inconsistent results may reflect differences among particular localities (Southern counties versus metropolitan areas nationwide), in the extent of white flight across these localities, or in the dependent variables analyzed (racial attitudes and policy positions versus votes). The fact that people can choose their ethnic context rather than fall into a locality more or less randomly, say by the accident of birth, creates a causal ambiguity complicating analyses of the effect of demographic diversity in today’s California. “Threat” promotes a positive association between the prevalence of minorities and prejudice, whereas “white flight” predicts a negative correlation between them, with prejudice more widespread in all-white areas (Voss, 1996). Unless these offsetting processes can be disentangled, which is difficult with cross-sectional data, the interpretation of results can only be tentative. This is an important caveat for the present study also. However, the research reported here differs from previous studies of contextual influences in several ways. First, the unit of analysis is the neighborhood (i.e., specifically, consolidated precincts), a relatively small area where one can assume face-to-face contact (although not necessarily real social interaction). Second, we examine the effect of ethnic context on both racial attitudes and voting on a specific measure affecting the distribution of public benefits among ethnic groups, namely, Proposition 209. Third, as noted above, we consider white reactions to several distinct ethnic minority populations. Finally, we are almost unique in analyzing the influence of ethnic context on the conduct of minority group members as well as whites. The threat hypothesis applies only to the reactions of the dominant group to ethnic change. Theoretically, the size of minority groups in a given area should have quite different consequences for the political behavior of racial and ethnic minorities, representing an opportunity rather than a danger. A heavy concentration of Latinos, for example, should improve their capacity for political mobilization. The density of social networks and the frequency of communications facilitate the development of a distinctive ethnic consciousness and the formation of organizations to advocate on behalf of group interests. On the other 9 hand, where an ethnic group is a small and scattered minority, as are blacks in many parts of California, group assertiveness should be more difficult, both psychologically and organizationally. Methodology This study addresses two basic questions: the nature of ethnic group differences in racial attitudes and political preferences in California, and the influence of local contexts on politics in a multiethnic society. The research design is tailored to these purposes and to overcome several methodological problems identified by previous analyses of contextual influences. The most significant problem is the need for a multilevel analysis based on the merger of micro-level and macro-level information. Confining comparisons to differences across contexts does not allow one to choose between the contextual and compositional hypotheses outlined above because there can be no full adjustment for the effects of the distribution of individual-level attributes on electoral choices. To identify contextual effects, therefore, one must be able to link survey information about particular individuals with census information about the respondents’ localities. Another difficulty relates to the frequent tendency to define context at the county, state, or national levels of aggregation. This is problematic because the territorial unit that denotes a psychologically meaningful context, regardless of whether one adheres to the “threat” or the “contact” hypothesis, may be smaller than the state or even the county; aggregation can mask meaningful variation. In addition, statewide samples typically underrepresent minority group members, particularly the poor and less-educated ones. To address these methodological issues, we took the data for this study from a unique survey of the California electorate conducted just before the vote on Proposition 209 in 1996. Respondents were sampled from a stratified set of contexts, and to determine whether genuine context effects are present, information about respondents’ social background and attitudes are linked to data about their localities. The crucial statistical analyses thus estimate the effect of ethnic context, controlling for theoretically relevant individual-level variables. Because of a primary interest in the implications of California’s increasing ethnic 10 diversity on group relations, we focus on the potential effect of the ethnic composition of neighborhoods rather than on the influences of other contextual attributes such as economic profiles. Hence, information about economic context was used only to confirm that any contextual effects detected were not spurious. Our sampling strategy used the electoral precinct (a good proxy for local neighborhood) as the jurisdiction defining a context. We deliberately oversampled areas in California with a high concentration of ethnic minorities to facilitate meaningful comparisons across ethnic groups (see Appendix A for details about the survey sample and data sources.) Finally, to focus attention more precisely on the political effect of ethnic contexts, we limited the sample to the registered voters who have been shaping official policy toward ethnic group relations through voting on initiatives such as Proposition 209. The remaining chapters of this monograph proceed as follows. We first describe the relationships between respondents’ ethnicity and residential context and their racial attitudes and voting intentions separately. We then report the results of a multivariate analysis incorporating both individual-level and contextual variables to determine whether or not any apparent contextual effects reflect more than differences in the ethnic composition of localities. These analyses are conducted for the sample as a whole and then for each ethnic group separately. Implications for Public Policy Policymakers in a diverse community have a broad interest in forging consensual solutions to tensions arising from ethnic group interaction and competition. The demographic trend toward increasing ethnic diversity in California is unlikely to be reversed. Neither is the existence of widespread variation in the degree of ethnic heterogeneity across residential locales likely to be reversed. The public opinion data examined here identify the fundamental reactions toward ethnic diversity among the varied racial and ethnic groups making up the state. To what extent is the presence of a growing minority population in California perceived as a threat, and by whom? Are attitudes in multiethnic contexts more or less hostile or ethnocentric? Answers to these questions 11 can provide clues about popular support for policies concerning immigration, housing, or local education by locating the sources of intergroup conflict, both geographically and demographically. The existence of contextual influences on reactions to members of other ethnic groups may indicate the effect of direct contact and personal experience, as opposed to that of media communications, on opinion formation. The nature of the relationship between ethnic diversity and group conflict may indicate not only where to concentrate efforts to alleviate racial and ethnic tensions, but also which mechanisms are likely to be effective in fostering harmony and goodwill. For example, one might learn how to structure programs for community organization, immigrant integration, and public service delivery at the local level. We discuss the policy implications of the study’s findings in the concluding chapter. 12 2. Neighborhood Context and Reactions to Ethnic Diversity Increasingly, California’s changing demography is bringing members of different racial and ethnic groups into more personal contact with one another. This is not uniformly the case, as we noted above, because some parts of California have become much more diverse while others have remained relatively homogeneous. The design of this study tries to capture the heterogeneity of neighborhood settings to discover whether local experiences make a difference in racial and ethnic attitudes. We begin with relations between groups and their perceptions of one another. Group perceptions lie at the core of race relations. When racial and ethnic groups view each other in a hostile manner, it erodes the basis for harmonious racial policies and makes building consensus difficult, if not impossible. People become aware of different racial and ethnic groups when people of varying backgrounds move into their neighborhoods, perhaps for the first time. In the course of daily life, they observe their neighbors’ comings and goings. Children from various families intermingle at school and play. Neighbors confront common problems such as traffic, drugs, crime, or the appearance of their surroundings. These shared concerns often engender discussion and group meetings to forge common solutions. Social interchange of this kind informs and alters stereotypes and evaluations of members of other racial and ethnic groups. Context affects learning by shaping direct experience and firsthand observation. Of course, group perceptions are not solely determined by what happens at or near one’s home. People from different racial and ethnic backgrounds also interact while commuting, shopping, working, or studying. Because many Californians do not work near their homes, the ethnic context of workplace relationships may differ from those in their 13 residential neighborhoods. Which setting is likely to influence racial attitudes is therefore an empirical question. For example, it might be that job and school experiences are more relevant than neighborhood experiences to people’s beliefs about affirmative action, since those are the locales where one might have encountered or heard about racism or “reverse discrimination.” In addition to direct learning about other ethnic groups in the neighborhood or at work and school, there are important indirect, nonexperiential sources of attitudes, such as the mass media, popular culture, and reference groups such as political parties. Television and the movies are an obvious source of imagery about other groups. For example, if movies typically cast blacks as criminals, whites as victims, and Asians as shopkeepers, they may create or reinforce stereotypes about people one rarely encounters. Political messages similarly can transmit messages about what a particular ethnic group is like. The question, then, is whether learning at the neighborhood level, indexed in part by the ethnic composition of the locality, matters, given the well-documented strength of early socialization, the media, and the larger political culture. Does living in a diverse area and being exposed to people of different racial and ethnic backgrounds foster positive or negative perceptions of other groups? The answers are important because if neighborhood learning matters and the number of diverse areas is growing, the content of local interactions will play a larger role in determining the quality of race relations in California. The nature of contextual influences also can indicate to policymakers whether they should target the more diverse areas as potentially problematic for ethnic harmony or whether it is more effective to address race relations in a more global way (e.g., such as through changing media images or general education). Group Perceptions at the Neighborhood Level To explore group perceptions, respondents in our survey were asked a series of questions about the effects that different groups of people have had on the quality of life in the neighborhoods where they live. In particular, they were asked to say whether the presence of particular groups improved, made worse, or had no effect on their neighborhoods. 14 The groups in question were whites, blacks, Latinos, Asians, legal immigrants, and illegal immigrants. We also asked our survey respondents to indicate whether they themselves lived in an area that was becoming more diverse, less diverse, or staying about the same. As a general observation, we found that California’s racial and ethnic groups were more likely to hold positive than negative perceptions of one another. Given some of the highly publicized racial conflicts in this state over the last decade, this was not a foregone conclusion. At the time of our survey California had only recently emerged from a serious economic recession, and contentious ballot measures dealing with ethnic issues (i.e., Propositions 187 and 209) had generated substantial protest and political turmoil. Nevertheless, among all four major racial and ethnic groups, negative images of ethnic diversity were very limited. Consider the evidence of Table 2.1, which examines beliefs about how blacks, Latinos, Asian-Americans, and whites, respectively, affect the quality of life in one’s neighborhood. No more than 10 percent in any ethnic group felt that the presence of blacks made neighborhood conditions worse, whereas over 89 percent thought that it made no difference or improved conditions. Overall, perceptions of the neighborhood effect of whites and Asians were even less negative, and perceptions of Latinos only marginally more critical. The most common view of Californians toward the presence of other ethnic groups in one’s neighborhood was neutral, not prejudiced. These data also reveal a general tendency of minority respondents to perceive the neighborhood effect of their own racial and ethnic group most positively. For example, blacks were twice as likely as Asians or Latinos, and almost three times as likely as whites, to say that the presence of more blacks had improved neighborhood conditions. The findings in Table 2.1 suggest that Asians are the most ethnocentric group, with over 30 percent saying that the presence of fellow Asians improved their neighborhoods. Only whites deviated from the norm of preferential assessment of their own group’s effect, perhaps because group identity is less salient for members of the dominant ethnic group (Prentice and Miller 1999). White respondents also were somewhat less likely to view the presence of minority groups in their neighborhoods positively. The second column of Table 2.1, for example, indicates that 15 16 Whites (n = 603) Blacks (n = 217) Latinos (n = 408) Asians (n = 270) Table 2.1 Neighborhood Effect of Different Groups, by Race/Ethnicity of Respondents (responses in percent) Effect of Blacks Im- No Made proved Effect Worse 11 81 8 28 62 10 13 81 6 15 79 7 Effect of Latinos Im- No Made proved Effect Worse 14 74 13 21 66 13 28 63 9 13 80 7 Effect of Asians Im- No Made proved Effect Worse 18 74 8 17 79 4 20 75 6 33 65 3 Effect of Whites Im- No Made proved Effect Worse 17 80 3 19 74 7 19 76 5 24 72 4 the ratio of favorable to unfavorable assessments of the effect of Latinos among black and Asian respondents was about two to one, compared to an even split among whites. In the next chapter, we report additional evidence that people tend to prefer living among others with the same ethnic background. So, although California’s ethnic groups do not oppose diversity at the local level, a high level of heterogeneity and integration is not necessarily their preferred choice. The notable exception to the finding of more positive than negative perceptions of a group’s effect on the neighborhood concerns beliefs about illegal immigrants. Table 2.2 reports responses to questions about the effect of both legal and illegal immigrants on the quality of neighborhood life, with answers broken down by the racial and ethnic identity of the respondent. In a familiar pattern, legal immigrants enjoyed approximately a three to one advantage in positive to negative perceptions, whereas opinion about the effect of illegal immigrants ran two to one in the opposite direction. This balance of opinion prevailed in all ethnic groups, including Latinos and Asians, despite the fact that the majority of the undocumented population in California is made up of these groups, especially Latinos. Here it should be noted that respondents in this study are registered voters who necessarily have lived in the United States at least five years themselves (for reasons of citizenship status) and may be less sympathetic to the illegal newcomers. Still, the predominant evaluation of illegal immigrants, as for all other groups, is neutral, not hostile. Table 2.2 Neighborhood Effect of Immigrants, by Race/Ethnicity of Respondents (responses in percent) Whites (n = 603) Blacks (n = 217) Latinos (n = 408) Asians (n = 270) Effect of Legal Immigrants Im- No Made proved Effect Worse 19 73 8 21 71 8 25 68 7 27 69 5 Effect of Illegal Immigrants Im- No Made proved Effect Worse 7 65 27 9 70 21 12 70 18 9 69 23 17 Group Perceptions Across Ethnic Contexts The next question to address is whether perceptions of other ethnic groups, which differ little by the respondent’s ethnicity, vary across ethnic contexts. Specifically, is the balance of opinion about a group’s influence on the neighborhood more negative in ethnically diverse than ethnically homogeneous areas? The relevant evidence is displayed in Tables 2.3 and 2.4. These tables examine both the respondent’s race and the type of neighborhood that he or she lives in. That is, they distinguish between blacks in majority black areas, blacks in majority minority areas, blacks in majority white areas, and so on. The dominant result in this array of statistics is that responses did not vary in a dramatic way by neighborhood. Overall, people were as positive about the effect of other ethnic groups in the mixed areas as they were in the homogeneous areas. Take, for instance, perceptions of the effect that blacks have on the neighborhood. Table 2.3 shows that contrary to the threat hypothesis, there was no more antipathy to blacks among whites in majority minority than majority white areas. The same is true when we compare the opinions of Asians in majority minority areas and Asians in majority white areas. For the most part, the differences were small for Latinos and for blacks as well. If anything, blacks in majority white areas tended to be relatively more negative toward the presence of their own group (and other minorities) in their neighborhoods than were blacks in majority black or majority minority areas. But, once again, the modal response for all settings and groups was to think that the presence of other groups, including illegal immigrants, does not, per se, make the neighborhood either better or worse off. The slight tendency of whites in majority minority areas to have more positive perceptions of racial and ethnic minority groups than whites in majority white areas is worth noting. However, this could be due to compositional differences in their party affiliation or ideology as much as to context-driven experiential learning. If, for example, Republican and conservative areas are more homogeneously white than Democratic areas, then it could be this partisan difference rather than the “whiteness” of a neighborhood that underlies the differences in the 18 Table 2.3 Neighborhood Effect of Different Groups, by Race/Ethnicity and Context (responses in percent) 19 Effect of Blacks Effect of Latinos Effect of Asians Effect of Whites Im- No Made Im- No Made Im- No Made Im- No Made proved Effect Worse proved Effect Worse proved Effect Worse proved Effect Worse Blacks in majority black precincts (n = 139) 31 60 8 21 67 11 16 79 5 20 75 5 Blacks in majority minority precincts (n = 57) 29 58 14 27 54 19 20 76 4 18 68 14 Blacks in majority white precincts (n = 21) 5 85 10 5 85 10 14 86 0 10 85 5 Latinos in majority Latino precincts (n = 147) 14 82 5 35 58 8 25 71 4 22 75 3 Latinos in majority minority precincts (n = 135) 12 79 9 30 63 8 18 73 10 16 78 6 Latinos in majority white precincts (n = 126) 13 83 4 19 71 10 16 82 3 19 75 6 Asians in majority minority precincts (n = 145) 15 76 9 14 77 9 37 61 2 27 70 3 Asians in majority white precincts (n = 125) 13 82 5 13 84 4 28 70 3 21 74 4 Whites in majority minority precincts (n = 236) 13 74 13 19 67 14 22 67 12 19 77 4 Whites in majority white precincts (n = 367) 9 86 5 11 78 12 15 79 6 16 81 3 Table 2.4 Neighborhood Effect of Immigrants, by Race/Ethnicity and Context (responses in percent) Blacks in majority black precincts (n = 139) Blacks in majority minority precincts (n = 57) Blacks in majority white precincts (n = 21) Latinos in majority Latino precincts (n = 147) Latinos in majority minority precincts (n = 135) Latinos in majority white precincts (n = 126) Asians in majority minority precincts (n = 145) Asians in majority white precincts (n = 125) Whites in majority minority precincts (n = 236) Whites in majority white precincts (n = 367) Effect of Legal Effect of Illegal Immigrants Immigrants Im- No Made Im- No Made proved Effect Worse proved Effect Worse 23 70 7 8 69 22 21 65 15 12 70 19 11 89 0 6 78 17 36 57 7 14 62 24 22 71 8 11 72 17 16 79 5 9 78 12 29 66 6 9 61 29 25 72 3 8 78 15 26 64 10 9 58 33 15 78 7 6 70 24 outlook of majority white and majority nonwhite areas. Similarly, it is unclear whether racial or ethnic minorities living in majority white areas have a lower opinion of their own ethnic group moving in because of some sort of self-selection process or because they are conforming to the dominant viewpoint among the white residents in the area. Another hint of learning deriving from the diversity of the neighborhood is that in almost all cases, those who lived in majority minority areas were less likely to be neutral about the presence of a given group in their neighborhood. Consider, for example, the white respondents. Living in a majority minority area increased the percentage of those with either a positive or negative view of blacks by about 12 20 points, of Latinos by about 11, and of Asians by about 12. The same pattern holds for black, Asian, and Latino respondents. We reported earlier that images of illegal immigrants tended to be more negative than responses elicited by questions about other groups. How does this vary by neighborhood context? Here, the more dramatic effects do suggest a pattern of negative learning. For all racial groups, including Latinos, people had more negative opinions about illegal immigrants if they lived in ethnically mixed areas than if they resided in largely white areas. Whether this opinion is founded on sensitivity to heightened economic competition or some other negative reaction, hostility to illegal immigrants seems stronger in precincts where they are more likely to live. The Influence of Subjective Contexts An alternative to defining context in terms of the current ethnic composition of a neighborhood is to focus on whether the context is changing. Advocates of the threat hypothesis (e.g., Green et al., 1999; Taylor, 1998) often adopt this approach. In the absence of trend data on the precincts, however, one can measure neighborhood context in subjective terms and compare neighborhoods that are viewed as becoming more diverse with those that are perceived as staying the same or becoming less diverse. A perception of rapid change, whatever the empirical reality, may engender a stronger sense of anxiety than would occur in areas with inter-racial stability. As actual demographic trends in California imply, about 36 percent of the respondents in our sample thought their area was becoming more diverse, compared to just 4 percent who perceived declining ethnic diversity. Whites and Asians (40 percent) were most likely to perceive their areas as changing in this way, and Latinos (28 percent) were the least likely to think so. Tables 2.5 and 2.6 show how beliefs about the changing composition of one’s neighborhood were linked to evaluations of the neighborhood effect of particular groups. A now familiar pattern emerges. The effect of perceived change in the character of the neighborhoods by the major racial and ethnic groups was relatively slight. By small margins, the images of Asians and Latinos were more negative 21 Table 2.5 Neighborhood Diversity, by Neighborhood Effect of Different Groups (responses in percent) Neighborhood less diverse (n = 56) Neighborhood about the same (n = 858) Effect of Blacks Improved/ Made No Effect Worse 90 10 Effect of Latinos Improved/ Made No Effect Worse 90 10 94 6 92 8 Effect of Asians Improved/ Made No Effect Worse 96 4 96 4 Effect of Whites Improved/ Made No Effect Worse 92 8 96 4 22 Neighborhood more diverse (n = 512) 90 10 85 15 91 9 95 5 Table 2.6 Neighborhood Diversity, by Neighborhood Effect of Immigrants (responses in percent) Neighborhood less diverse (n = 56) Neighborhood about the same (n = 858) Neighborhood more diverse (n = 512) Effect of Legal Immigrants Improved/ Made No Effect Worse 87 13 95 5 91 9 Effect of Illegal Immigrants Improved/ Made No Effect Worse 71 29 82 18 69 31 among respondents in the areas believed to be changing. But otherwise, perceived changes in the diversity of one’s neighborhood did not seem to alter perceptions of the groups a great deal. The pessimistic assumption that the increasing number of minorities in California is resulting in more hostile attitudes at the local level is not borne out here. Once again, however, respondents singled out illegal immigrants for criticism (Table 2.6). People in changing areas (those perceived as becoming either more or less diverse) were much more likely than those in stable neighborhoods to think that illegal immigrants make the quality of life in their neighborhood worse. Multilevel Analysis of Context Effects Our analysis so far has explored whether perceptions vary by neighborhood context, but not whether the differences observed reflect genuine contextual influences rather than compositional variation in the makeup of localities. To address this central issue, we need to estimate a statistical model that includes the following attributes of individual respondents as well as measures of their ethnic context as predictors: 1. Party and ideology, since the Republican party and conservatives generally tend to take a harder stance on immigration and affirmative action; 23 2. Sex, age and education, since these factors are often related to differences in racial and ethnic tolerance; 3. Perceptions of the state’s and the respondent’s personal financial situations, since those who are economically vulnerable may feel more threatened by members of other ethnic groups or by immigrants; and 4. Feelings of racial or ethnic identification, since those with a higher level of group consciousness to begin with may have more well-defined views about the presence of other racial and ethnic groups. The complete results of the analysis, including the statistical estimates, are presented in Appendix B. As a general conclusion, we found that the relatively minor neighborhood differences in beliefs about the effect of other groups reflected compositional rather than contextual effects. Controlling for the respondent’s background and political outlook, we found no statistically significant effect for the size of the minority group population in a neighborhood. This held true whether we measured context dichotomously as majority minority or majority white, continuously in terms of the proportion of black, Latino, Asian, or total minority population, or subjectively in terms of perceptions of ethnic change. The only consistent evidence of a contextual effect—and this, too, might be interpreted as a compositional factor—concerned the perceptions that people had of the neighborhood effect of their own groups. When asked about the effects of blacks on the neighborhood, blacks living in an area where they were the majority ethnic group tended to view the effect of more blacks in the neighborhood more favorably than their counterparts residing in mainly white areas. The same pattern held true among Latino and Asian respondents, suggesting that contextual influences may be specific to particular groups rather than generalized. It seems that favoritism toward one’s own group was stronger among those living in areas heavily populated by people with the same racial or ethnic background, either because of self-selected residential location or because the general tendency toward ethnocentrism (Tajfel, 1978) is reinforced by interacting with one’s own 24 group. Either way, where one lives seems to reveal something about a level of comfort with one’s own ethnic group that is not simply explained by partisan orientation, ideology, or social background. The equations summarized in Appendix B indicate that economic conditions affected perceptions of Latinos and Asians more than beliefs about blacks. This confirms earlier findings about the influence of economic adversity in stimulating opposition to immigration (Citrin et al., 1997), a factor that helps explain the passage of Proposition 187. In the present survey, respondents who perceived either the state as a whole or themselves to be in a worse economic situation than a year ago were more likely to see the presence of Asians and Latinos as having an unfavorable effect on their neighborhoods. Also, among Latino and Asian respondents, a strong sense of in-group identification was associated with a less-favorable attitude toward other ethnic groups. The models with perceptions of legal and illegal immigrants as dependent variables also contain some new information. Republicans and respondents who resided in neighborhoods perceived as becoming more diverse seemed to have the most negative view of the illegal immigrants. Whites in majority minority areas, controlling for party identification, also seemed to have more negative views—and this may actually be an instance of experiential learning. As for legal immigrants, other things being equal, respondents from all groups who lived in majority minority areas had more favorable views regarding their effect on the neighborhood. On the other hand, respondents who felt economically vulnerable had more negative views of legal as well as illegal immigrants, as did those with a stronger sense of racial identification. Summary of Findings Overall, California’s racial and ethnic groups had similar views about the effects of ethnic diversity on their neighborhoods. For all target groups, the dominant reaction was that their overall effect on one’s neighborhood was neutral. Beyond this, the ratio of perceived positive to negative effects was favorable, usually by a margin of at least two to one. The only exception was the negative image of illegal immigrants. In this case, the balance of opinion was reversed, and every ethnic group in the 25 sample felt that illegal immigrants were making conditions in their neighborhood worse rather than better. Perhaps predictably, white respondents were less positive about the effect of more ethnic diversity than were minority group members themselves, suggesting the existence of a discernible, though not deep, ethnic divide on this issue. More generally, even with multiple statistical controls, all ethnic groups expressed more positive views about their own group than any other. Last, on the central matter of contextual influences, there was no evidence of distinctively negative learning in ethnically diverse areas. The results of this initial test of the threat hypothesis are largely negative. 26 3. Racial Attitudes and Ethnic Context in California The expectation that sizable minority populations increase white hostility assumes that people experience the presence of a different ethnic group in their locality as a diffuse threat (Liska et al., 1985; Giles and Buckner, 1993; Quillian, 1996; Wright, 1977; Voss, 1996). This inference often is made without direct measures of perceived threat or other attitudes that may mediate between the individual’s environment and political behavior. The evidence presented in Chapter 2 indicated only minor contextual influences on perceptions of how ethnic change was affecting the quality of neighborhood life. This chapter explores the effects of context on normative beliefs about race relations and on the strength of ethnic identifications. These orientations may get at feelings of threat and anxiety, and thus may potentially link ethnic context and voting on Proposition 209, a relationship examined directly in Chapter 4. How do Californians evaluate the current state of ethnic group relations in society and what do they think of the policies frequently proposed for dealing with the issue? What are the ethnic differences in these views and how are they affected by residential context and feelings of group identification? For example, are people living in ethnically mixed neighborhoods more or less optimistic about achieving the ideal of a color-blind society? Are they more or less likely to agree that people are happier when living with their own group? How does context relate to the strength of people’s sense of racial and ethnic identity? We explore whether mixing with members of other racial and ethnic groups sensitizes one to the problem of discrimination or leads one to downplay its existence. Finally, we examine whether the most important divisions on these issues are between whites and nonwhites, between blacks and nonblacks, or between all the ethnic groups separately. 27 Once again, the order of presentation is to show contextual and individual-level differences separately, and then to report the results of multivariate models. To preview the results, our analyses indicate only minor effects of neighborhood context on racial attitudes, but also point to an indirect influence of context on beliefs about discrimination and social interactions, mediated through people’s feelings of identification with their particular ethnic group. Racial and Ethnic Identification One building block for beliefs about racial policies is feelings of identification with one’s own racial or ethnic group. A strong sense of group consciousness should foster support for measures favoring one’s own group. Accordingly, a pervasive sense of group consciousness should promote a common outlook on political issues visibly linked to the group’s status. But at the same time, a strong sense of identification with one’s own group often engenders hostility toward outsiders, and this poses a problem for an ethnically diverse society such as that of the United States (Brown, 1986). Indeed, an important element of the integrationist ideal is that people should not think of themselves primarily in racial or ethnic terms when they are acting politically. By this reasoning, a society is more likely to be race-neutral or color-blind in law and practice if people do not think that racial or ethnic groups are the primary vehicle for pursuing their interests. People can become aware of the political relevance of their racial or ethnic identity for many reasons: because the actions of others toward them are based on racial prejudice or bias; because other groups organize themselves on racial or ethnic lines; or because the state recognizes racial and ethnic categories in its programs and policies. Among the multiple stimuli in California’s political environment that could heighten or lessen racial and ethnic political identity, the focus here is the role of neighborhood context. Respondents in our survey were asked the following question about their personal political identity: “When it comes to social and political matters, some people think of themselves mainly as black, white, Latino, Asian, or Jewish, and that is very important to how they think of themselves. Other people don’t give much thought to these things. 28 When it comes to social and political matters, how important is your race and ethnicity to how you think of yourself?” Opinions about racial and ethnic identity divided whites and nonwhites (see Table 3.1). Two-thirds of white respondents said that their race and ethnicity were “not very important” or “not at all important” when considering social or political matters. By comparison, two-thirds of Asians and Latinos and 74 percent of blacks said that their racial and ethnic identity was “very important” or “somewhat important.” Of the minority groups, blacks expressed the most intense sense of ethnic group identification: Among them, but not among Latinos or Asians, a majority said that their racial identity was “very important.” These findings are consistent with the results of research based on national surveys, which also showed stronger and more widespread feelings of ethnic group identification among minorities than among whites (Wong, 1998). Previous studies indicate that the extent of group consciousness has political consequences (Tate, 1993; Dawson, 1994). Identifying strongly with one’s ethnic group increases the likelihood that one will mobilize and participate on behalf of that group. Identification also affects policy attitudes and vote choices. Looking at the effects of race and neighborhood context together (see Table 3.2), we find that for blacks, context had only a small effect on whether respondents emphasized a racial identity.1 But for Latinos and Table 3.1 Racial and Ethnic Identification, by Race/Ethnicity of Respondents (in percent) Whites (n = 580) Blacks (n = 209) Latinos (n = 386) Asians (n = 263) Very Important 15 56 42 30 Somewhat Important 20 18 24 35 Not Very Important 27 12 15 19 Not at All Important 39 14 19 16 ____________ 1Because few blacks in our sample live in majority white areas, it is difficult to make any generalizations with confidence based on this subgroup. 29 Table 3.2 Racial and Ethnic Identification, by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 135) Blacks in majority minority precincts (n = 54) Blacks in majority white precincts (n = 20) Latinos in majority Latino precincts (n = 141) Latinos in majority minority precincts (n = 125) Latinos in majority white precincts (n = 120) Asians in majority minority precincts (n = 142) Asians in majority white precincts (n = 121) Whites in majority minority precincts (n = 227) Whites in majority white precincts (n = 353) Very Somewhat Not Very Not at All Important Important Important Important 60 19 7 13 54 19 15 13 35 10 35 20 45 21 15 19 48 27 11 14 33 26 18 23 28 37 22 14 34 34 15 17 20 20 25 35 11 19 29 41 whites, particularly, living in a majority minority precinct led to a stronger sense of racial and ethnic identity than in other types of precincts. Specifically, 75 percent of Latinos who lived in majority minority precincts thought their race and ethnicity was “very” or “somewhat” important when it came to political and social matters, as opposed to 59 percent of Latinos in majority white precincts. For whites, a sense of racial identity was much weaker, but there was some variation across contexts: 40 percent of whites in majority minority precincts said their race was important when they decided political issues, whereas only 30 percent of whites in majority white precincts felt this way. In sum, living among racial minorities—regardless of whether one is white or nonwhite—does seem to promote a greater sense of ethnic identification among some groups. 30 Images of Race Relations Respondents were asked several questions about the current state of race relations in the United States. One item asked whether they agreed or disagreed with the following statement: “In our lifetime, America can become a color-blind society.” A majority of all ethnic groups felt this would not occur in the near future (see Table 3.3)2 : 52 percent of whites, 62 percent of blacks, 54 percent of Latinos, and 56 percent of Asians disagreed with the statement. The relatively greater pessimism of the black respondents is consistent with other data in our poll and with previous studies showing that perception of ongoing discrimination is more widespread among blacks (Uhlaner, 1991). Still, this survey item sets a very high standard for overcoming ethnic divisions in society; the fact that about 40 percent of the sample felt that a “color-blind” society could emerge quite soon points to the presence of a substantial degree of optimism about progress in race relations. Furthermore, opinion about the possibility of achieving a color-blind society was largely unaffected by neighborhood (see Table 3.4); only for Asians was there a statistically significant contextual effect. Forty-two percent of Asians living in majority minority precincts agreed that America could become a color-blind society in the near future, compared to 31 percent of those living in majority white precincts. One Table 3.3 “Color-Blind Society,” by Race/Ethnicity of Respondents (in percent) Whites (n = 593) Blacks (n = 210) Latinos (n = 399) Asians (n = 261) Disagree Strongly 27 47 36 33 Disagree Somewhat 25 16 18 23 Neither Agree Nor Disagree 5 6 6 7 Agree Somewhat 25 15 19 20 Agree Strongly 19 17 22 17 ____________ 2Implicit in the interpretation of this statement as a sign of optimism or pessimism is the idea that a color-blind world is an ideal; however, it is possible that “pessimism” may in fact reflect simple pragmatism. 31 Table 3.4 “Color-Blind Society,” by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 134) Blacks in majority minority precincts (n = 56) Blacks in majority white precincts (n = 20) Latinos in majority Latino precincts (n = 143) Latinos in majority minority precincts (n = 131) Latinos in majority white precincts (n = 125) Asians in majority minority precincts (n = 139) Asians in majority white precincts (n = 122) Whites in majority minority precincts (n = 231) Whites in majority white precincts (n = 361) Disagree Strongly 45 54 40 39 36 34 33 33 26 27 Disagree Somewhat 16 13 25 16 18 19 19 28 23 26 Neither Agree Nor Agree Agree Disagree Somewhat Strongly 6 16 18 5 13 16 5 15 15 4 16 25 6 20 20 6 21 20 7 24 18 8 16 15 4 26 20 5 24 18 interpretation of this is that assimilation into white society may not be so smooth for Asians in California as is often assumed. Another question about progress in race relations asked about equal opportunity. When given the statement, “Members of certain ethnic or racial groups still have less opportunities to get ahead than other people,” a majority of all racial groups, including whites, agreed on the presence of continuing discrimination (see Table 3.5). Ethnic minorities, and blacks in particular, were more likely to perceive a continued lack of equal opportunity “to get ahead.” These results are consistent with the previously noted opinion about the prospect of quickly achieving a colorblind society. There was no significant variation by neighborhood context for this item (see Table 3.6), and neighborhood experiences may 32 Table 3.5 “Minorities Have Less Opportunity,” by Race/Ethnicity of Respondents (in percent) Whites (n = 593) Blacks (n = 215) Latinos (n = 401) Asians (n = 263) Agree Strongly 34 66 51 48 Agree Somewhat 26 15 20 24 Neither Agree Nor Disagree 7 2 5 8 Disagree Somewhat 15 7 10 11 Disagree Strongly 19 10 15 9 Table 3.6 “Minorities Have Less Opportunity,” by Race/Ethnicity and Context (in percent) Agree Strongly Blacks in majority black precincts (n = 138) 67 Blacks in majority minority precincts (n = 56) 61 Blacks in majority white precincts (n = 21) 71 Latinos in majority Latino precincts (n = 145) 50 Latinos in majority minority precincts (n = 134) 53 Latinos in majority white precincts (n = 122) 52 Asians in majority minority precincts (n = 142) 47 Asians in majority white precincts (n = 121) 49 Whites in majority minority precincts (n = 231) 40 Whites in majority white precincts (n = 362) 30 Agree Somewhat 16 14 10 18 19 23 25 22 23 28 Neither Agree Nor Disagree 2 2 5 6 5 2 8 8 5 8 Disagree Somewhat 8 7 5 10 10 9 10 12 14 15 Disagree Strongly 7 16 10 17 12 15 10 8 19 19 33 therefore be less relevant to perceptions about opportunities than the work or school environment. It often is assumed that since Americans tend to endorse the principle of racial equality, perceptions of ongoing discrimination will lead to greater support for policies intended to benefit racial minorities. Still, our failure to detect contextual effects suggests that neighborhoodlevel interactions with members of other ethnic groups do not play a role in how one views access to opportunities for all Californians. Social Distance A third question about ethnic group relations was more concerned with social behavior than politics, asking respondents whether they agreed or disagreed that “people of different ethnic and racial groups are generally happier when they live and socialize with others of the same background.” One cannot tell if people treated this as a prescriptive or descriptive statement. Nevertheless, there was a general consensus on the benefit of living with people of the same ethnicity, a preference that might contribute to self-selection as a factor in promoting homogeneous neighborhoods. Table 3.7 shows that 66 percent of whites agreed, as did 57 percent of blacks, 61 percent of Latinos, and 70 percent of Asians. One possibility is that it is the most recent immigrants who most prefer living “with one’s own kind.” If that is the case here, then as their length of residence in California grows, the tendency of Hispanic and Asian residents to express this preference might diminish. (Nevertheless, since the respondents in the survey were registered voters, they have all lived in the United States at least five years.) The only evidence of a contextual effect that was statistically significant was for Latinos (see Table 3.8): 65 percent of Latinos in majority Latino areas felt that people were happier with others of their own background, compared to 57 percent of those living in majority white areas. Given the normative emphasis on the value of diversity in contemporary political culture, it is somewhat surprising that a majority of all racial groups think that an ethnically homogeneous residential environment is related to happiness. This is not the socially desirable response, and may imply a recognition of the psychological obstacles 34 Table 3.7 “People Happier with Others of Same Background,” by Race/Ethnicity of Respondents (in percent) Whites (n = 573) Blacks (n = 207) Latinos (n = 393) Asians (n = 263) Agree Strongly 35 32 36 43 Agree Somewhat 31 24 24 27 Neither Agree Nor Disagree 13 14 10 8 Disagree Somewhat 11 14 16 12 Disagree Strongly 9 16 13 10 Table 3.8 “People Happier with Others of Same Background,” by Race/Ethnicity and Context (in percent) Agree Strongly Blacks in majority black precincts (n = 132) 35 Blacks in majority minority precincts (n = 55) 27 Blacks in majority white precincts (n = 20) 30 Latinos in majority Latino precincts (n = 141) 42 Latinos in majority minority precincts (n = 131) 40 Latinos in majority white precincts (n = 121) 26 Asians in majority minority precincts (n = 140) 44 Asians in majority white precincts (n = 123) 42 Whites in majority minority precincts (n = 228) 35 Agree Somewhat 23 26 30 23 20 31 27 26 31 Neither Agree Nor Disagree 11 22 15 10 8 13 9 8 12 Disagree Somewhat 14 11 20 14 18 16 10 15 11 Disagree Strongly 17 15 5 11 15 14 10 9 11 35 facing the development of more intimate and friendly inter-group relations. Affirmative Action An obvious possibility is that the perception of racial threat would lead whites to oppose affirmative action, a policy explicitly designed to increase the number of minorities in desirable positions. One question in our survey asked respondents whether they regarded affirmative action programs as having a zero-sum quality. Whites (50 percent) were most likely to agree with the statement that “The more good jobs and places in college provided to minorities, the fewer there are for people who are not members of those groups.” A majority of blacks and Latinos disagreed, and Asians were evenly split. The differences among minority groups were relatively small, though, and opinions were not affected by the racial context in which respondents lived (see Tables 3.9 and 3.10). A second question concerning affirmative action is more pointed, in that it asks for a judgment about whether minorities are getting an unfair advantage. We asked respondents whether they agreed or disagreed with the following statement: “Members of particular ethnic or racial groups use special programs to get more benefits than they deserve.” Here, a majority of Latinos (51 percent) and Asians (55 percent), as well as whites (56 percent), agreed with this criticism of affirmative action (see Table 3.11). However, a majority (53 percent) of black respondents disagreed, a possible reflection of their stronger conviction that Table 3.9 “Minorities Gain at Expense of Others,” by Race/Ethnicity of Respondents (in percent) Whites (n = 576) Blacks (n = 204) Latinos (n = 390) Asians (n = 256) Disagree Strongly 18 33 27 24 Disagree Somewhat 19 20 23 18 Neither Agree Nor Disagree 13 10 11 15 Agree Somewhat 27 18 22 25 Agree Strongly 23 19 17 18 36 Table 3.10 “Minorities Gain at Expense of Others,” by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 131) Blacks in majority minority precincts (n = 53) Blacks in majority white precincts (n = 20) Latinos in majority Latino precincts (n = 137) Latinos in majority minority precincts (n = 131) Latinos in majority white precincts (n = 122) Asians in majority minority precincts (n = 139) Asians in majority white precincts (n = 117) Whites in majority minority precincts (n = 226) Whites in majority white precincts (n = 350) Disagree Strongly 32 30 45 30 26 24 25 22 22 15 Disagree Somewhat 17 23 30 18 24 28 22 15 18 20 Neither Agree Nor Agree Agree Disagree Somewhat Strongly 12 18 21 11 21 15 0 10 15 15 23 15 12 20 19 7 24 18 12 22 19 18 28 17 12 25 23 14 28 23 Table 3.11 “Minorities Get More Than They Deserve,” by Race/Ethnicity of Respondents (in percent) Whites (n = 579) Blacks (n = 210) Latinos (n = 391) Asians (n = 251) Disagree Strongly 17 32 28 19 Disagree Somewhat 15 21 14 14 Neither Agree Nor Disagree 12 8 7 12 Agree Somewhat 25 17 24 29 Agree Strongly 31 22 28 26 37 minorities still do not have an equal opportunity to get ahead. Again, there were no contextual effects affecting this pattern of results (see Table 3.12). To the extent that ethnic context captures variation in social contact with minority group members, then, white attitudes toward the state of race relations in this country are unaffected by such interchange. An important overall finding about California opinion is the coexistence among whites of a general recognition that prejudice and a lack of equal opportunity persist and a lack of strong support for affirmative action, a policy intended to alleviate the effects of these problems. Table 3.12 “Minorities Get More Than They Deserve,” by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 133) Blacks in majority minority precincts (n = 56) Blacks in majority white precincts (n = 21) Latinos in majority Latino precincts (n = 139) Latinos in majority minority precincts (n = 132) Latinos in majority white precincts (n = 120) Asians in majority minority precincts (n = 129) Asians in majority white precincts (n = 122) Whites in majority minority precincts (n = 229) Disagree Strongly 27 38 48 27 26 31 21 16 16 Disagree Somewhat 23 16 24 9 17 16 14 15 17 Neither Agree Nor Agree Agree Disagree Somewhat Strongly 10 16 24 5 21 20 5 10 14 10 25 29 4 21 33 8 25 21 12 23 30 12 35 22 12 21 35 38 Explaining Ethnic Identification and Attitudes Toward Race Relations Determining the meaning of the reported differences in outlook about race relations requires additional analysis. For example, past research has shown that blacks are much more likely to be Democrats than are whites; that the longer Latinos are in the United States, the more likely they are to be Democrats; and that the pattern of party affiliation varies across Asian subgroups (Cain et al., 1991). In other words, what appears to be ethnic group differences may simply reflect the influence of partisanship. There thus is a need to control for the compositional factors underlying group differences in outlook, while keeping in mind the possibility that a strong sense of ethnic identity may itself contribute to choosing a particular party affiliation. In our sample of Californians, 46 percent of the whites were Democrats, compared to 84 percent of the blacks, 68 percent of the Latinos, and 37 percent of the Asians. Party identification was significantly related to all the racial attitude items at the bivariate level, with Democrats more likely to perceive persistent discrimination and to regard affirmative action in positive terms. Ideology and educational attainment—to name only a couple of other influences—also were related to several of the racial attitude items. For this reason, it is necessary to control for these individual factors in a multivariate model to determine whether ethnic background or ethnic context has an independent effect on how people perceive race relations in the United States. As in Chapter 2, our statistical models therefore included the following variables as predictors (and controls): 1. Demographics: gender, age, and education; 2. Political predispositions: party identification and ideological self-identification; 3. Perceptions of diversity in the respondent’s neighborhood; 4. Economic outlooks: perception of whether the state economy has gotten better or worse over the last year and perception of whether the respondent’s own personal financial situation has gotten better or worse over the last year; 39 5. Ethnic context; and 6. Racial and ethnic identification, when this item is not itself the dependent variable of interest. The models are estimated in two different ways: in the first set of analyses, we employed the total sample, and included as predictors dummy variables for race and ethnicity and a continuous measure of the percentage of minorities living in a precinct as a measure of ethnic context (see Table 3.13). In the second set of equations, separate models were run for each race separately (see Appendix B).3 Results for Ethnic Identification Because group identification is an individual predisposition that possibly precedes a number of different political attitudes, the only variables included as predictors in the model for racial and ethnic identification are demographic factors and ethnic context. The main finding is that blacks, Latinos, and Asians were all more likely than whites to feel that their racial and ethnic identification was important, even when all other factors were held constant. Also, as we saw earlier, ethnic context turned out to be a significant predictor of group consciousness; the greater the percentage of minorities in one’s precinct, the more likely one was to express a strong sense of racial or ethnic identification. Living in areas with greater concentration of minorities did heighten one’s racial and ethnic political identity, regardless of whether one was white or a minority resident. Another notable finding is that the more educated respondents were less likely to think that their racial or ethnic identification was important when it came to social and political matters. When separate models were run for whites, blacks, Latinos, and Asians, the results indicate nuances in the underpinnings of the racial and ethnic identification of the separate groups. For whites and blacks, ethnic context was a significant predictor; the greater the percentage of minority residents in the area where they lived, the more likely they were to think their racial and ethnic identity was politically important. ____________ 3We also ran these models with other measures of ethnic context, as in Chapter 2, and obtained the same results. 40 Table 3.13 Multivariate Analyses of Racial Attitudes (entire sample, OLS regression) 41 % minority in precinct Black Hispanic Asian Gender Education Age Party identification Ideology Neighborhood diversity Neighborhood effect of blacks Neighborhood effect of Latinos Neighborhood effect of Asians Neighborhood effect of illegal immigrants Neighborhood effect of legal immigrants Neighborhood effect of whites Personal financial situation State economic situation Racial identifiication Constant Adjusted R–squared n Racial Identification B SE –0.47** 0.12 –0.89** 0.10 –0.66** 0.08 –0.70** 0.09 0.07 0.06 0.07* 0.03 0.04 0.04 2.80** 0.15 0.16 1,321 Minorities Color-Blind Have Less Society Opportunity B SE B SE 0.05 0.23 –0.47* 0.21 –0.58** 0.21 –0.47* 0.19 –0.38* 0.16 –0.41** 0.15 –0.25 0.18 –0.46** 0.16 –0.12 0.12 –0.02 0.11 –0.23** 0.05 –0.13** 0.05 0.04 0.08 0.01 0.07 0.20** 0.08 –0.01 0.07 –0.06 0.08 0.16* 0.07 –0.25* 0.11 –0.03 0.10 0.07 0.23 0.25 0.21 –0.35# 0.21 0.04 0.19 0.30 0.27 0.11 0.24 0.00 0.16 0.35* 0.14 0.28 0.26 –0.07 0.23 –0.14 0.33 –0.03 0.29 –0.11 0.08 –0.02 0.07 0.10 0.08 0.33** 0.07 0.03 0.05 0.11* 0.05 3.83** 0.64 1.33* 0.58 0.05 0.11 726 728 Happier with Others of Same Background B SE –0.08 0.21 0.38* 0.19 0.14 0.15 –0.02 0.16 0.18 0.11 0.07 0.05 0.00 0.07 –0.03 0.07 0.03 0.07 0.05 0.10 –0.27 0.21 –0.02 0.19 0.02 0.24 –0.25# 0.14 0.11 0.23 0.28 0.29 0.00 0.08 0.05 0.07 0.10# 0.05 1.75** 0.58 0.00 709 Minorities Minorities Get Gain at Ex- More Than pense of Others They Deserve B SE B SE –0.41# 0.22 –0.10 0.22 –0.32 0.20 –0.53** 0.20 –0.51** 0.15 –0.38* 0.16 –0.11 0.17 0.09 0.18 –0.16 0.11 –0.10 0.12 –0.17** 0.05 –0.26** 0.05 –0.06 0.07 –0.10 0.08 0.08 0.07 0.02 0.08 0.13# 0.08 0.12 0.08 –0.06 0.10 0.18# 0.11 –0.06 0.22 0.25 0.22 0.20 0.20 0.36# 0.21 0.18 0.25 0.28 0.26 0.35* 0.15 0.42** 0.16 0.23 0.24 –0.11 0.24 0.14 0.30 0.08 0.31 0.17* 0.08 0.02 0.08 0.04 0.07 0.03 0.07 –0.06 0.05 0.04 0.05 2.45** 0.61 2.10** 0.63 0.07 0.08 710 719 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Among Latinos, the effect of racial context and education approached conventional levels of statistical significance: The more educated respondents and those living in precincts with fewer minorities were less likely to emphasize their ethnic identification than their counterparts with less education or living in less predominantly white neighborhoods. Results for Beliefs About Race Relations Race, but not ethnic context, was a significant predictor in the multivariate models of attitudes about the general state of race relations, even after multiple controls. When asked about the possibility of a colorblind country in the near future, blacks and the well-educated were more likely to be dubious than were white respondents or those with less education, respectively. Democrats were less likely to envisage the creation of a color-blind world than Republicans, and respondents who thought their personal finances had worsened over the last year were more pessimistic than those with improving circumstances. Finally, respondents who thought that their neighborhoods were becoming more ethnically diverse were less likely to agree that the United States will be a color-blind society in the near future than those who saw their neighborhoods as unchanging or becoming less diverse. Objective ethnic context was not a significant factor, but the perception that one’s context was changing was associated with less optimistic views about racial progress. When separate models were run for each of the four racial groups, the results were varied. Among whites, as in the sample as a whole, Republicans were more likely to say that America could become a colorblind society. Among blacks, the better-educated respondents were more pessimistic about this prospect, a possible sign of relative deprivation and frustration among the upwardly mobile in this ethnic group. Among Latinos, the more-educated, the conservatives, and those who saw their neighborhoods as more diverse were more pessimistic. Among Asians, the older were more optimistic than the young, and the better-educated were more skeptical than those with less schooling. Similar results were obtained for the item concerning equal opportunity for minorities. The better-educated respondents, blacks, Latinos, and Asians all were more likely to believe that minorities had 42 fewer opportunities to get ahead than were the less-educated and white respondents. Group identification was a significant factor here: The greater the political importance of one’s racial or ethnic identity, the greater the agreement that minorities had fewer opportunities than whites. Conservatives and people who thought the state economy had gotten worse over the last year tended to disagree with the idea that minorities had limited opportunities. Among white, Latino, and Asian respondents, those who thought the state economy had gotten worse over the last year were more likely to believe that equal opportunity for minorities already existed. And Latinos and Asians lacking a strong sense of group identity also were more likely to feel this way. One’s sense of racial or ethnic identity was a significant predictor of attitudes about living in an ethnically diverse context. The stronger the sense of ethnic group consciousness—in the sample as a whole and among whites, blacks, and Latinos—the more likely a respondent would agree that people were happier when living and socializing with others of the same background. Interestingly, even controlling for racial and ethnic identification, blacks remained more likely than whites to disagree that people were happier when living apart. So, even though blacks were the most likely to express strong feelings of ethnic consciousness, they were also more in favor of social and residential mixing than any other ethnic group. Results for Beliefs About Affirmative Action Turning now to the underpinnings of attitudes about affirmative action, the multivariate analysis revealed the persistence of some social group differences after the imposition of controls. Specifically, black, Latino, and well-educated respondents were less likely to view affirmative action as a zero-sum policy in which benefits were taken from whites and allocated to minorities. Conversely, Republicans and self-identified conservatives were more likely to regard affirmative action as a matter of some groups winning and others losing. Among white respondents, the less-educated, Republicans, and conservatives again were more likely to think affirmative action meant that minorities gained at the expense of others. Among blacks, however, 43 conservatives and those who saw their neighborhoods as becoming more diverse were more likely to disagree that jobs and school placements for minorities took away opportunities for whites. Among Latino respondents, the better-educated and those lacking a strong sense of racial and ethnic identification were also inclined to disagree with the zero-sum portrayal of affirmative action. Among Asians, women, liberals, and younger respondents were less likely to think minorities gained at the expense of others. Finally, when asked whether minorities deserved the benefits of government programs, again, the better-educated, liberal, black, and Latino respondents were less likely to think that minorities got more than they deserved. On the other hand, respondents who perceived their neighborhoods as becoming more diverse were more likely to believe that minorities got more benefits than they should. Looking at the results for just the white respondents reveals the familiar finding that liberals and the well-educated were less likely to view affirmative action as an unfair reward for the undeserving. Education had a similar effect on the opinions of blacks and Asians. Among Latinos, respondents who felt that the state economy had gotten worse over the last year were also more likely to disagree that minorities got more than they deserved. Summary of Findings The more general racial attitudes discussed in this chapter were all significantly related to voting intentions on Proposition 209 and are potentially related to other more concrete policy questions that arise from election to election. Determining the underpinnings of these beliefs therefore helps identify the sources of more directly political conduct. The principal findings of this chapter are as follows. There are significant ethnic differences in the salience and strength of racial or ethnic identification. Minority group members consistently expressed a stronger sense of ethnic identity than whites. In addition, ethnic context made a difference for whites, blacks, and Latinos, with respondents living among sizable minority populations expressing a stronger sense of racial or ethnic identification. 44 A majority of respondents doubted that America could soon become a color-blind society, although this opinion was most widespread among blacks and least pervasive among whites. A majority of all ethnic groups conceded that members of certain racial and ethnic groups still had fewer opportunities to get ahead, but white and minority respondents were much more divided on whether minority gains from affirmative action were unfair and came at the expense of whites. Most people in the survey seemed to think that people were happier when living among their own ethnic groups, and those who had a stronger sense of racial or ethnic identification felt this way about selfsegregation most strongly. Taken together with earlier evidence that respondents were largely neutral about the effect of other ethnic groups moving into their neighborhood, this suggests that Californians tolerate ethnic diversity without feeling strong enthusiasm for it. In a multivariate analysis controlling for party affiliation, ideology, and other factors, significant differences between whites and nonwhites remained, with a smaller gap between whites and Asians than between whites and Latinos or blacks. In addition to race, education was a strong predictor of racial attitudes. Education diminished the strength of one’s racial or ethnic identification, but, confirming the results of previous studies (Schuman et al., 1997), more educated respondents were more pessimistic about the current state of race relations in the country and more positive about how affirmative action policies work. Finally, the analysis revealed few contextual influences on the racial beliefs examined. Aside from the effect of ethnic context on feelings of racial or ethnic identity, the size of the minority population in one’s neighborhood did not have an independent effect on people’s more general racial attitudes. 45 4. Ethnic Conflict, Ethnic Context, and Voting on Proposition 209 Proposition 209 targeted the ethnic and gender preferences embedded in California’s affirmative action programs for public employment, college admissions, and government contracts. As such, this ballot initiative plausibly can be regarded as attractive to people who feel a sense of racial threat. Analyzing the vote on Proposition 209 thus provides an important case for determining the political implications of California’s ethnic diversity. The passage of Proposition 209 in November 1996 was one of several occasions on which California voters have employed direct democracy to alter policies providing tangible or symbolic benefits to ethnic minorities. As long ago as 1964, they approved an initiative, later ruled unconstitutional, overturning the Rumford Fair Housing Act. In 1986, voters passed an English-only amendment to the state constitution. In 1994, they overwhelmingly adopted Proposition 187, a measure designed to restrict the access of illegal immigrants to most state services. And 1998 brought the victory of Proposition 227, an initiative requiring the elimination of most established bilingual education programs in public schools. Ethnic differences in voting for these initiatives were marked, although varying in magnitude. The Los Angeles Times exit polls (see Table 4.1) showed that 62 percent of whites, compared to 27 percent of blacks, 32 percent of Latinos, and 45 percent of Asian voters, supported the California Civil Rights Initiative (Proposition 209). In the case of Propositions 63 (English-only), 187 (illegal immigration), and 227 (bilingual education), however, the exit polls showed that majorities of white and Asian voters were favorable, whereas the yes vote among 47 Table 4.1 Percentage Voting “Yes” on “Ethnic” Initiatives, by Race/Ethnicity Voting Group Proposition 63a Proposition 187b Proposition 209c English Illegal Affirmative Only Immigration Action Total 68 59 55 Whites 72 83 62 Blacks 67 55 27 Latinos 39 31 30 Asians 58 55 45 aSOURCE: California Exit Poll, November 1986. bSOURCE: California Exit Poll, November 1994. cSOURCE: California Exit Poll, November 1996. dSOURCE: California Exit Poll, November 1998. Proposition 227d Bilingual Education 61 67 48 37 57 Latinos always was less than 40 percent. A majority of black voters supported Propositions 63 and 187, but not 209 or 227. In sum, there is a consistent tendency of whites to be most favorable toward initiatives that eliminate benefits targeted toward ethnic minorities, but the pattern of voting among blacks, Latinos, and Asians varies from issue to issue. Racial Threat as an Explanation One interpretation of ethnic differences in voting emphasizes the role of group competition for jobs, power, and status. As spelled out in Chapter 1, this theory holds that the dominant ethnic group in a community experiences the influx of minorities as a visible threat to its established position and responds with defensive actions designed to shore up its economic and cultural advantages (Quillian, 1995; Tolbert and Hero, 1996). The relative size of minority groups in one’s community is assumed to measure the intensity of group competition and, therefore, the extent of the perceived collective threat to the interests of whites, the numerical majority and culturally dominant ethnic group. The presumed causal chain, then, is from the objective ethnic mix of one’s environment to the subjective experience of threat and anxiety to 48 political polarization. The alternative contact theory proposes a different path—from residential diversity to racial tolerance to compromise and consensus. Our survey was conducted just a week before the vote on Proposition 209. The stated voting intentions anticipated the pattern of ethnic differences in voting that ultimately emerged (see Table 4.2). A majority of whites said they would vote for the measure and a majority of minority respondents were opposed. At first glance, however, the racial threat prediction that whites in neighborhoods with large minority populations would be more likely to vote for Proposition 209 was not confirmed. Of the whites living in districts where they were the majority, 61 percent said they would vote for Proposition 209, compared to 56 percent of those living in the ethnically heterogeneous majority Table 4.2 Objective Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity Voting Group Percentagea Whites in majority minority precincts 56 Whites in majority white precincts 61 Total whites 59 Blacks in majority black precincts 26 Blacks in majority minority precincts 14 Total blacks 22 Latinos in majority Latino precincts 35 Latinos in majority minority precincts 28 Latinos in majority white precincts 36 Total Latinos 33 Asians in majority minority precincts 34 Asians in majority white precincts 40 Total Asians 37 a“Don’t know” responses are excluded from this analysis. 49 minority districts.1 One interpretation of these findings is that whites, as a group, did perceive affirmative action as a threat to their collective interests, but that this belief did not vary across residential environments. An alternative view is that whites living in majority minority areas have adapted to their environment and become more accepting of liberal racial policies. Latinos and Asians in majority minority precincts, although not blacks, were more likely to oppose Proposition 209 than those in majority white precincts. This finding, as others reported above, suggests that the numerical dominance of one’s own ethnic group in a neighborhood may facilitate the development of a sense of group consciousness and identity. Living as a small minority within the mainstream, by contrast, can foster more conformity to the dominant group’s norms. When we consider subjective reactions to ethnic diversity more directly, the contextual differences among white voters predicted by the threat hypothesis do emerge. Those perceiving their neighborhood as becoming more diverse were 10 percent more likely to favor Proposition 209 than those perceiving no change. Similar differences appear when one compares whites with negative and favorable or neutral opinions about the effect of minorities or immigrants on the quality of life in their neighborhoods (see Table 4.3). This held true for whites in both majority white and majority minority precincts, indicating a potential rise in political tensions in the aftermath of rapid ethnic change (Green et al., 1999). The two sides of the campaign over Proposition 209 disagreed about whether affirmative action programs still were needed in California and ____________ 1When queried in the survey, 15 percent of the respondents said they were still undecided about their vote. For clarity of interpretation, these respondents are excluded from the models reported here. One possibility is that these respondents were not genuinely undecided but rather pro-209 voters unwilling to express an anti-minority opinion to interviewers. Examining the social background and political views of these “don’t know” respondents, however, showed that in terms of age, education, party affiliation, ideology, and general racial attitudes, the aggregate position of this group fell in between respondents with definite pro and anti voting intentions, respectively. In addition, we repeated the analyses with these respondents included, thereby creating a three-point dependent variable, using a multinomial logit model. There were no significant differences in the results and no change in the nature of the observed contextual effects. 50 Table 4.3 Perceived Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity Question Do you think your neighborhood is becoming more ethnically diverse, less ethnically diverse, or is about the same? Whitesa Less More Diverse or Diverse Same Blacks Less More Diverse Diverse or Same Latinos Less More Diverse Diverse or Same Asians Less More Diverse Diverse or Same 64b 54 14 27 34 30 30 32 51 Please tell me whether (blacks/Latinos/Asians/ whites) have improved, made worse, or had no effect on the quality of life in your neighbohood. Made Worse Improved or No Effect Improved Improved Made or No Made or No Worse Effect Worse Effect Made Worse Improved or No Effect 65 56 26 22 30 32 39 31 Please tell me whether (illegal immigrants/legal immigrants) have improved, made worse, or had no effect on the quality of life in your neighborhood. 70 54 24 22 42 29 33 31 aFor whites, n = 516; for blacks, n = 180; for Latinos, n = 347; and for Asians, n = 229. b64 percent refers to the percentage of whites who believe their neighborhood is becoming more ethnically diverse and who voted for Proposition 209. whether such programs resulted in qualified applicants for jobs being passed over in favor of undeserving minorities. Our survey posed these questions to respondents. Predictably, supporters of the initiative were less likely to say that affirmative action programs were still needed and more likely to report knowing someone who had personally suffered as a result of group preferences. Here, too, differences in racial and ethnic background, but not in one’s ethnic context, affected these beliefs about affirmative action (see Table 4.4). Whites in majority white neighborhoods were a little less inclined to say that blacks still needed Table 4.4 Percentage Answering “Yes” to Affirmative Action Questions, by Race/Ethnicity and Ethnic Context Group Whites in majority minority precincts Whites in majority white precincts Total whites Blacks in majority black precincts Blacks in majority minority precincts Total blacks Latinos in majority Latino precincts Latinos in majority minority precincts Latinos in majority white precincts Total Latinos Asians in majority minority precincts Asians in majority white precincts Total Asians Is Affirmative Action Still Necessary For For For For Blacks? Latinos? Asians? Women? 58 54 40 59 48 48 35 51 52 51 37 54 92 90 71 89 86 83 70 84 89 87 70 87 69 74 54 76 70 75 63 80 73 75 61 80 70 74 59 79 70 76 67 76 57 62 46 56 64 69 57 67 Has Affirmative Action Helped Someone Undeserving? 31 37 35 15 14 14 15 16 25 18 28 24 26 52 affirmative action and were more likely to say that they knew of a case in which affirmative action had helped someone undeserving. What is more striking about these data, however, is the fact that a majority of all four ethnic groups believed that affirmative action was still needed to help blacks, Latinos, and women. In all four groups, too, just a minority reported knowing someone who suffered from reverse discrimination. If anything, this pattern of opinion points to a sense of empathy rather than threat. It also confirms that voting for Proposition 209 was strongly influenced by the proponents’ success in framing the initiative as a vote against rigid group preferences rather than against all forms of assisting minorities to overcome discrimination and disadvantage. Accounting for the Vote on Proposition 209 The findings presented above failed to show strong or consistent differences in support for Proposition 209 between people of the same ethnicity living in different racial environments. Beliefs about the consequences of increased ethnic diversity mattered more than where one lived. With these uneven results as background, the next step is to provide a more complex account that incorporates the influences of both individual-level and neighborhood-level factors. Such an analysis helps to identify the social and political sources of voting on Proposition 209 and to estimate the residual effects of neighborhood context on support for or opposition to affirmative action. The statistical models discussed below incorporate the same set of predictors used in Chapter 3, with voting intentions on Proposition 209 as the dependent variable. The results address these main issues: 1. Whether the respondent’s ethnicity and strength of racial or ethnic identification had similar effects on voting on Proposition 209 in all types of neighborhood contexts; 2. Which other individual-level characteristics predicted voting on Proposition 209 and whether these same attributes mattered in every ethnic group; and 3. Whether contextual effects in voting for Proposition 209 emerge when one adjusts for the influences of the respondent’s 53 individual characteristics, and hence for compositional differences among neighborhoods. In describing the underpinnings of the vote on Proposition 209, we again report findings first for the sample as a whole and then for each ethnic group separately. With regard to the basic test for the racial threat hypothesis, the multivariate model confirmed the negative results presented above. Ethnic context had no direct effect on how Californians voted on affirmative action. The significant influences we detected were individual-level characteristics: ethnicity, general political orientations, and subjective perceptions of the effect of ethnic change. Of course, there could be relationships between objective neighborhood conditions and these attributes of residents; as mentioned above, minority residents of predominantly white neighborhoods were more likely to have a relatively high socioeconomic profile. Table 4.5 provides an overview of the model estimating the vote on Proposition 209 for the sample as a whole. This table indicates whether the respondent’s ethnicity, racial or ethnic identification, and residential context had statistically significant influences on voting intention. The table also lists the other variables included in the model with significant effects. The more complete summary of the equations for the sample as a whole and for each ethnic subgroup in Table 4.6 gives the numerical value of the logistic regression coefficients for each predictor. Looking first at the results for the entire sample (first column in Table 4.5), we find strong relationships between ethnic background and voting intentions. Even after adjusting for other background characteristics, political orientation, and neighborhood context, minority Table 4.5 Effects of Race/Ethnicity, Racial Context, and Racial and Ethnic Identification on Proposition 209 Vote Race/Ethnicity, Racial Context, and Racial and Ethnic Identification Race/ethnicity: significant Racial context: insignificant Racial and ethnic identification: significant Other Significant Coefficients Age Party identification Ideological self-identification State economic situation 54 Table 4.6 Explaining Voting Intentions on Proposition 209 (logit analyses) 55 % minority in precinct Black Hispanic Asian Gender Education Age Party identification Ideology Neighborhood diversity Neighborhood effect of blacks Neighborhood effect of Latinos Neighborhood effect of Asians Neighborhood effect of illegal immigrants Neighborhood effect of legal immigrants Neighborhood effect of whites Personal financial situation State economic situation Racial identifiication Constant Total Sample B SE 0.01 0.36 –1.15** 0.34 –0.98** 0.26 –1.31** 0.28 –0.21 0.19 –0.14 0.09 0.36** 0.12 0.47** 0.12 0.44** 0.13 –0.13 0.18 –0.07 0.36 0.37 0.34 –0.22 0.42 0.28 0.25 –0.26 0.39 –0.17 0.55 0.06 0.13 0.36** 0.12 0.19* 0.09 –2.40* 1.02 Adjusted R–squared n 0.17 640 NOTES: B is logit coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Whites B SE 0.49 0.62 –0.57# –0.01 0.70** 0.95** 0.37 0.02 0.18 1.00 –0.45 –0.41 –0.95 1.10 –0.15 1.10** 0.06 –5.52** 0.26 241 0.35 0.16 0.24 0.21 0.24 0.34 0.64 0.63 0.69 0.48 0.73 1.30 0.25 0.26 0.16 2.07 Blacks B SE 1.74 1.59 0.08 –0.62* 0.16 0.50 0.43 –0.72 –0.02 –0.90 –0.03 1.24 –0.58 0.58 0.29 0.35 0.61 0.41 0.60 1.13 1.03 1.51 0.89 1.26 –0.18 0.39 0.53# –2.35 0.19 91 0.45 0.38 0.27 3.13 Hispanics B SE –0.79 0.67 –0.18 –0.18 –0.11 0.21 0.20 –0.10 –1.13 0.43 0.00 0.78# 0.25 –0.44 0.14 0.08 0.07 –0.51 0.07 195 0.34 0.17 0.23 0.22 0.22 0.32 0.88 0.64 0.77 0.46 0.70 0.77 0.22 0.21 0.15 1.79 Asians B SE 0.89 1.00 –0.19 0.09 0.63# 0.03 0.51 0.03 1.43 0.82 0.53 0.26 0.35 0.35 0.37 0.52 1.10 1.14 –0.44 0.69 0.44 1.29 0.33 –0.05 0.46# –7.91** 0.13 104 0.39 0.33 0.26 3.09 respondents were less likely than whites to favor the anti-affirmative action initiative. As noted, ethnic context did not alter these results. Neither the proportion of minority residents of a neighborhood nor the perception of increasing ethnic diversity had an independent influence on voting intentions. Table 4.6 reveals that feelings of racial identity did affect voting decisions. Overall, respondents who said that their racial and ethnic background was important to them when they decided political and social questions were more likely to oppose Proposition 209. Because strong feelings of ethnic identity were much more prevalent among minority group voters than whites, this result suggests that the mobilization of feelings of ethnic consciousness may have made the connection between affirmative action policies and their group interests more visible and psychologically salient. If this is true, then there may have been an indirect influence of ethnic context upon voting on Proposition 209, captured by the effect of residential context on the strength of group consciousness. Table 4.6 reveals the strong effects of party affiliation and ideology on voting on Proposition 209. Even with multiple controls for personal background and local context, Republicans and self-defined conservatives were more likely to vote against affirmative action. So, too, were older voters and those who felt that the state economy had deteriorated in the last year, a finding that is consistent with the idea that feelings of economic vulnerability often heighten ethnic tension and competition. With differences in general political outlook controlled for, however, women respondents in our sample were as likely as men to say they would vote for Proposition 209. Despite the efforts of the opposition campaign to frame the initiative as dangerous to women’s interests, this argument may not have had a special resonance among female voters as a whole. This is not to say that gender consciousness did not play a similar role as feelings of racial or ethnic identity. The gender gap in party affiliation and ideology—with women more likely than men to be Democrats and liberals—suggests that the role of group consciousness among women is already captured by these control variables. The results for the entire sample blur some interesting differences between whites and minority voters. The pattern of effects among whites 56 generally approximated those reported for the overall sample: age, a Republican party affiliation, and the feeling that the state economy had deteriorated boosted support for Proposition 209 among white respondents. In general, however, party affiliation and ideology had stronger influences on the choices of whites than on those of minority group members. Among black respondents, a higher level of formal education and strong feelings of racial identification significantly increased opposition to Proposition 209. Ethnic context had no effect among this group. Among the Latino respondents in the sample, there were hints of an influence of ethnic context, as measured by the proportion of minority residents in one’s neighborhood; the coefficient for this predictor approached conventional levels of statistical significance. No other predictor in the model had a statistically significant effect for the Latino respondents. Finally, among Asians, only age and pessimistic views of the state’s economy had statistically significant effects on how they intended to vote on Proposition 209. Once again, residential context did not seem to matter. Reviewing the Role of Context To summarize, this chapter’s investigation of voting on Proposition 209 found that ethnic group interest and broad political identifications were the main determinants of how Californians regarded affirmative action. The ethnic composition of one’s neighborhood did not have a direct effect on how people voted, disconfirming the racial threat hypothesis that anti-minority reactions among whites are related to the size of the local minority group population. This latter result deviates from a recent national study that found a significant relationship between the size of the black population in an area and racial hostility among the local white residents (Taylor, 1998). One possible explanation for these divergent findings is that the links in the causal chain from neighborhood composition to perception of threat to group identification to political conduct are more complex than generally assumed by the racial threat theory. The concentration of minority populations is not a necessary condition for perceived threat. What may matter more is change in the ethnic composition of a 57 neighborhood, the nature of inter-group contacts, and the economic status, cultural distinctiveness, or political organization of minority group members. One might argue that a more differentiated treatment of context might have revealed effects undetected by the present study’s emphasis on the proportion of minority group residents. Limitations in the available data restricted our ability to pursue some interesting alternatives; for example, we lacked census information about change in the composition of the neighborhood units of analysis. Nevertheless, analyses using the proportion of black and proportion of Latino residents in a neighborhood—rather than the proportion of minority residents as a whole—as the measure of ethnic context did not change the results reported above. Another possibility is that economic, rather than ethnic, context mattered, with the reactions of white voters on racial issues conditioned by local economic circumstances. In addition, if neighborhoods with large minority populations also tended to be poor—as measured by the aggregate level of unemployment, the extent of home-ownership, and median income, for example—any observed effect of ethnic context might be a spurious reflection of local economic conditions. To address these possibilities, we also estimated models predicting voting intentions on Proposition 209 that incorporated the proportion of the households in a neighborhood with annual incomes under $10,000, between $10,000 and $20,000, and so on, as a measure of economic context. Including economic context as a control did not alter the estimated coefficient for ethnic context; nor did economic context have an independent effect on stated voting intentions. It also could be argued that using the precinct or neighborhood as the unit for measuring context may not capture the pattern of group interactions that actually foster a sense of threat. It may be that the metropolitan area or county is the focal point for political or economic competition between ethnic groups. In the present study, we were able to merge county-level contextual data with information about individual respondents. Analyses showed that the size of the minority population at the county level similarly had no effect on the racial attitudes or beliefs about Proposition 209 among white respondents. 58 The divergent results about contextual effects also may be related to differences in the attitudinal or behavioral responses to minority populations being studied. For example, Taylor (1998, pp. 522–523) found strong effects of the proportion of blacks in an area on the level of traditional anti-black prejudice and stereotyping of blacks. But her study found no statistically significant effect of ethnic context on hostility to race-targeting policies designed to achieve equal outcomes—a measure that is close in meaning to support for Proposition 209. More generally, beliefs about the relative costs and benefits of ethnic diversity are founded on vicarious learning as well as on actual proximity to members of other groups. Competition for admission into the University of California occurs statewide, so the proportion of minority group members in one’s own neighborhood or county is unlikely to be the only source of opinions about how whites are likely to be affected by affirmative action programs in higher education. In addition, the campaign over Proposition 209 centered on how to define affirmative action in the public’s mind, with references to group interest, party, ideology, and principles of fairness as the main cues to which voters in all contexts were exposed. Under these circumstances, generalized political orientations rather than localized concerns might be expected to influence voting behavior. The oft-assumed tendency of whites to react to ethnic change around them with increased opposition to policies, parties, or candidates perceived as pro-minority is not automatic. This may vary with the visible effect locally of large minority populations, as with the effect of immigration on the makeup of public schools, or with how issues are defined and justified.2 One result worth pursuing in further research is the influence of ethnic context, defined in both objective and subjective terms, on feelings of group consciousness among both whites and minorities. ____________ 2Several studies (e.g., Tolbert and Hero, 1998) suggest that the proportion of Latinos and immigrants in a county influenced the vote on Proposition 187. The results are problematic, since those analyses did not control for compositional effects. The present study asked respondents whether they approved of Proposition 187. However, this retrospective answer coming two years after the actual election is an unreliable indicator of support, so our failure to detect contextual effects on responses to this question cannot be regarded as definitive. 59 Identity politics—the idea that politics should be motivated by the drive to enhance the standing of one’s own ethnic group—may be more widespread in locales with a substantial number of minority group numbers. One reason for this increased likelihood is that greater numbers engender denser networks of intra-group communications and facilitate the mobilization of resources for political participation. If this is so, then the influence of context on racial attitudes and voting may be quite potent, but indirect. We found that broad political orientations such as party affiliation and ideological outlook were stronger predictors of voting on Proposition 209 among whites than among minority groups. More generally, our models did poorly when it came to accounting for the choice of nonwhite respondents. To improve our understanding of cleavages within minority groups, subsequent analyses should attempt to incorporate better measures of the social status, residential mobility, and immigrant generation among respondents from these groups. 60 5. Conclusion In many respects, the relations among ethnic groups are more a product of civil society than public policy. Group attitudes and stereotypes are formulated largely by daily experience and the cultural transmission of images—processes that are substantially beyond the realm of governmental action and leadership. In fact, there are inherent limitations on the capacity of democratic governments to modify ethnic group relations. In the United States, for instance, the First Amendment freedoms of speech and association hamper the ability of public officials to mold social attitudes in desired directions. Whether the issue is teenage violence, civic malaise, or hate speech, Americans seem to fear the potential remedies of thought control and censorship more than the problem per se. Yet, there are important reasons why public policies that improve the quality of inter-group relations in California are needed: The ethnic diversity of the state’s population is a given, and it seems axiomatic that hostility and conflict among the various ethnic groups can have debilitating consequences for society. Even when racial and ethnic tensions do not lead to overt unrest and violence, they can contribute to economic inefficiency and undermine the legitimacy of government. Prejudice may result in a failure to hire the best people or negotiate the best contracts because people refuse to work or deal with others from different racial and ethnic groups. Deep ethnic cleavages also can make bad politics. Decisions by a government that is controlled by one racial or ethnic group will have less moral authority if the members of the outgroups feel that they have not had a fair chance to affect political outcomes. Second, even if the pattern of inter-group relations is primarily formed in civil society, governmental actions can accelerate or moderate trends that emerge in the private sector. Even when a government does 61 not legally discriminate against particular groups, it can favor some groups over others in more subtle ways. Holding power gives some groups access to jobs and economic opportunities that are denied to others. The needs of some communities may receive more attention than the needs of others. The effects of these differences can exacerbate or lessen racial tensions. Also, elected officials can influence attitudes about race and ethnicity by legitimating some positions and attacking others. Their campaigns help frame public debate by choosing either to emphasize or downplay ethnic issues. These strategic decisions influence the salience of group interests and identifications, potentially evoking strong emotional responses from voters. The struggle over affirmative action is illustrative. As a policy, Proposition 209 altered the conditions under which people of different genders, races, and ethnicities competed for jobs and educational opportunities. But the fact that the debate over Proposition 209 was framed within the context of partisan presidential politics affected the perceptions that different groups had of the measure. This study examined public reactions to ethnic diversity in California in the context of political contention about the future of affirmative action in the state. One important purpose was to understand whether the prevailing pattern of group attitudes made harmonious relations among the state’s different racial and ethnic communities more difficult to achieve. A second goal was to examine whether the ethnic composition of residential areas affected inter-group relations. In particular, we tested the widely held theory that white prejudice and hostility are related to the size of minority group populations in an area. There are four major conclusions. The first concerns the pattern of conflict and consensus across ethnic groups. There were substantial ethnic group differences on a number of racial attitudes and policies. Specifically, whites were less likely than members of the three ethnic minority groups to say that their race or ethnicity was important for their political identity. Whites were more likely than voters in the other ethnic groups to support Propositions 209 and 187 and were more likely to regard affirmative action programs as unnecessary and unfair. These ethnic group differences cannot be simply explained by other factors such 62 as party identification, ideology, social background, or residential context (see Table 4.6). At the same time, there was consensus among the four ethnic groups on a number of questions related to the state’s ethnic composition. An important point is that a large majority in every group was either neutral or positive about the effect of people of a different ethnic or cultural group moving into their neighborhood. They also were united in expressing more negative views about illegal immigrants than about any other group. Despite their differences about the need for affirmative action, respondents from all four ethnic groups also tended to agree on the egalitarian principles of merit and nondiscrimination as the primary criteria for job promotion and educational opportunity. Even if a desire to say the socially acceptable thing was an influence on responses in our survey, the generally sanguine outlook on ethnic heterogeneity at the neighborhood level is comforting. Certainly we found little evidence that Californians, whatever their racial and ethnic background or residential context, viewed current demographic trends in the state as creating the “ethnic cauldron” depicted by some journalists and scholars. At several points we noted that a majority in every racial and ethnic group seemed to view the effect of their own ethnic group on the neighborhood most favorably and to agree that people tended to be happier living and socializing with others of the same background. It would be highly misleading, however, to conclude that Californians consider residential and social segregation as a desirable goal. As other studies have shown (Farley et al., 1994), people of all ethnic groups indicate that their preferred neighborhood is one in which people of their own background are a majority, but not their only neighbors. Whites seemed most worried about the effect on property values of a heavy influx of black residents; blacks were most concerned about the prevalence of hostile attitudes in a largely white community. In addition, realism, not prejudice, may underlie the belief that it is easier to socialize with people with a similar culture and background. This separatist opinion is not the same as saying that people would not value friendship with people from other ethnic groups. Indeed, the rising rates of intermarriage between whites and Latinos and between whites and Asians 63 in California belie such a claim. Although there was considerable doubt about the feasibility of achieving a color-blind society, these data do not gainsay the accumulated evidence of public acceptance of integration (or diversity) in the workplace, in higher education, and at public gatherings such as movies, sporting events, and the like. To be sure, some people do move to live in ethnically more homogeneous neighborhoods; the phenomenon of white flight is real. Doubtless, too, there also are people who seek to live in culturally diverse communities. As long as there is the freedom to make these residential choices, there will be ethnic clusterings of all sorts, and only some neighborhoods will be ethnically mixed. What this outcome means for the quality of group relations depends in part on how ethnic context affects inter-group relations. The critical question that deserves additional research is whether the tolerance that groups express toward one another is genuine, or whether it is simply the culturally acceptable response that masks deeper hostilities. If it is the latter, we could be lulled into underestimating the extent of the potential problems arising from the changing ethnic composition of the state. The second major conclusion of this study is that multiethnic neighborhoods are not the primary locus of ethnic political divisions in this state. A number of previous academic studies have argued for the threat hypothesis, suggesting that racial tension should be highest in contexts where different groups interact the most. The present data showed that in most instances there was very little variation in perceptions and attitudes between ethnically mixed and homogeneous areas. Indeed, where there were differences between whites and nonwhites, they often were in the direction of greater tolerance and agreement between whites and nonwhites in mixed areas than in homogeneous ones. Clearly, a fuller understanding of the conditions under which ethnic context has a distinctive influence on white attitudes and political behavior requires more direct measures of perceived threat, inter-group contact, and residential choices reflecting different tastes for diversity. As noted above, ethnic context may be a proxy for several, potentially offsetting, social processes. The implication of our findings about the lack of contextual effects is that there is no reason to assume that as California’s growing ethnic 64 diversity multiplies the number of mixed neighborhoods, the level of ethnic tensions necessarily will increase. The fact that whites in multiethnic neighborhoods were not less tolerant than those living in mainly white areas means that efforts to reduce prejudice cannot simply be focused on specific localities experiencing ethnic change, such as the large cities. California’s political climate on racial and cultural issues depends as much (perhaps even more) upon the perceptions of the more numerous whites who live in majority white areas than whites who live near sizable minority populations. Concentrating only on the ethnically diverse areas would therefore neglect important sources of political tension. Among all ethnic groups, living in an ethnically mixed neighborhood boosted feelings of ethnic and racial identification. Clearly, how group consciousness—what social psychologists call in-group favoritism— affects political conduct is a vital issue that deserves close attention. From the perspective of minority groups, a stronger sense of racial and ethnic identity may fuel efforts to advance collective goals and redress inequality. Obviously, one implication of the finding that racial and ethnic identification among minorities is stronger in areas with a higher density of minorities suggests that efforts at political organization may be more successful by targeting residents in those neighborhoods. However, raising the salience of ethnic identity also may foster a backlash and intensify group conflict as the collision of in-group favoritisms makes for more out-group hostility. The third major conclusion is that attitude formation on racial and ethnic issues in California politics seems to occur more on a global than a local basis. There were few differences across neighborhoods sorted by their ethnic makeup, and where these did emerge they seemed to be more compositional than contextual. Differences in the background characteristics, ideology, and party affiliation of respondents primarily account for the observed variations across neighborhoods. To use the example of Proposition 209, people’s voting intentions were not so much influenced by what people perceived in their local areas as by their general political orientations and what they learned from the respective pro- and anti-209 campaigns. 65 This suggests that how political leaders and the mass media frame issues may be more critical to racial and ethnic attitudes than what goes on in local contexts. Because most people get their political information from the newspapers and the electronic media, local experience may be swamped by information diffused on a much broader basis. Even as they strive to improve group relations, policymakers must be aware that media coverage they cannot control may have a stronger influence on opinion formation. The final main conclusion of this study concerns the relatively negative orientation of Californians toward undocumented immigrants. Respondents in our poll—conducted two years after Proposition 187— had distinctively more positive views about legal immigrants than illegal ones. This finding held for all of the major ethnic groups. An implication of this is that policymakers should be careful not to lump legal and illegal immigrants together in assessing the public support for programs that affect these groups. Educating the public that the legal Latino population is much larger than the undocumented group also might reduce resentment about immigration and increase the public’s willingness to support policies that aid the larger Latino community. In sum, there is no evidence that ethnic group relations are particularly troubled in highly diverse areas or that these neighborhoods will serve as cauldrons for future problems. Racial attitudes seem to be more influenced by a person’s political and social background than by local context. Resistance to ethnic diversity is not more likely among those who directly experience it near where they live. In some ways, this is a hopeful note on which to conclude, because it tells us that ethnic tensions in California are not preordained to increase as the state continues to cope with the effect of its changing demography. 66 Appendix A Survey Sample and Data Sources Our data are drawn from a telephone survey of 1,500 registered voters conducted between October 26 and November 2, 1996, just before the general election. The sample was designed by Dr. Ken McCue. Using California’s Statewide Database, formerly known as the Assembly’s Redistricting Database, the entire state was divided into “consolidated precincts.” The consolidated precinct level is a unit of analysis whose creation becomes necessary when merging census data with election results or registration data. For the purpose of this sample design, the consolidated precinct level is used to represent a “neighborhood” or context. Census data merged with registration data had to be used to draw the sample because the objective was to find registered voters of certain racial categories who lived in the same neighborhood. The Census does not report registration data, however, and the statement of registration for voting does not report race or ethnicity. Census data are block-level data that are, in most cases, reported by the block group. Election data are collected on the electoral precinct level and registration data on the registration precinct level. Both sets of precincts encompass census blocks as the smallest unit of analysis, but they do not share all boundaries in most cases. One important objective of this sample design was to locate minority groups in varying racial contexts to enable comparisons of, for example, blacks living in majority black neighborhoods with blacks in majority minority neighborhoods. The sample was constructed by dividing registered voters into 10 categories and then drawing a random sample from each stratum, listed below as distinctive “objective” contexts. Thus, our purpose was not to represent the state’s electorate; rather, we wanted to include sufficient numbers of minority respondents to enable systematic comparisons of all ethnic groups across contexts. Given this research design, it is not surprising that the lower level of 67 support for Proposition 209 within our sample was lower than the actual vote. The first question in our survey summarized Proposition 209 and asked respondents: “If the election were being held today, would you vote yes or no on this initiative?” Those in favor constituted 41 percent of the sample, and another 15 percent said they were still undecided. As noted above, the contextual variable we employ as an indicator of racial threat is the size of the minority population within a consolidated precinct. As also noted above, 1990 Census data about the ethnic “mix” within a precinct is our “objective” measure of context. Respondents were grouped according to whether they lived in majority white precincts; majority minority precincts, in which the total of black, Latino, and Asian residents rather than a single ethnic group constituted a majority; majority black precincts; or majority Latino districts. Clearly, though, in an ethnically diverse society such as California, census data enable a number of alternative categorizations of ethnic context. For example, one might measure the size of each minority group separately or employ different combinations depending on the particular issue (e.g., language policy or affirmative action). Another possibility is to define context by the relative economic status of particular ethnic groups rather than their relative size. Similarly, the specific nature of the hypothesized group threat might dictate one’s choice of contextual variable. For example, if one believed that ethnic change posed a “cultural” threat to the majority group, then language use rather than ethnic origin might be the appropriate grouplevel variable to use. By the same token, if the hypothesized collective threat were economic, then economic conditions within a jurisdiction, such as mean per capita income or unemployment level, might be considered potential contextual sources of voting. 68 Appendix B Multivariate Analyses from Chapters 2 and 3 69 Table B.1 Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 (OLS regression) 70 % minority in precinct Black Hispanic Asian Gender Education Age PID Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Effect of Blacks B SE 0.03 0.06 –0.15** 0.05 –0.05 0.04 –0.05 0.04 0.06* 0.03 –0.01 0.01 0.01 0.02 0.03 0.02 0.00 0.02 –0.01 0.03 0.02 0.02 0.02 0.02 0.01 0.01 1.82** 0.12 Effect of Latinos B SE –0.08 0.06 0.01 0.06 –0.15** 0.05 0.00 0.05 0.05# 0.03 –0.02 0.02 0.01 0.02 0.05* 0.02 –0.02 0.02 0.03 0.03 0.06** 0.02 0.05* 0.02 0.03* 0.02 1.60** 0.14 Effect of Asians B SE –0.11# 0.06 0.03 0.06 –0.01 0.04 –0.14** 0.05 0.07* 0.03 –0.04** 0.01 0.00 0.02 0.00 0.02 –0.01 0.02 0.01 0.03 0.03 0.02 0.03 0.02 0.02 0.01 1.82** 0.13 Effect of Whites B SE –0.07 0.05 0.05 0.05 0.03 0.04 –0.05 0.04 0.03 0.03 0.01 0.01 –0.02 0.02 –0.02 0.02 0.03 0.02 –0.05* 0.03 0.06** 0.02 0.00 0.02 0.02 0.01 1.81** 0.12 Effect of Legal Immigrants B SE –0.04 0.07 –0.01 0.06 –0.09# 0.05 –0.10* 0.05 –0.01 0.04 –0.03# 0.02 –0.04 0.02 –0.01 0.02 –0.01 0.02 0.00 0.03 0.04# 0.03 0.04 0.02 0.02 0.02 1.93** 0.15 Adjusted R–squared n 0.02 1,036 0.05 1,041 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. 0.03 1,010 0.01 1,043 0.01 922 Effect of Illegal Immigrants B SE 0.20** 0.08 –0.08 0.07 –0.10# 0.05 –0.02 0.06 –0.01 0.04 0.00 0.02 –0.03 0.03 0.06* 0.03 0.00 0.03 0.09** 0.04 0.04 0.03 0.01 0.03 0.03 0.02 1.68** 0.16 0.02 829 Table B.2 Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 (OLS regression) 71 Blacks in majority black precincts Blacks in majority minority precincts Blacks in majority white precincts Latinos in majority Latino precincts Latinos in majority minority precincts Latinos in majority white precincts Asians in majority minority precincts Asians in majority white precincts Whites in majority minority precincts Gender Education Age PID Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Effect of Blacks B SE –0.19** 0.06 –0.08 0.08 0.11 0.12 –0.03 0.06 0.01 0.06 –0.05 0.06 –0.02 0.06 –0.06 0.06 0.04 0.05 0.07* 0.03 –0.02 0.01 0.01 0.02 0.03 0.02 0.00 0.02 –0.01 0.03 0.02 0.02 0.02 0.02 0.01 0.01 1.82** 0.12 Effect of Latinos B SE –0.04 0.07 –0.10 0.09 0.11 0.13 –0.26** 0.06 –0.19** 0.07 –0.06 0.07 –0.02 0.06 –0.02 0.07 –0.02 0.05 0.06# 0.03 –0.02 0.02 0.01 0.02 0.05* 0.02 –0.01 0.02 0.03 0.03 0.06** 0.02 0.05* 0.02 0.03* 0.02 1.58** 0.13 Effect of Asians B SE 0.02 0.06 –0.11 0.09 –0.04 0.12 –0.11# 0.06 0.02 0.06 0.01 0.07 –0.23** 0.06 –0.08 0.06 –0.01 0.05 0.07* 0.03 –0.04** 0.01 –0.01 0.02 0.01 0.02 –0.01 0.02 0.01 0.03 0.04# 0.02 0.03 0.02 0.03# 0.01 1.76** 0.13 Adjusted R-squared n 0.02 1,036 0.05 1,041 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. 0.04 1,010 Effect of Whites B SE –0.06 0.06 0.09 0.08 0.06 0.11 –0.07 0.05 0.03 0.06 –0.01 0.06 –0.13** 0.05 –0.04 0.06 –0.07 0.05 0.03 0.03 0.01 0.01 –0.02 0.02 –0.02 0.02 0.03 0.02 –0.06* 0.03 0.06** 0.02 0.01 0.02 0.02 0.01 1.82** 0.11 0.02 1,043 Effect of Legal Immigrants B SE –0.12# 0.07 0.01 0.10 –0.07 0.14 –0.23** 0.07 –0.10 0.07 –0.11 0.07 –0.19** 0.07 –0.13# 0.07 –0.13* 0.06 –0.01 0.04 –0.03# 0.02 –0.04 0.02 –0.01 0.02 –0.02 0.02 –0.01 0.03 0.04# 0.03 0.04# 0.02 0.02 0.02 1.98** 0.14 0.02 922 Effect of Illegal Immigrants B SE 0.09 0.08 –0.09 0.11 0.05 0.15 0.04 0.08 –0.01 0.08 –0.07 0.08 0.09 0.08 –0.01 0.08 0.11# 0.07 –0.01 0.04 –0.01 0.02 –0.03 0.03 0.05* 0.03 0.01 0.03 0.10** 0.04 0.04 0.03 0.01 0.03 0.03 0.02 1.73** 0.16 0.02 829 Table B.3 Racial Identification and Racial Attitudes—Black Respondents, from Chapter 3 (OLS regression) 72 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –1.27** 0.36 0.19 0.15 0.01 0.07 0.01 0.10 2.71** 0.45 Color-Blind Society B SE –0.38 0.58 –0.47# 0.26 –0.34** 0.12 –0.05 0.16 0.04 0.25 0.00 0.17 –0.31 0.22 –0.19 0.18 0.11 0.16 0.04 0.12 4.65** 1.20 Adjusted R–squared 0.05 0.03 n 190 159 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE 0.30 0.49 –0.08 0.22 –0.12 0.10 –0.08 0.13 –0.10 0.21 –0.07 0.14 0.01 0.19 0.19 0.15 –0.07 0.13 0.13 0.10 1.81# 1.02 0.02 161 Happier with Others of Same Background B SE 0.04 0.54 –0.04 0.24 0.11 0.11 –0.13 0.15 –0.34 0.25 0.09 0.16 –0.06 0.21 0.39* 0.17 0.07 0.15 0.12 0.11 1.74 1.14 Minorities Gain at Expense of Others B SE 1.08# 0.56 –0.19 0.25 –0.18 0.12 –0.27# 0.15 –0.06 0.24 –0.33* 0.17 –0.59** 0.22 0.24 0.17 0.03 0.15 0.20# 0.12 4.14** 1.18 Minorities Get More Than They Deserve B SE –0.08 0.57 –0.04 0.25 –0.39** 0.12 –0.10 0.16 –0.05 0.26 0.04 0.17 0.16 0.23 0.11 0.17 0.16 0.16 0.03 0.12 3.15** 1.20 0.01 0.07 0.04 156 156 158 Table B.4 Racial Identification and Racial Attitudes—Latino Respondents, from Chapter 3 (OLS regression) 73 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –0.42# 0.24 0.06 0.12 0.11* 0.06 0.06 0.08 1.96** 0.29 Color-Blind Society B SE –0.09 0.38 0.03 0.19 –0.23** 0.09 0.21 0.13 –0.02 0.13 –0.29* 0.12 –0.44* 0.18 –0.15 0.13 0.14 0.12 –0.03 0.08 4.64** 0.76 Adjusted R-squared 0.01 0.03 n 367 303 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE –0.03 0.33 –0.07 0.17 –0.21** 0.08 –0.12 0.11 0.00 0.12 0.15 0.11 0.07 0.16 –0.01 0.12 0.21* 0.11 0.23** 0.07 1.58* 0.67 Happier with Others of Same Background B SE –0.52 0.33 0.08 0.17 –0.10 0.08 0.05 0.11 0.14 0.12 0.23* 0.11 –0.06 0.16 –0.18 0.12 0.03 0.11 0.20** 0.07 2.14** 0.67 Minorities Gain at Expense of Others B SE –0.32 0.35 –0.29 0.18 –0.21** 0.08 0.05 0.12 0.05 0.12 –0.11 0.12 0.01 0.17 –0.04 0.12 –0.05 0.11 –0.15* 0.08 4.22** 0.70 Minorities Get More Than They Deserve B SE 0.59 0.38 0.24 0.19 –0.12 0.09 –0.10 0.13 0.02 0.13 –0.12 0.12 0.04 0.18 –0.04 0.13 –0.34** 0.12 0.05 0.08 3.76** 0.76 0.05 0.04 307 301 0.02 0.02 300 303 Table B.5 Racial Identification and Racial Attitudes—Asian Respondents, from Chapter 3 (OLS regression) 74 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –0.09 0.27 –0.01 0.13 0.06 0.06 0.06 0.09 1.90** 0.35 Color-Blind Society B SE 0.53 0.44 0.05 0.21 –0.26* 0.10 0.38** 0.15 0.10 0.14 –0.17 0.15 –0.36# 0.20 –0.32# 0.17 0.00 0.14 0.12 0.10 3.92** 0.84 Adjusted R–squared 0.01 0.08 n 248 196 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE –0.44 0.41 0.14 0.20 –0.13 0.09 0.10 0.13 –0.09 0.13 0.17 0.14 –0.19 0.18 –0.01 0.16 0.34** 0.13 0.22* 0.10 1.74* 0.77 Happier with Others of Same Background B SE –0.39 0.42 0.32 0.20 –0.03 0.10 0.02 0.14 –0.02 0.13 –0.17 0.14 0.13 0.18 0.01 0.16 0.14 0.13 0.03 0.10 2.04** 0.79 Minorities Gain at Expense of Others B SE –0.13 0.42 –0.49* 0.21 –0.09 0.10 0.36* 0.14 0.04 0.14 0.30* 0.15 0.12 0.19 –0.11 0.17 0.14 0.13 –0.12 0.10 2.20** 0.82 Minorities Get More Than They Deserve B SE 0.05 0.44 –0.31 0.22 –0.31** 0.10 0.04 0.15 0.12 0.14 0.19 0.16 0.29 0.20 –0.11 0.18 –0.05 0.14 0.02 0.10 3.47** 0.85 0.05 0.02 195 194 0.07 191 0.04 185 Table B.6 Racial Identification and Racial Attitudes—White Respondents, from Chapter 3 (OLS regression) 75 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –0.47** 0.17 0.08 0.09 0.06 0.04 0.02 0.06 2.86** 0.23 Color–Blind Society B SE 0.05 0.27 –0.22 0.15 –0.12# 0.07 –0.05 0.10 0.19* 0.09 0.11 0.11 –0.08 0.13 –0.07 0.10 0.04 0.10 0.03 0.07 3.00** 0.58 Adjusted R-squared n 0.01 0.01 516 438 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE –0.36 0.26 –0.10 0.14 –0.10 0.06 0.08 0.09 0.22** 0.08 0.13 0.11 –0.20 0.13 –0.04 0.10 0.39** 0.10 0.00 0.07 2.01** 0.56 0.09 439 Happier with Others of Same Background B SE –0.06 0.23 0.22# 0.13 0.23** 0.06 –0.06 0.08 –0.10 0.08 –0.06 0.09 –0.10 0.12 –0.02 0.09 –0.04 0.09 0.12* 0.06 1.93** 0.50 Minorities Gain at Expense of Others B SE –0.11 0.25 0.01 0.14 –0.15** 0.06 –0.04 0.09 0.24** 0.08 0.37** 0.10 0.02 0.12 0.17# 0.10 –0.05 0.10 –0.06 0.07 2.46** 0.54 Minorities Get More Than They Deserve B SE 0.10 0.26 –0.16 0.14 –0.32** 0.06 0.06 0.09 0.21** 0.08 0.10 0.10 0.16 0.13 0.05 0.10 0.14 0.10 0.08 0.07 2.78** 0.56 0.05 0.09 426 429 0.10 431 Appendix C Survey Question Wording Prop 209 Proposition 209, referred to as the California Civil Right Initiative, sometimes known as CCRI, would prohibit state and local governments from using race, sex, color, ethnicity or national origin as criteria for either discriminating against or granting preferential treatment to individuals or groups in hiring, promoting, granting admissions to college or selecting public contractors. If the election were being held today, would you vote yes or no on this initiative? Vote for President Who do you think you will vote for in the election for president? Neighborhood Diversity Do you think your neighborhood is becoming more ethnically diverse, less ethnically diverse, or is about the same? Neighborhood Effect of Groups Now I am going to ask you what effect different groups of people have had on the quality of life in the neighborhood where you live. After I give you the name of a group, please tell me whether that group has improved, made worse, or had no effect on the quality of life in your neighborhood. Blacks Latinos Asians Illegal Immigrants 77 Legal Immigrants Whites Who Still Needs Affirmative Action? As you know, Proposition 209 will eliminate all state and local government actions which in the past have provided special programs intended to benefit certain groups of people. Which of the following groups do you think still need the programs eliminated by Proposition 209? Blacks Latinos Asians Women Meaning of 209 Vote When you vote on initiatives like Prop 209, what do you most want to accomplish, change or preserve the law, or send a message to government officials? Will Private Companies Continue AA? Proposition 209 does not prohibit companies and businesses from providing special programs which benefit certain groups of people. Do you think that private companies and businesses would continue to provide these programs if Proposition 209 passes? At what level? If so, would they provide them at the same level as in the past or at a lesser level? Racial Attitudes Battery I am going to read you some statements which some people agree with, while others do not. Please tell me if you agree strongly, agree somewhat, are neutral, disagree somewhat, or disagree strongly. 78 1. The more jobs and places in college provided to minorities, the fewer there are for people who are not members of those groups. 2. In our lifetime, America can become a color-blind society. 3. Members of certain ethnic or racial groups still have less opportunities to get ahead than other people. 4. Members of particular ethnic or racial groups use special programs to get more benefits than they deserve. 5. People of different ethnic and racial groups are generally happier when they live and socialize with others of the same background. Diversity or Merit? Which of these statements come closer to the way you feel? 1. Diversity benefits our country economically and socially, so race, ethnicity, and gender should be a factor in determining the type of person who is hired, promoted or admitted to college. 2. Hiring, promotion and college admission should be based solely on merit and qualifications and not on characteristics of race, ethnicity or gender. Racial Identity in Politics When it comes to social and political matters, some people think of themselves mainly as black, white, Latino, Asian or Jewish, and that is very important to how they think of themselves. Other people don’t give as much thought to these things. When it comes to social and political matters, how important is your race and ethnicity to how you think of yourself? Personal Knowledge of Undeserved Job Do you have personal knowledge of a situation where a woman or a member of a racial minority got a job or promotion that he or she did not deserve as a result of affirmative action programs, or do you have no such knowledge? 79 Vote in 1994? Did you vote in the 1994 November election? Prop 187 Vote [If yes on 14] In 1994, there was a proposition on the ballot in California to deny governmental services to illegal immigrants called Proposition 187. How did you vote on Proposition 187? Ideological Self-Identification Do you consider yourself a liberal, moderate, or conservative? 209 Effect on Presidential Vote Would you change your vote for president based on whether or not that candidate had endorsed or opposed Proposition 209? Personal Financial Situation Would you say over the past year your personal financial situation has gotten better, worse, or stayed the same? California Economic Situation What about the economy of California? Would you say over the past year it has gotten better, worse, or stayed the same? Main Effect of 209 If Proposition 209 passes, do you think that its main effect will be to change California law or to send a message to government officials? Education What was the last grade of school you completed? Race What race do you consider yourself? (White, Black, Asian, or other?) 80 Asian Nationality [If Asian on question 22] What nationality do you consider yourself? (Japanese, Chinese, Korean, Vietnamese, Cambodian, Other?) Latino [If not Asian on question 22] What ethnicity do you consider yourself? (Latino or Other?) 81 Bibliography Allport, Gordon W., The Nature of Prejudice, Doubleday Anchor Books, Garden City, NJ, 1954. Blalock, Hubert M., Toward a Theory of Minority-Group Relations, Capricorn Books, New York, 1967. Bowers, Jake, “Black Threat and Black Context: New Evidence from the 1997 NES Pilot Study,” paper presented at the Annual Meeting of the International Society of Political Psychology, Montreal, Canada, 1998. Brown, Roger, Social Psychology (2nd ed.), Basic Books, New York, 1986. Cain, Bruce E., D. Roderick Kiewiet, and Carole J. Uhlaner, “The Acquisition of Partisanship by Latinos and Asian Americans,” American Journal of Political Science, Vol. 35, 1991, pp. 390–422. Citrin, Jack, Donald P. Green, Christopher Muste, and Cara Wong, “Public Opinion Toward Immigration Reform: The Role of Economic Motivations,” Journal of Politics, Vol. 59, 1997, pp. 858– 881. Clark, William A. V., “Residential Patterns: Avoidance, Assimilation, and Succession,” in Roger Waldinger and Mehdi Bozorgmehr, eds., Ethnic Los Angeles, Russell Sage Foundation, New York, 1996. Dawson, Michael C., Behind the Mule: Race and Class in AfricanAmerican Politics, Princeton University Press, Princeton, NJ, 1994. Farley, Reynolds, Charlotte Steeh, Maria Krysan, Tara Jackson, and Keith Reeves, “Stereotypes and Segregation: Neighborhoods in the Detroit Area,” American Journal of Sociology, Vol. 100, 1994, pp. 750– 780. Giles, Michael W., and Melanie A. Buckner, “David Duke and Black Threat: An Old Hypothesis Revisited,” The Journal of Politics, Vol. 55, 1993, pp. 702–713. 83 Glaser, James, “Back to the Black Belt: Racial Environment and White Racial Attitudes in the South,” Journal of Politics, Vol. 56, 1994, pp. 21–41. Green, Donald P., Robert P. Abelson, and Margaret Garnett, “The Distinctive Political Views of Hate-Crime Perpetrators and White Supremacists,” in Deborah A. Prentice and Dale T. Miller, eds., Cultural Divides: Understanding and Overcoming Group Conflict, Russell Sage Foundation, New York, 1999. Hood, M.V., III, and Irwin L. Morris, “Amigo o Enemigo? Context, Attitudes, and Anglo Public Opinion Toward Immigration,” Social Science Quarterly, Vol. 78, 1997, pp. 309–323. Key, V.O., Southern Politics in State and Nation, University of Tennessee Press, Knoxville, 1949. Levine, Robert, and Donald Campbell, Ethnocentrism, John Wiley & Sons, New York, 1972. Liska, Allen E., Mitchell B. Chamlin, and Mark D. Reed, “Testing the Economic Production and Conflict Models of Crime Control,” Social Forces, Vol. 64, 1985, pp. 119–138. Olzak, Susan, The Dynamics of Ethnic Competition and Conflict, Stanford University Press, 1992. Prentice, Deborah A., and Dale T. Miller, Cultural Divides: Understanding and Overcoming Group Conflict, Russell Sage Foundation, New York, 1999. Quillian, Lincoln, “Prejudice as a Response to Group Threat: Population Composition and Anti-Immigrant and Racial Prejudice in Europe,” American Sociological Review, Vol. 60, 1995, pp. 586–611. Quillian, Lincoln, “Group Threat and Regional Change in Attitudes Toward African Americans,” American Journal of Sociology, Vol. 102, 1996, pp. 816–860. Schuman, Howard, Charlotte Steeh, Lawrence Bobo, and Maria Krysan, Racial Attitudes in America: Trends and Interpretations (rev. ed.), Harvard University Press, Cambridge, MA, 1997. Stephan, Walter, “The Contact Hypothesis in Intergroup Relations,” in C. Hendrick, ed., Review of Personality and Social Psychology, Vol. 9, Sage Publications, Newbury Park, CA, 1987, pp. 13–40. 84 Tajfel, Henri, “Social Categorization, Social Identity, and Social Comparison,” in H Tajfel, ed., Differentiation Between Social Groups, Academic Press, London, 1978. Tate, Katherine, From Protest to Politics: The New Black Voters in American Elections, Harvard University Press, Cambridge, MA, 1993. Taylor, Marylee C., “How White Attitudes Vary with the Racial Composition of Local Populations: Numbers Count,” American Sociological Review, Vol. 63, 1998, pp. 512–535. Tolbert, Caroline, and Rodney Hero, “Race/Ethnicity and Direct Democracy: An Analysis of California’s Illegal Immigration Initiative,” Journal of Politics, Vol. 58, 1996, pp. 806–818. Uhlaner, Carole J., “Perceived Discrimination and Prejudice and the Coalition Prospects of Blacks, Latinos, and Asian Americans,” in Byran O. Jackson and Michael B. Preston, eds., Racial and Ethnic Politics in California, IGS Press, Institute of Governmental Studies, University of California at Berkeley, Berkeley, CA, 1991. Valenty, Linda O., Ron Sylvia, and Peter Haas, “Thresholds for Tolerance: The Impact of Racial and Ethnic Composition on California Propositions 209 and 187,” paper presented at the Annual Meeting of the International Society of Political Psychology, Montreal, Canada, 1998. Voss, D. Stephen, “Beyond Racial Threat: Failure of an Old Hypothesis in the New South,” Journal of Politics, Vol. 58, 1996, pp. 1156–1170. Wong, Cara, “Who Is This ‘We’ We Keep Talking About: Conceptualization and Measurement of Community,” paper presented at the Annual Meeting of the American Political Science Association, Boston, 1998. Wright, Gerald C., “Contextual Models of Electoral Behavior: The Southern Wallace Vote,” American Political Science Review, Vol. 71, 1977, pp. 497–508. 85 About the Authors BRUCE CAIN Bruce Cain is the Director of the Institute of Governmental Studies and Robson Professor of Political Science at the University of California, Berkeley. A graduate of Bowdoin College in 1970, he was a Rhodes Scholar at Trinity College, Oxford, and received his Ph.D. in political science from Harvard University in 1976. From 1976 to 1989, he taught California politics, political theory, and comparative government at the California Institute of Technology. His writings include The Reapportionment Puzzle, The Personal Vote (with John Ferejohn and Morris Fiorina), Congressional Redistricting (with David Butler), several co-edited volumes, and numerous articles for professional journals. He has served as a polling consultant for several state senate races and as a redistricting consultant to the California State Assembly, Los Angeles City Council, and Attorney General of the State of Massachusetts. He has been a consultant to the Los Angeles Times and a frequent commentator for radio and television stations in Los Angeles and the Bay Area. JACK CITRIN Jack Citrin is Professor of Political Science at the University of California, Berkeley, where he has taught since 1990. He received his B.A. from McGill University and his Ph.D from Berkeley. His research interests are American national politics, comparative politics in advanced industrial societies and California politics, with a focus on public opinion, political trust, and ethnic conflict. He is the author of The Politics of Disaffection among British and American Youth (1975), Tax Revolt: Something for Nothing in California (1982), and numerous articles and book chapters. With Professor David Sears of UCLA, he is presently completing a book entitled The Politics of Multiculturalism: Crisis of American Indentities. CARA WONG Cara Wong is a doctoral student at the University of California, Berkeley. Beginning in the Fall of 2000, she will be an assistant professor of political science at the University of Michigan. Her research interests include racial and ethnic politics in the United States, public opinion and voting behavior, and research methodology. 87" } ["___content":protected]=> string(102) "

R 800BCR

" ["_permalink":protected]=> string(96) "https://www.ppic.org/publication/ethnic-context-race-relations-and-california-politics/r_800bcr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8125) ["ID"]=> int(8125) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:35:04" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3237) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 800BCR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_800bcr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_800BCR.pdf" ["wpmf_size"]=> string(6) "174271" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(173785) "Ethnic Context, Race Relations, and California Politics ••• Bruce Cain Jack Citrin Cara Wong 2000 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Cain, Bruce E. Ethnic context, race relations and California politics / Bruce Cain, Jack Citrin, and Cara Wong. p. cm. Includes bibliographical references. ISBN 1-58213-051-5 1. California—Ethnic relations—Political aspects. 2. California—Race relations—Political aspects. 3. California— Politics and government—1951. 4. Minorities—California— Political activity. 5. Neighborhood—California. I. Citrin, Jack. II. Wong, Cara, 1970– III. Title. F870.A1 .C33 2000 305.8'009794—dc21 00-040312 Copyright © 2000 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword At some time in the early part of this century, no racial or ethnic group will make up a majority of California’s population. Demographers describe this state of affairs with all the wonderment of astronomers exploring the outer reaches of the universe. Other social scientists, politicians, and planners are more sobered by it. What does this mean for California, the Golden Dream, and life at the neighborhood level? What do we know about how people will get along? Will there be harmony, or will racial or ethnic tensions threaten California’s governance, let alone the smooth running of day-to-day community life? However we imagine this majority minority status and the changes it may bring, California has embarked on a journey that is unprecedented in U.S. history, with all the risk and excitement that such a journey entails. To help map this journey, PPIC has begun a series of research projects to describe this new demographic reality and to study the consequences of ethnic and racial diversity for the formation of public policy. As one of the first contributions to that series, Ethnic Context, Race Relations, and California Politics takes a careful look at ethnic group relations at the neighborhood level. After reviewing precinct-level data on racial attitudes and voting behavior, Bruce Cain, Jack Citrin, and Cara Wong find “no evidence that ethnic group relations are particularly troubled in highly diverse areas or that these areas will serve as cauldrons for future problems.” Even more encouraging is their finding that “where there were differences between whites and nonwhites, they often were in the direction of greater tolerance and agreement between whites and nonwhites in mixed areas than in homogenous ones.” At the same time, the authors found significant ethnic group differences in racial attitudes and voting behavior. They also found that most residents prefer to live in neighborhoods in which members of their own ethnic group constitute a majority. (They note, however, that this iii preference does not necessarily contradict an acceptance of diversity in the workplace, in higher education, or in public life generally.) Finally, the authors predict that “California’s political climate on racial and cultural issues will depend in large part on the perceptions of the more numerous whites who live in majority white areas.” What shapes these perceptions and voting preferences is unclear, but they do not seem to arise from daily interactions with neighbors of different ethnic groups. This study and its companion, How Different Ethnic Groups React to Legal Authority, by Yuen Huo and Tom Tyler, have been supported by PPIC with two objectives in mind. The first is to describe and understand California’s changing demography to anticipate the challenges we will inevitably face. Demographers are fond of saying that demography is destiny, and in California’s case, there is a lot to be learned before we fulfill our destiny. The second objective is to remind Californians that public policy is fundamentally tied to politics at the local level, where tensions between the old and the new are most likely to be felt. These tensions may resolve themselves gradually, smoothly, even indiscernibly. Or the demographic fault lines could shift precariously, perhaps through the workings of either representative or direct democracy. The prudent course is for California’s leaders to be ready for either outcome, and PPIC intends to map that course with reliable research. David W. Lyon President and CEO Public Policy Institute of California iv Summary Immigration has transformed California’s ethnic landscape. Since 1965, the Latino and Asian populations have grown rapidly, and demographers project that by 2030, no ethnic group will constitute more than 50 percent of the state’s population. This transformation has important implications for government policy and electoral politics. Given California’s history, it also raises questions about how much and what kind of ethnic group competition and conflict is likely to occur over the next decades. This increased diversity of California’s population has complicated its ethnic politics. In many cases, racial and ethnic issues cannot be reduced to disagreements between a white majority and a coalition of ethnic minorities. Blacks, Latinos, and Asian Americans may share an interest in anti-discrimination policies, but they compete for political offices and jobs, especially at the local level. Likewise, the main immigrant groups have shared concerns, such as bilingual education, which may not be a priority for blacks. In this context, competition among minority groups over the allocation of public benefits is commonplace, and the explosive growth of the Latino and Asian communities has generated a larger and more complex debate over ethnicity and public policy. In this study, we explore how reactions to ethnic diversity relate to important policy questions. These questions include, but are not limited to, whether government should use ethnicity as a criterion for distributing public benefits. Analyzing a survey of California voters conducted just before the vote on Proposition 209, which forbids the use of such preferences by public agencies in California, we investigate how personal experiences with and attitudes toward other ethnic groups might influence attitudes and electoral choices about racial policies. v Research Design and Methodology The study assumes that one’s ethnic context—which we define as the ethnic composition of one’s neighborhood—affects the frequency and character of one’s interactions with members of different groups. Our main concern is to identify the influence, if any, of such contexts on racial attitudes and behavior. In doing so, we distinguish between two hypotheses. The compositional hypothesis recognizes that people living in ethnically mixed neighborhoods may differ systematically from those who do not. For example, whites who live in such neighborhoods may be younger or more economically vulnerable than those who live in predominantly white neighborhoods. If so, these characteristics could shape their attitudes and behavior more than the ethnic composition of their neighborhoods as such. Thus, the compositional hypothesis is alert to the ways in which ethnic contexts might reflect, rather than shape, attitudes and personal characteristics. In contrast, the contextual hypothesis focuses on the extent to which ethnic context actually shapes attitudes and behavior. Although these hypotheses can be distinguished analytically, in practice they are difficult to disentangle. The fact that people can and do choose their residences—that is, people are not assigned to neighborhoods randomly—creates a causal ambiguity that cannot be ignored or eliminated in any study of this type. Even so, this study differs from previous attempts to understand these contextual effects in several ways. First, the unit of analysis is the consolidated precinct, where some degree of face-to-face contact can be assumed. Second, this study examines the relationships between ethnic context, racial attitudes, and voting patterns on a specific measure, namely, Proposition 209. Third, most reports focus on contact between two ethnic groups, but we consider white reactions to several minority populations. Finally, we analyze the influence of ethnic context on minority group members as well as on whites. The data come from a unique survey of the California electorate conducted just before the vote on Proposition 209 in 1996. The survey asked respondents about Proposition 209, neighborhood diversity, racial attitudes, racial identity, political ideology, personal finances, and education. To determine whether genuine contextual effects are present, vi information about the respondents’ social backgrounds and attitudes are linked to data about their localities. The statistical analyses thus estimate the effect of ethnic context by controlling for other individual-level variables. We deliberately oversampled areas with high concentrations of ethnic minorities to facilitate meaningful comparisons across ethnic groups. We also limited the sample to registered voters to focus more precisely on the political effects of ethnic contexts. Major Findings and Implications The study revealed substantial differences across ethnic groups. Compared to the other ethnic groups, whites were less likely to say that their race or ethnicity was an important part of their political identity. They were also more likely to support Propositions 209 and 187 and to regard affirmative action programs as unnecessary and unfair. These differences cannot be explained by either ethnic context or the other factors we considered, such as party, ideology, or social background. At the same time, there was consensus among the four ethnic groups on a number of questions related to the state’s ethnic composition. A large majority in every ethnic group was either neutral or positive about the effect of people of a different ethnic or cultural group moving into their neighborhood. They also were united in expressing more negative views about the effect of illegal immigrants than any other group. Despite differing attitudes about the need for affirmative action, respondents from all four ethnic groups tended to agree on the egalitarian principles of merit and nondiscrimination as the primary criteria for job promotion and educational opportunity. A majority in every ethnic group seemed to favor the effect of its own group on the neighborhood and to agree that people tended to be happier living and socializing with others of the same background. Whites seemed most worried about the effects of a heavy influx of black residents on property values; blacks were most concerned about the prevalence of hostile attitudes in a largely white community. It would be highly misleading, however, to conclude that Californians favor residential or social segregation. As other studies have shown, people of all ethnic groups indicate that they prefer to live in neighborhoods where members of their own group make up the majority, but not all, of the vii neighborhood’s households. In addition, realism rather than prejudice may underlie the belief that it is easier to socialize with people of similar backgrounds. This “separatist” opinion does not imply that people do not value friendships with members of other ethnic groups. Indeed, the rising rates of intermarriage in California between whites and Latinos and between whites and Asians belie such a claim. Although respondents expressed considerable doubt about the feasibility of achieving a “colorblind society,” the data do not contradict public acceptance of diversity in the workplace, in higher education, or in public life generally. The second major conclusion of this study is that multiethnic neighborhoods are not the primary locus of ethnic political divisions in California. Previous studies have argued for the so-called “threat hypothesis,” which predicts that racial tension will be highest where different groups interact the most. In general, however, the data indicate that perceptions and attitudes varied little by ethnic context. Indeed, where there were differences between whites and nonwhites, they often were in the direction of greater tolerance and agreement between whites and nonwhites in mixed areas than in homogeneous ones. These findings suggest that California’s political climate on racial and cultural issues will depend in large part on the perceptions of the more numerous whites who live in majority white areas. We found that living in an ethnically mixed neighborhood boosted feelings of ethnic and racial identification for all four groups. How such identifications affect political conduct is a more complicated question. From the perspective of minority groups, a stronger sense of racial and ethnic identity may fuel efforts to advance collective goals and redress inequality. Racial and ethnic identifications among minorities were stronger in areas with higher densities of minorities, a finding that suggests that efforts at political organization may be more successful in such neighborhoods. However, foregrounding ethnic identity also may foster backlash and intensify group conflict. The third major conclusion is that attitudes on racial and ethnic issues do not seem to be formed at the local level. The few observed neighborhood variations seem to be compositional rather than contextual: That is, these variations were best explained by differences in the ideologies, party affiliations, and background characteristics of viii respondents. This pattern indicates that racial attitudes are largely acquired through experiences that cut across localities. To use the example of Proposition 209, voting intentions were not influenced by what residents perceived in their local areas so much as by their general political orientations and what they learned from the respective pro- and anti-209 campaigns. This finding suggests that mass media, for example, may be more critical to racial and ethnic attitudes than are local contexts. Our final major conclusion concerns the relatively negative feelings many Californians have about undocumented immigrants. Respondents from all four ethnic groups had distinctively more positive views about legal immigrants than illegal ones. One implication of this finding is that policymakers should be careful not to lump legal and illegal immigrants together in assessing the public support for programs that affect these groups. Policymakers might also emphasize the fact that the legal Latino population is much larger than the undocumented group. In addition to reducing resentment about immigration, this measure might increase the public’s willingness to support policies that aid the larger Latino community. In sum, there is no evidence that ethnic group relations are particularly troubled in highly diverse areas or that these areas will serve as cauldrons for future problems. Racial attitudes seem to be influenced more by one’s political and social background than by one’s local context. Resistance to ethnic diversity is not more likely among those who directly experience it in their neighborhoods. We conclude that ethnic tensions are not preordained to increase as Californians continue to cope with the state’s shifting ethnic landscape. ix Contents Foreword..................................... iii Summary..................................... v Tables ....................................... xiii 1. THE POLITICS OF ETHNICITY IN A CHANGING CALIFORNIA .............................. Purposes of This Study ......................... Residential Patterns ........................... The Motivational Bases of Contextual Effects ........... Methodology ............................... Implications for Public Policy ..................... 1 3 5 7 10 11 2. NEIGHBORHOOD CONTEXT AND REACTIONS TO ETHNIC DIVERSITY ......................... Group Perceptions at the Neighborhood Level .......... Group Perceptions Across Ethnic Contexts ............ The Influence of Subjective Contexts ................ Multilevel Analysis of Context Effects ................ Summary of Findings .......................... 13 14 18 21 23 25 3. RACIAL ATTITUDES AND ETHNIC CONTEXT IN CALIFORNIA .............................. Racial and Ethnic Identification ................... Images of Race Relations ........................ Social Distance .............................. Affirmative Action ............................ Explaining Ethnic Identification and Attitudes Toward Race Relations ............................... Results for Ethnic Identification ................... Results for Beliefs About Race Relations .............. Results for Beliefs About Affirmative Action ........... Summary of Findings .......................... 27 28 31 34 36 39 40 42 43 44 xi 4. ETHNIC CONFLICT, ETHNIC CONTEXT, AND VOTING ON PROPOSITION 209 ................ Racial Threat as an Explanation ................... Accounting for the Vote on Proposition 209 ........... Reviewing the Role of Context .................... 47 48 53 57 5. CONCLUSION ............................. Appendix A. Survey Sample and Data Sources ................... B. Multivariate Analyses from Chapters 2 and 3 ........... C. Survey Question Wording ....................... Bibliography .................................. About the Authors ............................... 61 67 69 77 83 87 xii Tables 2.1. Neighborhood Effect of Different Groups, by Race/Ethnicity of Respondents ................. 2.2. Neighborhood Effect of Immigrants, by Race/Ethnicity of Respondents ........................... 2.3. Neighborhood Effect of Different Groups, by Race/Ethnicity and Context ................... 2.4. Neighborhood Effect of Immigrants, by Race/Ethnicity and Context ............................. 2.5. Neighborhood Diversity, by Neighborhood Effect of Different Groups .......................... 2.6. Neighborhood Diversity, by Neighborhood Effect of Immigrants ............................. 3.1. Racial and Ethnic Identification, by Race/Ethnicity of Respondents ............................. 3.2. Racial and Ethnic Identification, by Race/Ethnicity and Context ................................ 3.3. “Color-Blind Society,” by Race/Ethnicity of Respondents ............................. 3.4. “Color-Blind Society,” by Race/Ethnicity and Context ................................ 3.5. “Minorities Have Less Opportunity,” by Race/Ethnicity of Respondents ........................... 3.6. “Minorities Have Less Opportunity,” by Race/Ethnicity and Context ............................. 3.7. “People Happier with Others of Same Background,” by Race/Ethnicity of Respondents ................. 3.8. “People Happier with Others of Same Background,” by Race/Ethnicity and Context ................... 3.9. “Minorities Gain at Expense of Others,” by Race/ Ethnicity of Respondents ..................... 3.10. “Minorities Gain at Expense of Others,” by Race/ Ethnicity and Context ....................... 16 17 19 20 22 23 29 30 31 32 33 33 35 35 36 37 xiii 3.11. “Minorities Get More Than They Deserve,” by Race/Ethnicity of Respondents ................. 3.12. “Minorities Get More Than They Deserve,” by Race/Ethnicity and Context ................... 3.13. Multivariate Analyses of Racial Attitudes ........... 4.1. Percentage Voting “Yes” on “Ethnic” Initiatives, by Race/Ethnicity ............................ 4.2. Objective Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity .............. 4.3. Perceived Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity .............. 4.4. Percentage Answering “Yes” to Affirmative Action Questions, by Race/Ethnicity and Ethnic Context ..... 4.5. Effects of Race/Ethnicity, Racial Context, and Racial and Ethnic Identification on Proposition 209 Vote ....... 4.6. Explaining Voting Intentions on Proposition 209 ..... B.1. Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 ................ B.2. Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 ................ B.3. Racial Identification and Racial Attitudes—Black Respondents, from Chapter 3 .................. B.4. Racial Identification and Racial Attitudes—Latino Respondents, from Chapter 3 .................. B.5. Racial Identification and Racial Attitudes—Asian Respondents, from Chapter 3 .................. B.6. Racial Identification and Racial Attitudes—White Respondents, from Chapter 3 .................. 37 38 41 48 49 51 52 54 55 70 71 72 73 74 75 xiv 1. The Politics of Ethnicity in a Changing California Immigration has transformed the ethnic composition of Californian society. Since 1965, the Latino and Asian segments of the population have grown rapidly, and demographers project that by 2030, California will be a “majority minority” state.1 The regional concentration and age distribution of California’s population also are changing because of the residential concentration of recent immigrants and because they tend to be younger and tend to have larger families than the mainly white or black native residents.2 These demographic trends have important implications for both government policy and electoral politics. An obvious example is the effect of the influx of non-English speakers on the state’s public schools. The changing ethnic composition of California’s schoolchildren has fueled controversy over bilingual education programs and reshaped affirmative action programs in higher education. On the political front, the growing number of ethnic minorities and their geographic location has an immediate effect on redistricting and reapportionment, with consequences for minority representation in the state legislature and Congress. More generally, ethnic differences in voter turnout and party identification have a direct effect on electoral outcomes. Historically, rapid change in the ethnic composition of California society has engendered group competition and conflict. Beginning with the Gold Rush, the pull of labor demand and the push of poverty led to ____________ 1The term “majority minority” is used to refer to a situation where blacks, Latinos, and/or Asian Americans constitute a majority of the population in an area (and where non-Hispanic whites, therefore, make up less than 50 percent of the population) according to the 1990 Census. 2 In this monograph, we will be referring to the many different racial and ethnic groups in California. We want to clarify that when we use the term “whites,” we are referring to non-Hispanic whites. 1 successive waves of immigration to California from China, Japan, the Philippines, and Mexico. When the influx of new arrivals was particularly heavy and economic times turned bad, communal violence erupted and legislation was passed restricting immigration (Olzak, 1992). Thus, experience suggests that the current demographic changes might create new tensions in ethnic group relations. The very diversity of California’s population, however, complicates the nature of ethnic politics in the state. The assumption that disagreement over public policy generally reflects conflict between a white majority and a coalition of ethnic minorities seems simplistic. The civil rights movement centered on black-white relations and the need to improve conditions for the former group, but the explosive growth of population in Latino and Asian communities has created an expanded and more complex policy agenda. In this context, competition among minority groups over the allocation of public benefits is commonplace. For example, blacks, Latinos, and Asian Americans may share an interest in anti-discrimination policies, but they contend for scarce political offices and jobs, particularly at the local level. Affirmative action in college admissions benefits blacks and Latinos more than Asians. On the other hand, the main immigrant groups have concerns—in the area of bilingual education and the access of newcomers to government services, for example—that may not be a priority for blacks. A significant contrast between the Watts riot of 1965 and the Rodney King riot of 1992 illustrates the growing complexity of ethnic relations in California. The earlier outbreak of violence largely expressed the hostility of black residents toward the predominantly white Los Angeles police force; the more recent eruption revealed antagonism between blacks and Koreans as well as between blacks and whites. The climate of opinion facing recent Americans differs from that facing the earlier waves of newcomers from Europe in a way that may affect ethnic group relations. In the early 20th century, belief in assimilation and Americanization was pervasive among political elites. Today, a “multiculturalist” perspective stressing themes of ethnic identification and group rights challenges the traditional individualist ethos; the question of how government should use ethnicity as a criterion for distributing public benefits is highly controversial. 2 Public opinion influences policy formation in a democratic society, if only indirectly by limiting the range of choices elected officials can safely contemplate. In California, though, the increasing use of direct democracy for deciding important issues magnifies the effect of the views of ordinary citizens. Since the passage of Proposition 13 in 1978, voting on initiatives has determined state policy on taxing and spending, environmental protection, gay rights, gun control, auto insurance, language rights, illegal immigration, and affirmative action. Clearly, one stimulus for direct legislation is an impasse between the governor and the legislators; another source of the recent trend toward government by initiative is the interplay between institutional structures and current demographic developments. The ideological fulcrum in California differs in legislative and initiative politics. Legislative districts are drawn on the basis of population, not citizenship or even legal residency. Thus, immigrants who cannot vote provide additional legislative seats for certain geographic areas. Since these immigrants often cluster in ethnic “enclaves,” representation in the California legislature is weighted toward Democrats and minorities with liberal policy attitudes. On the other hand, the statewide electorate, which decides the gubernatorial race and ballot initiatives, is whiter, older, and wealthier than the population as a whole. Accordingly, outcomes in these contests tend to be more heavily influenced by Republican and conservative voters. This contrast in constituencies helps explain the passage of recent initiatives on illegal immigration, affirmative action, and bilingual education, when voters reversed policies with entrenched legislative support, and it is highly probable that the electoral arena will see continued battles on ethnic issues. Purposes of This Study This study describes public attitudes toward the changing ethnic composition of California and explores how these reactions to ethnic diversity are related to people’s preferences on important policy questions. The vote on Proposition 209, the initiative forbidding the use of racial or gender preferences by public agencies in California, is a useful case study for assessing the effect of the demographic, ideological, and 3 institutional factors briefly described above. Affirmative action uses an applicant’s race or ethnicity as one criterion for allocating places in college, government jobs, and public contracts. A frequent justification for this policy is that fairness requires that the distribution of such benefits should mirror the state’s ethnic diversity. This monograph analyzes a survey of California voters conducted just before the vote on Proposition 209 to investigate how personal experiences with and attitudes toward other ethnic groups influenced both electoral choices and opinions about racial policies. The focus of the empirical inquiry is the influence of one’s personal context—defined as the ethnic composition of one’s neighborhood—on the outlook of each of California’s four main racial and ethnic groups. More specifically: 1. How do subjective experiences of diversity differ across ethnic groups? By comparing people grouped by both ethnicity and residential context, Chapter 2 describes whether group conflict is more or less widespread in ethnically diverse locales. 2. What are Californians’ hopes and expectations concerning race relations? Do they anticipate progress toward a color-blind society with minimal discrimination? How strong is their sense of racial and ethnic group identity, and how does this influence the policy preferences of different groups? Chapter 3 assesses the strength of ethnic group identifications and the influence of these feelings on policy positions of Californians living in different residential contexts. 3. Does a strong sense of racial or ethnic identity have electoral consequences, boosting support for Proposition 209 among whites while enhancing opposition among minority groups? Chapter 4 addresses this issue and investigates whether contact with other ethnic groups, as indexed by one’s residential context, influenced voting on the ballot initiative and whether this effect was similar across ethnic groups. 4 Residential Patterns In a state as large and varied as California, there is no one typical residential mix. The state as a whole has a high level of ethnic diversity, but regional and local profiles vary widely. Some counties reflect the diversity of the state as a whole; others are virtually all white. Similarly, residential segregation persists within diverse areas; within Los Angeles County, for example, the Palos Verdes Peninsula, the Santa Monica Mountains, and the newer northern cities such as Santa Clarita are relatively homogeneous. In California, as nationally, ethnic heterogeneity at the aggregate level tends to mask uniformity at the level of tracts or blocks (Clark, 1996). Using precinct-level information derived from the University of California, Berkeley’s, Statewide Database, we determined that 86 percent of whites in California live in neighborhoods where more than half of the residents share their racial background (majority white areas). By contrast, 29 percent of the state’s blacks live in areas with a majority of blacks, and 44 percent of its Latinos live in majority Latino areas. To the extent that people live in neighborhoods populated by a majority of their own racial or ethnic group, more diversity at the state level may not mean a high level of intergroup contact in one’s daily life. Ethnic homogeneity in residential neighborhoods is not a matter of law but of personal choice and financial ability. In statistical terms, whites are the ethnic group in California most likely to live among people of the same background. Studies show that both whites and blacks would prefer not to live as an isolated minority in their neighborhoods, and that living with a majority of one’s own race is the dominant preference in both groups (Farley et al., 1994). Among whites, this preference rests partly on stereotypes of blacks that engender the belief that as the ratio of blacks to whites increases, property values decline. Among blacks, a reluctance to move into mainly white neighborhoods partly reflects fear of a hostile reception. In other words, attitudes toward one’s own and other ethnic groups influence the composition of neighborhoods by shaping individual choices to move in or out. Therefore, how people react to their neighborhood’s ethnic mix is an important element in the changing pattern of California politics. 5 In this study, the term “context” refers to the composition—ethnic or economic—of a particular area. Depending on one’s analytic purpose, the contextual unit could be the precinct, census tract, county, state, or even nation. The focus here is on the ethnic context of the neighborhood, measured by the relative proportions of the four main ethnic groups in California. Context can thus refer to the overall proportion of all minority groups, to the proportion of a particular group such as blacks or Latinos, or simply to whether or not a locality is a “majority minority” area. This study assumes that ethnic contexts affect the frequency and character of interactions between members of different groups. Its main concern is to identify the effect of the makeup of one’s locality on racial attitudes and behavior. In a statistical analysis delineating these effects, if any exist, on the outlook of individuals of a particular race or ethnicity, it is necessary to distinguish between compositional and contextual hypotheses. The compositional interpretation recognizes that whites living in majority minority locales may differ systematically from those who live in mainly white areas in ways that influence racial attitudes. They may be younger, for example, and therefore perhaps more liberal, or they may be economically vulnerable, and therefore more likely to scapegoat minority group members. More generally, prior attitudes and stereotypes can influence one’s residential choices, so that racially tolerant whites are willing to remain in an area experiencing an influx of minorities while their more prejudiced counterparts take flight. To the extent that people vote with their feet in this way, differences across ethnic contexts might simply reflect variation in the demographic and attitudinal characteristics of the particular ethnic group members living there. The alternative contextual interpretation searches for effects that go beyond the aggregation of individual-level differences. This hypothesis asserts that the residential setting itself influences the outlook of those who experience it, whatever their prior predispositions or background. The notion of a contextual effect thus emphasizes that individuals are not just autonomous units but rather are shaped by social interaction. Ethnic context is an aggregate-level concept that depends on the characteristics of individuals in a particular area, but its influence on behavior is 6 separate from the effects of these background factors on these individual actors. The power of cultural norms is one mechanism for the influence of the local community. Social interactions condition the flow of information, reinforcing some beliefs and undermining others through exposure and conformity pressures. For example, a Democrat living in a community composed mainly of partisans will tend to find her views shared by her neighbors, whereas a Republican in the same area may modify his views to fit in better with the majority. To the extent that one self-selects friends and social contacts, the psychological tendency to prefer similar and reinforcing views will dominate one’s interactions, but the opportunities for such self-selection are limited to some degree by the externally imposed context. Finding a fellow-conservative in North Berkeley, for example, is not easy. The Motivational Bases of Contextual Effects Prior research on the political effect of ethnic context has centered on white responses to the size of the black population in their community. Following V. O. Key’s famous study of the ‘black belt” in the South, a conventional hypothesis is that the larger the proportion of blacks, the more hostile or prejudiced the political outlook of white residents (Key, 1949; Blalock, 1967; Glaser, 1994). Large concentrations of blacks are said to represent a threat to whites because of ethnic competition for economic, political, and cultural power. Numbers matter because voting and market power help determine the allocation of resources across ethnic groups. Accordingly, large concentrations of blacks make the economic and political threat more acute and visible, heightening feelings of vulnerability among whites and increasing their antipathy to minority political demands. In other words, as minority populations in their immediate context grow in size, the resulting diverse ethnic context increases the likelihood of defensive and prejudiced reactions among whites. Transporting Key’s “threat” hypothesis from the South to California involves broadening the definition of ethnic context to refer to the concentrations of Latino and Asian as well as black residents. However, theoretical discussions (Taylor, 1998) generally suggest that 7 concentrations of other minorities will yield the same kind of antipathy among whites as would a larger share of blacks. This means that the more diverse contexts within California should be characterized by a heightened sense of ethnic consciousness among all groups and by political polarization along ethnic lines. Clearly, the “threat” hypothesis implies pessimism about the possibility of achieving harmony in a multiethnic society. An alternative hypothesis with some support in social psychology (Allport, 1954; Prentice and Miller, 1999) is the more optimistic “contact” theory, which proposes that direct experience with members of other groups tends to ease conflict and promote mutual tolerance and understanding. This occurs partly through the erosion of prior stereotypes and partly because the necessities of peaceful coexistence help develop norms of compromise and power-sharing where there are large numbers of minority group residents. If the processes identified by the “contact” theory prevail, then ethnically diverse contexts should be characterized by more positive images of race relations and greater support among whites for policies designed to assist minorities, such as affirmative action. Research indicates that contact between racial groups reduces conflict and prejudice only under special conditions (Stephan, 1987). Interdependence, common goals, equal status, and encouragement by authorities are necessary for intergroup contact to promote tolerance and goodwill. The mere presence of a large number of minority residents in a locality does not guarantee these conditions. Nor does it assure intimate contact with someone of a different background, particularly when the contextual unit is large. Previous studies regarding the relationship between the numerical size of the black community and racial hostility among whites also give mixed support for the threat hypothesis. For example, Voss (1996) concludes that the size of the black population in one’s locality did not boost white voting for David Duke in Louisiana. By contrast, analysis of the 1990 national General Social Survey (Taylor, 1998) found that traditional prejudice and opposition to race-targeted policies among whites in metropolitan areas swell as the local black share of the population expands. However, concentrations of local Asian American 8 and Latino populations did not engender white antipathy toward these groups (Taylor, 1998; Hood and Morris, 1997). These inconsistent results may reflect differences among particular localities (Southern counties versus metropolitan areas nationwide), in the extent of white flight across these localities, or in the dependent variables analyzed (racial attitudes and policy positions versus votes). The fact that people can choose their ethnic context rather than fall into a locality more or less randomly, say by the accident of birth, creates a causal ambiguity complicating analyses of the effect of demographic diversity in today’s California. “Threat” promotes a positive association between the prevalence of minorities and prejudice, whereas “white flight” predicts a negative correlation between them, with prejudice more widespread in all-white areas (Voss, 1996). Unless these offsetting processes can be disentangled, which is difficult with cross-sectional data, the interpretation of results can only be tentative. This is an important caveat for the present study also. However, the research reported here differs from previous studies of contextual influences in several ways. First, the unit of analysis is the neighborhood (i.e., specifically, consolidated precincts), a relatively small area where one can assume face-to-face contact (although not necessarily real social interaction). Second, we examine the effect of ethnic context on both racial attitudes and voting on a specific measure affecting the distribution of public benefits among ethnic groups, namely, Proposition 209. Third, as noted above, we consider white reactions to several distinct ethnic minority populations. Finally, we are almost unique in analyzing the influence of ethnic context on the conduct of minority group members as well as whites. The threat hypothesis applies only to the reactions of the dominant group to ethnic change. Theoretically, the size of minority groups in a given area should have quite different consequences for the political behavior of racial and ethnic minorities, representing an opportunity rather than a danger. A heavy concentration of Latinos, for example, should improve their capacity for political mobilization. The density of social networks and the frequency of communications facilitate the development of a distinctive ethnic consciousness and the formation of organizations to advocate on behalf of group interests. On the other 9 hand, where an ethnic group is a small and scattered minority, as are blacks in many parts of California, group assertiveness should be more difficult, both psychologically and organizationally. Methodology This study addresses two basic questions: the nature of ethnic group differences in racial attitudes and political preferences in California, and the influence of local contexts on politics in a multiethnic society. The research design is tailored to these purposes and to overcome several methodological problems identified by previous analyses of contextual influences. The most significant problem is the need for a multilevel analysis based on the merger of micro-level and macro-level information. Confining comparisons to differences across contexts does not allow one to choose between the contextual and compositional hypotheses outlined above because there can be no full adjustment for the effects of the distribution of individual-level attributes on electoral choices. To identify contextual effects, therefore, one must be able to link survey information about particular individuals with census information about the respondents’ localities. Another difficulty relates to the frequent tendency to define context at the county, state, or national levels of aggregation. This is problematic because the territorial unit that denotes a psychologically meaningful context, regardless of whether one adheres to the “threat” or the “contact” hypothesis, may be smaller than the state or even the county; aggregation can mask meaningful variation. In addition, statewide samples typically underrepresent minority group members, particularly the poor and less-educated ones. To address these methodological issues, we took the data for this study from a unique survey of the California electorate conducted just before the vote on Proposition 209 in 1996. Respondents were sampled from a stratified set of contexts, and to determine whether genuine context effects are present, information about respondents’ social background and attitudes are linked to data about their localities. The crucial statistical analyses thus estimate the effect of ethnic context, controlling for theoretically relevant individual-level variables. Because of a primary interest in the implications of California’s increasing ethnic 10 diversity on group relations, we focus on the potential effect of the ethnic composition of neighborhoods rather than on the influences of other contextual attributes such as economic profiles. Hence, information about economic context was used only to confirm that any contextual effects detected were not spurious. Our sampling strategy used the electoral precinct (a good proxy for local neighborhood) as the jurisdiction defining a context. We deliberately oversampled areas in California with a high concentration of ethnic minorities to facilitate meaningful comparisons across ethnic groups (see Appendix A for details about the survey sample and data sources.) Finally, to focus attention more precisely on the political effect of ethnic contexts, we limited the sample to the registered voters who have been shaping official policy toward ethnic group relations through voting on initiatives such as Proposition 209. The remaining chapters of this monograph proceed as follows. We first describe the relationships between respondents’ ethnicity and residential context and their racial attitudes and voting intentions separately. We then report the results of a multivariate analysis incorporating both individual-level and contextual variables to determine whether or not any apparent contextual effects reflect more than differences in the ethnic composition of localities. These analyses are conducted for the sample as a whole and then for each ethnic group separately. Implications for Public Policy Policymakers in a diverse community have a broad interest in forging consensual solutions to tensions arising from ethnic group interaction and competition. The demographic trend toward increasing ethnic diversity in California is unlikely to be reversed. Neither is the existence of widespread variation in the degree of ethnic heterogeneity across residential locales likely to be reversed. The public opinion data examined here identify the fundamental reactions toward ethnic diversity among the varied racial and ethnic groups making up the state. To what extent is the presence of a growing minority population in California perceived as a threat, and by whom? Are attitudes in multiethnic contexts more or less hostile or ethnocentric? Answers to these questions 11 can provide clues about popular support for policies concerning immigration, housing, or local education by locating the sources of intergroup conflict, both geographically and demographically. The existence of contextual influences on reactions to members of other ethnic groups may indicate the effect of direct contact and personal experience, as opposed to that of media communications, on opinion formation. The nature of the relationship between ethnic diversity and group conflict may indicate not only where to concentrate efforts to alleviate racial and ethnic tensions, but also which mechanisms are likely to be effective in fostering harmony and goodwill. For example, one might learn how to structure programs for community organization, immigrant integration, and public service delivery at the local level. We discuss the policy implications of the study’s findings in the concluding chapter. 12 2. Neighborhood Context and Reactions to Ethnic Diversity Increasingly, California’s changing demography is bringing members of different racial and ethnic groups into more personal contact with one another. This is not uniformly the case, as we noted above, because some parts of California have become much more diverse while others have remained relatively homogeneous. The design of this study tries to capture the heterogeneity of neighborhood settings to discover whether local experiences make a difference in racial and ethnic attitudes. We begin with relations between groups and their perceptions of one another. Group perceptions lie at the core of race relations. When racial and ethnic groups view each other in a hostile manner, it erodes the basis for harmonious racial policies and makes building consensus difficult, if not impossible. People become aware of different racial and ethnic groups when people of varying backgrounds move into their neighborhoods, perhaps for the first time. In the course of daily life, they observe their neighbors’ comings and goings. Children from various families intermingle at school and play. Neighbors confront common problems such as traffic, drugs, crime, or the appearance of their surroundings. These shared concerns often engender discussion and group meetings to forge common solutions. Social interchange of this kind informs and alters stereotypes and evaluations of members of other racial and ethnic groups. Context affects learning by shaping direct experience and firsthand observation. Of course, group perceptions are not solely determined by what happens at or near one’s home. People from different racial and ethnic backgrounds also interact while commuting, shopping, working, or studying. Because many Californians do not work near their homes, the ethnic context of workplace relationships may differ from those in their 13 residential neighborhoods. Which setting is likely to influence racial attitudes is therefore an empirical question. For example, it might be that job and school experiences are more relevant than neighborhood experiences to people’s beliefs about affirmative action, since those are the locales where one might have encountered or heard about racism or “reverse discrimination.” In addition to direct learning about other ethnic groups in the neighborhood or at work and school, there are important indirect, nonexperiential sources of attitudes, such as the mass media, popular culture, and reference groups such as political parties. Television and the movies are an obvious source of imagery about other groups. For example, if movies typically cast blacks as criminals, whites as victims, and Asians as shopkeepers, they may create or reinforce stereotypes about people one rarely encounters. Political messages similarly can transmit messages about what a particular ethnic group is like. The question, then, is whether learning at the neighborhood level, indexed in part by the ethnic composition of the locality, matters, given the well-documented strength of early socialization, the media, and the larger political culture. Does living in a diverse area and being exposed to people of different racial and ethnic backgrounds foster positive or negative perceptions of other groups? The answers are important because if neighborhood learning matters and the number of diverse areas is growing, the content of local interactions will play a larger role in determining the quality of race relations in California. The nature of contextual influences also can indicate to policymakers whether they should target the more diverse areas as potentially problematic for ethnic harmony or whether it is more effective to address race relations in a more global way (e.g., such as through changing media images or general education). Group Perceptions at the Neighborhood Level To explore group perceptions, respondents in our survey were asked a series of questions about the effects that different groups of people have had on the quality of life in the neighborhoods where they live. In particular, they were asked to say whether the presence of particular groups improved, made worse, or had no effect on their neighborhoods. 14 The groups in question were whites, blacks, Latinos, Asians, legal immigrants, and illegal immigrants. We also asked our survey respondents to indicate whether they themselves lived in an area that was becoming more diverse, less diverse, or staying about the same. As a general observation, we found that California’s racial and ethnic groups were more likely to hold positive than negative perceptions of one another. Given some of the highly publicized racial conflicts in this state over the last decade, this was not a foregone conclusion. At the time of our survey California had only recently emerged from a serious economic recession, and contentious ballot measures dealing with ethnic issues (i.e., Propositions 187 and 209) had generated substantial protest and political turmoil. Nevertheless, among all four major racial and ethnic groups, negative images of ethnic diversity were very limited. Consider the evidence of Table 2.1, which examines beliefs about how blacks, Latinos, Asian-Americans, and whites, respectively, affect the quality of life in one’s neighborhood. No more than 10 percent in any ethnic group felt that the presence of blacks made neighborhood conditions worse, whereas over 89 percent thought that it made no difference or improved conditions. Overall, perceptions of the neighborhood effect of whites and Asians were even less negative, and perceptions of Latinos only marginally more critical. The most common view of Californians toward the presence of other ethnic groups in one’s neighborhood was neutral, not prejudiced. These data also reveal a general tendency of minority respondents to perceive the neighborhood effect of their own racial and ethnic group most positively. For example, blacks were twice as likely as Asians or Latinos, and almost three times as likely as whites, to say that the presence of more blacks had improved neighborhood conditions. The findings in Table 2.1 suggest that Asians are the most ethnocentric group, with over 30 percent saying that the presence of fellow Asians improved their neighborhoods. Only whites deviated from the norm of preferential assessment of their own group’s effect, perhaps because group identity is less salient for members of the dominant ethnic group (Prentice and Miller 1999). White respondents also were somewhat less likely to view the presence of minority groups in their neighborhoods positively. The second column of Table 2.1, for example, indicates that 15 16 Whites (n = 603) Blacks (n = 217) Latinos (n = 408) Asians (n = 270) Table 2.1 Neighborhood Effect of Different Groups, by Race/Ethnicity of Respondents (responses in percent) Effect of Blacks Im- No Made proved Effect Worse 11 81 8 28 62 10 13 81 6 15 79 7 Effect of Latinos Im- No Made proved Effect Worse 14 74 13 21 66 13 28 63 9 13 80 7 Effect of Asians Im- No Made proved Effect Worse 18 74 8 17 79 4 20 75 6 33 65 3 Effect of Whites Im- No Made proved Effect Worse 17 80 3 19 74 7 19 76 5 24 72 4 the ratio of favorable to unfavorable assessments of the effect of Latinos among black and Asian respondents was about two to one, compared to an even split among whites. In the next chapter, we report additional evidence that people tend to prefer living among others with the same ethnic background. So, although California’s ethnic groups do not oppose diversity at the local level, a high level of heterogeneity and integration is not necessarily their preferred choice. The notable exception to the finding of more positive than negative perceptions of a group’s effect on the neighborhood concerns beliefs about illegal immigrants. Table 2.2 reports responses to questions about the effect of both legal and illegal immigrants on the quality of neighborhood life, with answers broken down by the racial and ethnic identity of the respondent. In a familiar pattern, legal immigrants enjoyed approximately a three to one advantage in positive to negative perceptions, whereas opinion about the effect of illegal immigrants ran two to one in the opposite direction. This balance of opinion prevailed in all ethnic groups, including Latinos and Asians, despite the fact that the majority of the undocumented population in California is made up of these groups, especially Latinos. Here it should be noted that respondents in this study are registered voters who necessarily have lived in the United States at least five years themselves (for reasons of citizenship status) and may be less sympathetic to the illegal newcomers. Still, the predominant evaluation of illegal immigrants, as for all other groups, is neutral, not hostile. Table 2.2 Neighborhood Effect of Immigrants, by Race/Ethnicity of Respondents (responses in percent) Whites (n = 603) Blacks (n = 217) Latinos (n = 408) Asians (n = 270) Effect of Legal Immigrants Im- No Made proved Effect Worse 19 73 8 21 71 8 25 68 7 27 69 5 Effect of Illegal Immigrants Im- No Made proved Effect Worse 7 65 27 9 70 21 12 70 18 9 69 23 17 Group Perceptions Across Ethnic Contexts The next question to address is whether perceptions of other ethnic groups, which differ little by the respondent’s ethnicity, vary across ethnic contexts. Specifically, is the balance of opinion about a group’s influence on the neighborhood more negative in ethnically diverse than ethnically homogeneous areas? The relevant evidence is displayed in Tables 2.3 and 2.4. These tables examine both the respondent’s race and the type of neighborhood that he or she lives in. That is, they distinguish between blacks in majority black areas, blacks in majority minority areas, blacks in majority white areas, and so on. The dominant result in this array of statistics is that responses did not vary in a dramatic way by neighborhood. Overall, people were as positive about the effect of other ethnic groups in the mixed areas as they were in the homogeneous areas. Take, for instance, perceptions of the effect that blacks have on the neighborhood. Table 2.3 shows that contrary to the threat hypothesis, there was no more antipathy to blacks among whites in majority minority than majority white areas. The same is true when we compare the opinions of Asians in majority minority areas and Asians in majority white areas. For the most part, the differences were small for Latinos and for blacks as well. If anything, blacks in majority white areas tended to be relatively more negative toward the presence of their own group (and other minorities) in their neighborhoods than were blacks in majority black or majority minority areas. But, once again, the modal response for all settings and groups was to think that the presence of other groups, including illegal immigrants, does not, per se, make the neighborhood either better or worse off. The slight tendency of whites in majority minority areas to have more positive perceptions of racial and ethnic minority groups than whites in majority white areas is worth noting. However, this could be due to compositional differences in their party affiliation or ideology as much as to context-driven experiential learning. If, for example, Republican and conservative areas are more homogeneously white than Democratic areas, then it could be this partisan difference rather than the “whiteness” of a neighborhood that underlies the differences in the 18 Table 2.3 Neighborhood Effect of Different Groups, by Race/Ethnicity and Context (responses in percent) 19 Effect of Blacks Effect of Latinos Effect of Asians Effect of Whites Im- No Made Im- No Made Im- No Made Im- No Made proved Effect Worse proved Effect Worse proved Effect Worse proved Effect Worse Blacks in majority black precincts (n = 139) 31 60 8 21 67 11 16 79 5 20 75 5 Blacks in majority minority precincts (n = 57) 29 58 14 27 54 19 20 76 4 18 68 14 Blacks in majority white precincts (n = 21) 5 85 10 5 85 10 14 86 0 10 85 5 Latinos in majority Latino precincts (n = 147) 14 82 5 35 58 8 25 71 4 22 75 3 Latinos in majority minority precincts (n = 135) 12 79 9 30 63 8 18 73 10 16 78 6 Latinos in majority white precincts (n = 126) 13 83 4 19 71 10 16 82 3 19 75 6 Asians in majority minority precincts (n = 145) 15 76 9 14 77 9 37 61 2 27 70 3 Asians in majority white precincts (n = 125) 13 82 5 13 84 4 28 70 3 21 74 4 Whites in majority minority precincts (n = 236) 13 74 13 19 67 14 22 67 12 19 77 4 Whites in majority white precincts (n = 367) 9 86 5 11 78 12 15 79 6 16 81 3 Table 2.4 Neighborhood Effect of Immigrants, by Race/Ethnicity and Context (responses in percent) Blacks in majority black precincts (n = 139) Blacks in majority minority precincts (n = 57) Blacks in majority white precincts (n = 21) Latinos in majority Latino precincts (n = 147) Latinos in majority minority precincts (n = 135) Latinos in majority white precincts (n = 126) Asians in majority minority precincts (n = 145) Asians in majority white precincts (n = 125) Whites in majority minority precincts (n = 236) Whites in majority white precincts (n = 367) Effect of Legal Effect of Illegal Immigrants Immigrants Im- No Made Im- No Made proved Effect Worse proved Effect Worse 23 70 7 8 69 22 21 65 15 12 70 19 11 89 0 6 78 17 36 57 7 14 62 24 22 71 8 11 72 17 16 79 5 9 78 12 29 66 6 9 61 29 25 72 3 8 78 15 26 64 10 9 58 33 15 78 7 6 70 24 outlook of majority white and majority nonwhite areas. Similarly, it is unclear whether racial or ethnic minorities living in majority white areas have a lower opinion of their own ethnic group moving in because of some sort of self-selection process or because they are conforming to the dominant viewpoint among the white residents in the area. Another hint of learning deriving from the diversity of the neighborhood is that in almost all cases, those who lived in majority minority areas were less likely to be neutral about the presence of a given group in their neighborhood. Consider, for example, the white respondents. Living in a majority minority area increased the percentage of those with either a positive or negative view of blacks by about 12 20 points, of Latinos by about 11, and of Asians by about 12. The same pattern holds for black, Asian, and Latino respondents. We reported earlier that images of illegal immigrants tended to be more negative than responses elicited by questions about other groups. How does this vary by neighborhood context? Here, the more dramatic effects do suggest a pattern of negative learning. For all racial groups, including Latinos, people had more negative opinions about illegal immigrants if they lived in ethnically mixed areas than if they resided in largely white areas. Whether this opinion is founded on sensitivity to heightened economic competition or some other negative reaction, hostility to illegal immigrants seems stronger in precincts where they are more likely to live. The Influence of Subjective Contexts An alternative to defining context in terms of the current ethnic composition of a neighborhood is to focus on whether the context is changing. Advocates of the threat hypothesis (e.g., Green et al., 1999; Taylor, 1998) often adopt this approach. In the absence of trend data on the precincts, however, one can measure neighborhood context in subjective terms and compare neighborhoods that are viewed as becoming more diverse with those that are perceived as staying the same or becoming less diverse. A perception of rapid change, whatever the empirical reality, may engender a stronger sense of anxiety than would occur in areas with inter-racial stability. As actual demographic trends in California imply, about 36 percent of the respondents in our sample thought their area was becoming more diverse, compared to just 4 percent who perceived declining ethnic diversity. Whites and Asians (40 percent) were most likely to perceive their areas as changing in this way, and Latinos (28 percent) were the least likely to think so. Tables 2.5 and 2.6 show how beliefs about the changing composition of one’s neighborhood were linked to evaluations of the neighborhood effect of particular groups. A now familiar pattern emerges. The effect of perceived change in the character of the neighborhoods by the major racial and ethnic groups was relatively slight. By small margins, the images of Asians and Latinos were more negative 21 Table 2.5 Neighborhood Diversity, by Neighborhood Effect of Different Groups (responses in percent) Neighborhood less diverse (n = 56) Neighborhood about the same (n = 858) Effect of Blacks Improved/ Made No Effect Worse 90 10 Effect of Latinos Improved/ Made No Effect Worse 90 10 94 6 92 8 Effect of Asians Improved/ Made No Effect Worse 96 4 96 4 Effect of Whites Improved/ Made No Effect Worse 92 8 96 4 22 Neighborhood more diverse (n = 512) 90 10 85 15 91 9 95 5 Table 2.6 Neighborhood Diversity, by Neighborhood Effect of Immigrants (responses in percent) Neighborhood less diverse (n = 56) Neighborhood about the same (n = 858) Neighborhood more diverse (n = 512) Effect of Legal Immigrants Improved/ Made No Effect Worse 87 13 95 5 91 9 Effect of Illegal Immigrants Improved/ Made No Effect Worse 71 29 82 18 69 31 among respondents in the areas believed to be changing. But otherwise, perceived changes in the diversity of one’s neighborhood did not seem to alter perceptions of the groups a great deal. The pessimistic assumption that the increasing number of minorities in California is resulting in more hostile attitudes at the local level is not borne out here. Once again, however, respondents singled out illegal immigrants for criticism (Table 2.6). People in changing areas (those perceived as becoming either more or less diverse) were much more likely than those in stable neighborhoods to think that illegal immigrants make the quality of life in their neighborhood worse. Multilevel Analysis of Context Effects Our analysis so far has explored whether perceptions vary by neighborhood context, but not whether the differences observed reflect genuine contextual influences rather than compositional variation in the makeup of localities. To address this central issue, we need to estimate a statistical model that includes the following attributes of individual respondents as well as measures of their ethnic context as predictors: 1. Party and ideology, since the Republican party and conservatives generally tend to take a harder stance on immigration and affirmative action; 23 2. Sex, age and education, since these factors are often related to differences in racial and ethnic tolerance; 3. Perceptions of the state’s and the respondent’s personal financial situations, since those who are economically vulnerable may feel more threatened by members of other ethnic groups or by immigrants; and 4. Feelings of racial or ethnic identification, since those with a higher level of group consciousness to begin with may have more well-defined views about the presence of other racial and ethnic groups. The complete results of the analysis, including the statistical estimates, are presented in Appendix B. As a general conclusion, we found that the relatively minor neighborhood differences in beliefs about the effect of other groups reflected compositional rather than contextual effects. Controlling for the respondent’s background and political outlook, we found no statistically significant effect for the size of the minority group population in a neighborhood. This held true whether we measured context dichotomously as majority minority or majority white, continuously in terms of the proportion of black, Latino, Asian, or total minority population, or subjectively in terms of perceptions of ethnic change. The only consistent evidence of a contextual effect—and this, too, might be interpreted as a compositional factor—concerned the perceptions that people had of the neighborhood effect of their own groups. When asked about the effects of blacks on the neighborhood, blacks living in an area where they were the majority ethnic group tended to view the effect of more blacks in the neighborhood more favorably than their counterparts residing in mainly white areas. The same pattern held true among Latino and Asian respondents, suggesting that contextual influences may be specific to particular groups rather than generalized. It seems that favoritism toward one’s own group was stronger among those living in areas heavily populated by people with the same racial or ethnic background, either because of self-selected residential location or because the general tendency toward ethnocentrism (Tajfel, 1978) is reinforced by interacting with one’s own 24 group. Either way, where one lives seems to reveal something about a level of comfort with one’s own ethnic group that is not simply explained by partisan orientation, ideology, or social background. The equations summarized in Appendix B indicate that economic conditions affected perceptions of Latinos and Asians more than beliefs about blacks. This confirms earlier findings about the influence of economic adversity in stimulating opposition to immigration (Citrin et al., 1997), a factor that helps explain the passage of Proposition 187. In the present survey, respondents who perceived either the state as a whole or themselves to be in a worse economic situation than a year ago were more likely to see the presence of Asians and Latinos as having an unfavorable effect on their neighborhoods. Also, among Latino and Asian respondents, a strong sense of in-group identification was associated with a less-favorable attitude toward other ethnic groups. The models with perceptions of legal and illegal immigrants as dependent variables also contain some new information. Republicans and respondents who resided in neighborhoods perceived as becoming more diverse seemed to have the most negative view of the illegal immigrants. Whites in majority minority areas, controlling for party identification, also seemed to have more negative views—and this may actually be an instance of experiential learning. As for legal immigrants, other things being equal, respondents from all groups who lived in majority minority areas had more favorable views regarding their effect on the neighborhood. On the other hand, respondents who felt economically vulnerable had more negative views of legal as well as illegal immigrants, as did those with a stronger sense of racial identification. Summary of Findings Overall, California’s racial and ethnic groups had similar views about the effects of ethnic diversity on their neighborhoods. For all target groups, the dominant reaction was that their overall effect on one’s neighborhood was neutral. Beyond this, the ratio of perceived positive to negative effects was favorable, usually by a margin of at least two to one. The only exception was the negative image of illegal immigrants. In this case, the balance of opinion was reversed, and every ethnic group in the 25 sample felt that illegal immigrants were making conditions in their neighborhood worse rather than better. Perhaps predictably, white respondents were less positive about the effect of more ethnic diversity than were minority group members themselves, suggesting the existence of a discernible, though not deep, ethnic divide on this issue. More generally, even with multiple statistical controls, all ethnic groups expressed more positive views about their own group than any other. Last, on the central matter of contextual influences, there was no evidence of distinctively negative learning in ethnically diverse areas. The results of this initial test of the threat hypothesis are largely negative. 26 3. Racial Attitudes and Ethnic Context in California The expectation that sizable minority populations increase white hostility assumes that people experience the presence of a different ethnic group in their locality as a diffuse threat (Liska et al., 1985; Giles and Buckner, 1993; Quillian, 1996; Wright, 1977; Voss, 1996). This inference often is made without direct measures of perceived threat or other attitudes that may mediate between the individual’s environment and political behavior. The evidence presented in Chapter 2 indicated only minor contextual influences on perceptions of how ethnic change was affecting the quality of neighborhood life. This chapter explores the effects of context on normative beliefs about race relations and on the strength of ethnic identifications. These orientations may get at feelings of threat and anxiety, and thus may potentially link ethnic context and voting on Proposition 209, a relationship examined directly in Chapter 4. How do Californians evaluate the current state of ethnic group relations in society and what do they think of the policies frequently proposed for dealing with the issue? What are the ethnic differences in these views and how are they affected by residential context and feelings of group identification? For example, are people living in ethnically mixed neighborhoods more or less optimistic about achieving the ideal of a color-blind society? Are they more or less likely to agree that people are happier when living with their own group? How does context relate to the strength of people’s sense of racial and ethnic identity? We explore whether mixing with members of other racial and ethnic groups sensitizes one to the problem of discrimination or leads one to downplay its existence. Finally, we examine whether the most important divisions on these issues are between whites and nonwhites, between blacks and nonblacks, or between all the ethnic groups separately. 27 Once again, the order of presentation is to show contextual and individual-level differences separately, and then to report the results of multivariate models. To preview the results, our analyses indicate only minor effects of neighborhood context on racial attitudes, but also point to an indirect influence of context on beliefs about discrimination and social interactions, mediated through people’s feelings of identification with their particular ethnic group. Racial and Ethnic Identification One building block for beliefs about racial policies is feelings of identification with one’s own racial or ethnic group. A strong sense of group consciousness should foster support for measures favoring one’s own group. Accordingly, a pervasive sense of group consciousness should promote a common outlook on political issues visibly linked to the group’s status. But at the same time, a strong sense of identification with one’s own group often engenders hostility toward outsiders, and this poses a problem for an ethnically diverse society such as that of the United States (Brown, 1986). Indeed, an important element of the integrationist ideal is that people should not think of themselves primarily in racial or ethnic terms when they are acting politically. By this reasoning, a society is more likely to be race-neutral or color-blind in law and practice if people do not think that racial or ethnic groups are the primary vehicle for pursuing their interests. People can become aware of the political relevance of their racial or ethnic identity for many reasons: because the actions of others toward them are based on racial prejudice or bias; because other groups organize themselves on racial or ethnic lines; or because the state recognizes racial and ethnic categories in its programs and policies. Among the multiple stimuli in California’s political environment that could heighten or lessen racial and ethnic political identity, the focus here is the role of neighborhood context. Respondents in our survey were asked the following question about their personal political identity: “When it comes to social and political matters, some people think of themselves mainly as black, white, Latino, Asian, or Jewish, and that is very important to how they think of themselves. Other people don’t give much thought to these things. 28 When it comes to social and political matters, how important is your race and ethnicity to how you think of yourself?” Opinions about racial and ethnic identity divided whites and nonwhites (see Table 3.1). Two-thirds of white respondents said that their race and ethnicity were “not very important” or “not at all important” when considering social or political matters. By comparison, two-thirds of Asians and Latinos and 74 percent of blacks said that their racial and ethnic identity was “very important” or “somewhat important.” Of the minority groups, blacks expressed the most intense sense of ethnic group identification: Among them, but not among Latinos or Asians, a majority said that their racial identity was “very important.” These findings are consistent with the results of research based on national surveys, which also showed stronger and more widespread feelings of ethnic group identification among minorities than among whites (Wong, 1998). Previous studies indicate that the extent of group consciousness has political consequences (Tate, 1993; Dawson, 1994). Identifying strongly with one’s ethnic group increases the likelihood that one will mobilize and participate on behalf of that group. Identification also affects policy attitudes and vote choices. Looking at the effects of race and neighborhood context together (see Table 3.2), we find that for blacks, context had only a small effect on whether respondents emphasized a racial identity.1 But for Latinos and Table 3.1 Racial and Ethnic Identification, by Race/Ethnicity of Respondents (in percent) Whites (n = 580) Blacks (n = 209) Latinos (n = 386) Asians (n = 263) Very Important 15 56 42 30 Somewhat Important 20 18 24 35 Not Very Important 27 12 15 19 Not at All Important 39 14 19 16 ____________ 1Because few blacks in our sample live in majority white areas, it is difficult to make any generalizations with confidence based on this subgroup. 29 Table 3.2 Racial and Ethnic Identification, by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 135) Blacks in majority minority precincts (n = 54) Blacks in majority white precincts (n = 20) Latinos in majority Latino precincts (n = 141) Latinos in majority minority precincts (n = 125) Latinos in majority white precincts (n = 120) Asians in majority minority precincts (n = 142) Asians in majority white precincts (n = 121) Whites in majority minority precincts (n = 227) Whites in majority white precincts (n = 353) Very Somewhat Not Very Not at All Important Important Important Important 60 19 7 13 54 19 15 13 35 10 35 20 45 21 15 19 48 27 11 14 33 26 18 23 28 37 22 14 34 34 15 17 20 20 25 35 11 19 29 41 whites, particularly, living in a majority minority precinct led to a stronger sense of racial and ethnic identity than in other types of precincts. Specifically, 75 percent of Latinos who lived in majority minority precincts thought their race and ethnicity was “very” or “somewhat” important when it came to political and social matters, as opposed to 59 percent of Latinos in majority white precincts. For whites, a sense of racial identity was much weaker, but there was some variation across contexts: 40 percent of whites in majority minority precincts said their race was important when they decided political issues, whereas only 30 percent of whites in majority white precincts felt this way. In sum, living among racial minorities—regardless of whether one is white or nonwhite—does seem to promote a greater sense of ethnic identification among some groups. 30 Images of Race Relations Respondents were asked several questions about the current state of race relations in the United States. One item asked whether they agreed or disagreed with the following statement: “In our lifetime, America can become a color-blind society.” A majority of all ethnic groups felt this would not occur in the near future (see Table 3.3)2 : 52 percent of whites, 62 percent of blacks, 54 percent of Latinos, and 56 percent of Asians disagreed with the statement. The relatively greater pessimism of the black respondents is consistent with other data in our poll and with previous studies showing that perception of ongoing discrimination is more widespread among blacks (Uhlaner, 1991). Still, this survey item sets a very high standard for overcoming ethnic divisions in society; the fact that about 40 percent of the sample felt that a “color-blind” society could emerge quite soon points to the presence of a substantial degree of optimism about progress in race relations. Furthermore, opinion about the possibility of achieving a color-blind society was largely unaffected by neighborhood (see Table 3.4); only for Asians was there a statistically significant contextual effect. Forty-two percent of Asians living in majority minority precincts agreed that America could become a color-blind society in the near future, compared to 31 percent of those living in majority white precincts. One Table 3.3 “Color-Blind Society,” by Race/Ethnicity of Respondents (in percent) Whites (n = 593) Blacks (n = 210) Latinos (n = 399) Asians (n = 261) Disagree Strongly 27 47 36 33 Disagree Somewhat 25 16 18 23 Neither Agree Nor Disagree 5 6 6 7 Agree Somewhat 25 15 19 20 Agree Strongly 19 17 22 17 ____________ 2Implicit in the interpretation of this statement as a sign of optimism or pessimism is the idea that a color-blind world is an ideal; however, it is possible that “pessimism” may in fact reflect simple pragmatism. 31 Table 3.4 “Color-Blind Society,” by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 134) Blacks in majority minority precincts (n = 56) Blacks in majority white precincts (n = 20) Latinos in majority Latino precincts (n = 143) Latinos in majority minority precincts (n = 131) Latinos in majority white precincts (n = 125) Asians in majority minority precincts (n = 139) Asians in majority white precincts (n = 122) Whites in majority minority precincts (n = 231) Whites in majority white precincts (n = 361) Disagree Strongly 45 54 40 39 36 34 33 33 26 27 Disagree Somewhat 16 13 25 16 18 19 19 28 23 26 Neither Agree Nor Agree Agree Disagree Somewhat Strongly 6 16 18 5 13 16 5 15 15 4 16 25 6 20 20 6 21 20 7 24 18 8 16 15 4 26 20 5 24 18 interpretation of this is that assimilation into white society may not be so smooth for Asians in California as is often assumed. Another question about progress in race relations asked about equal opportunity. When given the statement, “Members of certain ethnic or racial groups still have less opportunities to get ahead than other people,” a majority of all racial groups, including whites, agreed on the presence of continuing discrimination (see Table 3.5). Ethnic minorities, and blacks in particular, were more likely to perceive a continued lack of equal opportunity “to get ahead.” These results are consistent with the previously noted opinion about the prospect of quickly achieving a colorblind society. There was no significant variation by neighborhood context for this item (see Table 3.6), and neighborhood experiences may 32 Table 3.5 “Minorities Have Less Opportunity,” by Race/Ethnicity of Respondents (in percent) Whites (n = 593) Blacks (n = 215) Latinos (n = 401) Asians (n = 263) Agree Strongly 34 66 51 48 Agree Somewhat 26 15 20 24 Neither Agree Nor Disagree 7 2 5 8 Disagree Somewhat 15 7 10 11 Disagree Strongly 19 10 15 9 Table 3.6 “Minorities Have Less Opportunity,” by Race/Ethnicity and Context (in percent) Agree Strongly Blacks in majority black precincts (n = 138) 67 Blacks in majority minority precincts (n = 56) 61 Blacks in majority white precincts (n = 21) 71 Latinos in majority Latino precincts (n = 145) 50 Latinos in majority minority precincts (n = 134) 53 Latinos in majority white precincts (n = 122) 52 Asians in majority minority precincts (n = 142) 47 Asians in majority white precincts (n = 121) 49 Whites in majority minority precincts (n = 231) 40 Whites in majority white precincts (n = 362) 30 Agree Somewhat 16 14 10 18 19 23 25 22 23 28 Neither Agree Nor Disagree 2 2 5 6 5 2 8 8 5 8 Disagree Somewhat 8 7 5 10 10 9 10 12 14 15 Disagree Strongly 7 16 10 17 12 15 10 8 19 19 33 therefore be less relevant to perceptions about opportunities than the work or school environment. It often is assumed that since Americans tend to endorse the principle of racial equality, perceptions of ongoing discrimination will lead to greater support for policies intended to benefit racial minorities. Still, our failure to detect contextual effects suggests that neighborhoodlevel interactions with members of other ethnic groups do not play a role in how one views access to opportunities for all Californians. Social Distance A third question about ethnic group relations was more concerned with social behavior than politics, asking respondents whether they agreed or disagreed that “people of different ethnic and racial groups are generally happier when they live and socialize with others of the same background.” One cannot tell if people treated this as a prescriptive or descriptive statement. Nevertheless, there was a general consensus on the benefit of living with people of the same ethnicity, a preference that might contribute to self-selection as a factor in promoting homogeneous neighborhoods. Table 3.7 shows that 66 percent of whites agreed, as did 57 percent of blacks, 61 percent of Latinos, and 70 percent of Asians. One possibility is that it is the most recent immigrants who most prefer living “with one’s own kind.” If that is the case here, then as their length of residence in California grows, the tendency of Hispanic and Asian residents to express this preference might diminish. (Nevertheless, since the respondents in the survey were registered voters, they have all lived in the United States at least five years.) The only evidence of a contextual effect that was statistically significant was for Latinos (see Table 3.8): 65 percent of Latinos in majority Latino areas felt that people were happier with others of their own background, compared to 57 percent of those living in majority white areas. Given the normative emphasis on the value of diversity in contemporary political culture, it is somewhat surprising that a majority of all racial groups think that an ethnically homogeneous residential environment is related to happiness. This is not the socially desirable response, and may imply a recognition of the psychological obstacles 34 Table 3.7 “People Happier with Others of Same Background,” by Race/Ethnicity of Respondents (in percent) Whites (n = 573) Blacks (n = 207) Latinos (n = 393) Asians (n = 263) Agree Strongly 35 32 36 43 Agree Somewhat 31 24 24 27 Neither Agree Nor Disagree 13 14 10 8 Disagree Somewhat 11 14 16 12 Disagree Strongly 9 16 13 10 Table 3.8 “People Happier with Others of Same Background,” by Race/Ethnicity and Context (in percent) Agree Strongly Blacks in majority black precincts (n = 132) 35 Blacks in majority minority precincts (n = 55) 27 Blacks in majority white precincts (n = 20) 30 Latinos in majority Latino precincts (n = 141) 42 Latinos in majority minority precincts (n = 131) 40 Latinos in majority white precincts (n = 121) 26 Asians in majority minority precincts (n = 140) 44 Asians in majority white precincts (n = 123) 42 Whites in majority minority precincts (n = 228) 35 Agree Somewhat 23 26 30 23 20 31 27 26 31 Neither Agree Nor Disagree 11 22 15 10 8 13 9 8 12 Disagree Somewhat 14 11 20 14 18 16 10 15 11 Disagree Strongly 17 15 5 11 15 14 10 9 11 35 facing the development of more intimate and friendly inter-group relations. Affirmative Action An obvious possibility is that the perception of racial threat would lead whites to oppose affirmative action, a policy explicitly designed to increase the number of minorities in desirable positions. One question in our survey asked respondents whether they regarded affirmative action programs as having a zero-sum quality. Whites (50 percent) were most likely to agree with the statement that “The more good jobs and places in college provided to minorities, the fewer there are for people who are not members of those groups.” A majority of blacks and Latinos disagreed, and Asians were evenly split. The differences among minority groups were relatively small, though, and opinions were not affected by the racial context in which respondents lived (see Tables 3.9 and 3.10). A second question concerning affirmative action is more pointed, in that it asks for a judgment about whether minorities are getting an unfair advantage. We asked respondents whether they agreed or disagreed with the following statement: “Members of particular ethnic or racial groups use special programs to get more benefits than they deserve.” Here, a majority of Latinos (51 percent) and Asians (55 percent), as well as whites (56 percent), agreed with this criticism of affirmative action (see Table 3.11). However, a majority (53 percent) of black respondents disagreed, a possible reflection of their stronger conviction that Table 3.9 “Minorities Gain at Expense of Others,” by Race/Ethnicity of Respondents (in percent) Whites (n = 576) Blacks (n = 204) Latinos (n = 390) Asians (n = 256) Disagree Strongly 18 33 27 24 Disagree Somewhat 19 20 23 18 Neither Agree Nor Disagree 13 10 11 15 Agree Somewhat 27 18 22 25 Agree Strongly 23 19 17 18 36 Table 3.10 “Minorities Gain at Expense of Others,” by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 131) Blacks in majority minority precincts (n = 53) Blacks in majority white precincts (n = 20) Latinos in majority Latino precincts (n = 137) Latinos in majority minority precincts (n = 131) Latinos in majority white precincts (n = 122) Asians in majority minority precincts (n = 139) Asians in majority white precincts (n = 117) Whites in majority minority precincts (n = 226) Whites in majority white precincts (n = 350) Disagree Strongly 32 30 45 30 26 24 25 22 22 15 Disagree Somewhat 17 23 30 18 24 28 22 15 18 20 Neither Agree Nor Agree Agree Disagree Somewhat Strongly 12 18 21 11 21 15 0 10 15 15 23 15 12 20 19 7 24 18 12 22 19 18 28 17 12 25 23 14 28 23 Table 3.11 “Minorities Get More Than They Deserve,” by Race/Ethnicity of Respondents (in percent) Whites (n = 579) Blacks (n = 210) Latinos (n = 391) Asians (n = 251) Disagree Strongly 17 32 28 19 Disagree Somewhat 15 21 14 14 Neither Agree Nor Disagree 12 8 7 12 Agree Somewhat 25 17 24 29 Agree Strongly 31 22 28 26 37 minorities still do not have an equal opportunity to get ahead. Again, there were no contextual effects affecting this pattern of results (see Table 3.12). To the extent that ethnic context captures variation in social contact with minority group members, then, white attitudes toward the state of race relations in this country are unaffected by such interchange. An important overall finding about California opinion is the coexistence among whites of a general recognition that prejudice and a lack of equal opportunity persist and a lack of strong support for affirmative action, a policy intended to alleviate the effects of these problems. Table 3.12 “Minorities Get More Than They Deserve,” by Race/Ethnicity and Context (in percent) Blacks in majority black precincts (n = 133) Blacks in majority minority precincts (n = 56) Blacks in majority white precincts (n = 21) Latinos in majority Latino precincts (n = 139) Latinos in majority minority precincts (n = 132) Latinos in majority white precincts (n = 120) Asians in majority minority precincts (n = 129) Asians in majority white precincts (n = 122) Whites in majority minority precincts (n = 229) Disagree Strongly 27 38 48 27 26 31 21 16 16 Disagree Somewhat 23 16 24 9 17 16 14 15 17 Neither Agree Nor Agree Agree Disagree Somewhat Strongly 10 16 24 5 21 20 5 10 14 10 25 29 4 21 33 8 25 21 12 23 30 12 35 22 12 21 35 38 Explaining Ethnic Identification and Attitudes Toward Race Relations Determining the meaning of the reported differences in outlook about race relations requires additional analysis. For example, past research has shown that blacks are much more likely to be Democrats than are whites; that the longer Latinos are in the United States, the more likely they are to be Democrats; and that the pattern of party affiliation varies across Asian subgroups (Cain et al., 1991). In other words, what appears to be ethnic group differences may simply reflect the influence of partisanship. There thus is a need to control for the compositional factors underlying group differences in outlook, while keeping in mind the possibility that a strong sense of ethnic identity may itself contribute to choosing a particular party affiliation. In our sample of Californians, 46 percent of the whites were Democrats, compared to 84 percent of the blacks, 68 percent of the Latinos, and 37 percent of the Asians. Party identification was significantly related to all the racial attitude items at the bivariate level, with Democrats more likely to perceive persistent discrimination and to regard affirmative action in positive terms. Ideology and educational attainment—to name only a couple of other influences—also were related to several of the racial attitude items. For this reason, it is necessary to control for these individual factors in a multivariate model to determine whether ethnic background or ethnic context has an independent effect on how people perceive race relations in the United States. As in Chapter 2, our statistical models therefore included the following variables as predictors (and controls): 1. Demographics: gender, age, and education; 2. Political predispositions: party identification and ideological self-identification; 3. Perceptions of diversity in the respondent’s neighborhood; 4. Economic outlooks: perception of whether the state economy has gotten better or worse over the last year and perception of whether the respondent’s own personal financial situation has gotten better or worse over the last year; 39 5. Ethnic context; and 6. Racial and ethnic identification, when this item is not itself the dependent variable of interest. The models are estimated in two different ways: in the first set of analyses, we employed the total sample, and included as predictors dummy variables for race and ethnicity and a continuous measure of the percentage of minorities living in a precinct as a measure of ethnic context (see Table 3.13). In the second set of equations, separate models were run for each race separately (see Appendix B).3 Results for Ethnic Identification Because group identification is an individual predisposition that possibly precedes a number of different political attitudes, the only variables included as predictors in the model for racial and ethnic identification are demographic factors and ethnic context. The main finding is that blacks, Latinos, and Asians were all more likely than whites to feel that their racial and ethnic identification was important, even when all other factors were held constant. Also, as we saw earlier, ethnic context turned out to be a significant predictor of group consciousness; the greater the percentage of minorities in one’s precinct, the more likely one was to express a strong sense of racial or ethnic identification. Living in areas with greater concentration of minorities did heighten one’s racial and ethnic political identity, regardless of whether one was white or a minority resident. Another notable finding is that the more educated respondents were less likely to think that their racial or ethnic identification was important when it came to social and political matters. When separate models were run for whites, blacks, Latinos, and Asians, the results indicate nuances in the underpinnings of the racial and ethnic identification of the separate groups. For whites and blacks, ethnic context was a significant predictor; the greater the percentage of minority residents in the area where they lived, the more likely they were to think their racial and ethnic identity was politically important. ____________ 3We also ran these models with other measures of ethnic context, as in Chapter 2, and obtained the same results. 40 Table 3.13 Multivariate Analyses of Racial Attitudes (entire sample, OLS regression) 41 % minority in precinct Black Hispanic Asian Gender Education Age Party identification Ideology Neighborhood diversity Neighborhood effect of blacks Neighborhood effect of Latinos Neighborhood effect of Asians Neighborhood effect of illegal immigrants Neighborhood effect of legal immigrants Neighborhood effect of whites Personal financial situation State economic situation Racial identifiication Constant Adjusted R–squared n Racial Identification B SE –0.47** 0.12 –0.89** 0.10 –0.66** 0.08 –0.70** 0.09 0.07 0.06 0.07* 0.03 0.04 0.04 2.80** 0.15 0.16 1,321 Minorities Color-Blind Have Less Society Opportunity B SE B SE 0.05 0.23 –0.47* 0.21 –0.58** 0.21 –0.47* 0.19 –0.38* 0.16 –0.41** 0.15 –0.25 0.18 –0.46** 0.16 –0.12 0.12 –0.02 0.11 –0.23** 0.05 –0.13** 0.05 0.04 0.08 0.01 0.07 0.20** 0.08 –0.01 0.07 –0.06 0.08 0.16* 0.07 –0.25* 0.11 –0.03 0.10 0.07 0.23 0.25 0.21 –0.35# 0.21 0.04 0.19 0.30 0.27 0.11 0.24 0.00 0.16 0.35* 0.14 0.28 0.26 –0.07 0.23 –0.14 0.33 –0.03 0.29 –0.11 0.08 –0.02 0.07 0.10 0.08 0.33** 0.07 0.03 0.05 0.11* 0.05 3.83** 0.64 1.33* 0.58 0.05 0.11 726 728 Happier with Others of Same Background B SE –0.08 0.21 0.38* 0.19 0.14 0.15 –0.02 0.16 0.18 0.11 0.07 0.05 0.00 0.07 –0.03 0.07 0.03 0.07 0.05 0.10 –0.27 0.21 –0.02 0.19 0.02 0.24 –0.25# 0.14 0.11 0.23 0.28 0.29 0.00 0.08 0.05 0.07 0.10# 0.05 1.75** 0.58 0.00 709 Minorities Minorities Get Gain at Ex- More Than pense of Others They Deserve B SE B SE –0.41# 0.22 –0.10 0.22 –0.32 0.20 –0.53** 0.20 –0.51** 0.15 –0.38* 0.16 –0.11 0.17 0.09 0.18 –0.16 0.11 –0.10 0.12 –0.17** 0.05 –0.26** 0.05 –0.06 0.07 –0.10 0.08 0.08 0.07 0.02 0.08 0.13# 0.08 0.12 0.08 –0.06 0.10 0.18# 0.11 –0.06 0.22 0.25 0.22 0.20 0.20 0.36# 0.21 0.18 0.25 0.28 0.26 0.35* 0.15 0.42** 0.16 0.23 0.24 –0.11 0.24 0.14 0.30 0.08 0.31 0.17* 0.08 0.02 0.08 0.04 0.07 0.03 0.07 –0.06 0.05 0.04 0.05 2.45** 0.61 2.10** 0.63 0.07 0.08 710 719 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Among Latinos, the effect of racial context and education approached conventional levels of statistical significance: The more educated respondents and those living in precincts with fewer minorities were less likely to emphasize their ethnic identification than their counterparts with less education or living in less predominantly white neighborhoods. Results for Beliefs About Race Relations Race, but not ethnic context, was a significant predictor in the multivariate models of attitudes about the general state of race relations, even after multiple controls. When asked about the possibility of a colorblind country in the near future, blacks and the well-educated were more likely to be dubious than were white respondents or those with less education, respectively. Democrats were less likely to envisage the creation of a color-blind world than Republicans, and respondents who thought their personal finances had worsened over the last year were more pessimistic than those with improving circumstances. Finally, respondents who thought that their neighborhoods were becoming more ethnically diverse were less likely to agree that the United States will be a color-blind society in the near future than those who saw their neighborhoods as unchanging or becoming less diverse. Objective ethnic context was not a significant factor, but the perception that one’s context was changing was associated with less optimistic views about racial progress. When separate models were run for each of the four racial groups, the results were varied. Among whites, as in the sample as a whole, Republicans were more likely to say that America could become a colorblind society. Among blacks, the better-educated respondents were more pessimistic about this prospect, a possible sign of relative deprivation and frustration among the upwardly mobile in this ethnic group. Among Latinos, the more-educated, the conservatives, and those who saw their neighborhoods as more diverse were more pessimistic. Among Asians, the older were more optimistic than the young, and the better-educated were more skeptical than those with less schooling. Similar results were obtained for the item concerning equal opportunity for minorities. The better-educated respondents, blacks, Latinos, and Asians all were more likely to believe that minorities had 42 fewer opportunities to get ahead than were the less-educated and white respondents. Group identification was a significant factor here: The greater the political importance of one’s racial or ethnic identity, the greater the agreement that minorities had fewer opportunities than whites. Conservatives and people who thought the state economy had gotten worse over the last year tended to disagree with the idea that minorities had limited opportunities. Among white, Latino, and Asian respondents, those who thought the state economy had gotten worse over the last year were more likely to believe that equal opportunity for minorities already existed. And Latinos and Asians lacking a strong sense of group identity also were more likely to feel this way. One’s sense of racial or ethnic identity was a significant predictor of attitudes about living in an ethnically diverse context. The stronger the sense of ethnic group consciousness—in the sample as a whole and among whites, blacks, and Latinos—the more likely a respondent would agree that people were happier when living and socializing with others of the same background. Interestingly, even controlling for racial and ethnic identification, blacks remained more likely than whites to disagree that people were happier when living apart. So, even though blacks were the most likely to express strong feelings of ethnic consciousness, they were also more in favor of social and residential mixing than any other ethnic group. Results for Beliefs About Affirmative Action Turning now to the underpinnings of attitudes about affirmative action, the multivariate analysis revealed the persistence of some social group differences after the imposition of controls. Specifically, black, Latino, and well-educated respondents were less likely to view affirmative action as a zero-sum policy in which benefits were taken from whites and allocated to minorities. Conversely, Republicans and self-identified conservatives were more likely to regard affirmative action as a matter of some groups winning and others losing. Among white respondents, the less-educated, Republicans, and conservatives again were more likely to think affirmative action meant that minorities gained at the expense of others. Among blacks, however, 43 conservatives and those who saw their neighborhoods as becoming more diverse were more likely to disagree that jobs and school placements for minorities took away opportunities for whites. Among Latino respondents, the better-educated and those lacking a strong sense of racial and ethnic identification were also inclined to disagree with the zero-sum portrayal of affirmative action. Among Asians, women, liberals, and younger respondents were less likely to think minorities gained at the expense of others. Finally, when asked whether minorities deserved the benefits of government programs, again, the better-educated, liberal, black, and Latino respondents were less likely to think that minorities got more than they deserved. On the other hand, respondents who perceived their neighborhoods as becoming more diverse were more likely to believe that minorities got more benefits than they should. Looking at the results for just the white respondents reveals the familiar finding that liberals and the well-educated were less likely to view affirmative action as an unfair reward for the undeserving. Education had a similar effect on the opinions of blacks and Asians. Among Latinos, respondents who felt that the state economy had gotten worse over the last year were also more likely to disagree that minorities got more than they deserved. Summary of Findings The more general racial attitudes discussed in this chapter were all significantly related to voting intentions on Proposition 209 and are potentially related to other more concrete policy questions that arise from election to election. Determining the underpinnings of these beliefs therefore helps identify the sources of more directly political conduct. The principal findings of this chapter are as follows. There are significant ethnic differences in the salience and strength of racial or ethnic identification. Minority group members consistently expressed a stronger sense of ethnic identity than whites. In addition, ethnic context made a difference for whites, blacks, and Latinos, with respondents living among sizable minority populations expressing a stronger sense of racial or ethnic identification. 44 A majority of respondents doubted that America could soon become a color-blind society, although this opinion was most widespread among blacks and least pervasive among whites. A majority of all ethnic groups conceded that members of certain racial and ethnic groups still had fewer opportunities to get ahead, but white and minority respondents were much more divided on whether minority gains from affirmative action were unfair and came at the expense of whites. Most people in the survey seemed to think that people were happier when living among their own ethnic groups, and those who had a stronger sense of racial or ethnic identification felt this way about selfsegregation most strongly. Taken together with earlier evidence that respondents were largely neutral about the effect of other ethnic groups moving into their neighborhood, this suggests that Californians tolerate ethnic diversity without feeling strong enthusiasm for it. In a multivariate analysis controlling for party affiliation, ideology, and other factors, significant differences between whites and nonwhites remained, with a smaller gap between whites and Asians than between whites and Latinos or blacks. In addition to race, education was a strong predictor of racial attitudes. Education diminished the strength of one’s racial or ethnic identification, but, confirming the results of previous studies (Schuman et al., 1997), more educated respondents were more pessimistic about the current state of race relations in the country and more positive about how affirmative action policies work. Finally, the analysis revealed few contextual influences on the racial beliefs examined. Aside from the effect of ethnic context on feelings of racial or ethnic identity, the size of the minority population in one’s neighborhood did not have an independent effect on people’s more general racial attitudes. 45 4. Ethnic Conflict, Ethnic Context, and Voting on Proposition 209 Proposition 209 targeted the ethnic and gender preferences embedded in California’s affirmative action programs for public employment, college admissions, and government contracts. As such, this ballot initiative plausibly can be regarded as attractive to people who feel a sense of racial threat. Analyzing the vote on Proposition 209 thus provides an important case for determining the political implications of California’s ethnic diversity. The passage of Proposition 209 in November 1996 was one of several occasions on which California voters have employed direct democracy to alter policies providing tangible or symbolic benefits to ethnic minorities. As long ago as 1964, they approved an initiative, later ruled unconstitutional, overturning the Rumford Fair Housing Act. In 1986, voters passed an English-only amendment to the state constitution. In 1994, they overwhelmingly adopted Proposition 187, a measure designed to restrict the access of illegal immigrants to most state services. And 1998 brought the victory of Proposition 227, an initiative requiring the elimination of most established bilingual education programs in public schools. Ethnic differences in voting for these initiatives were marked, although varying in magnitude. The Los Angeles Times exit polls (see Table 4.1) showed that 62 percent of whites, compared to 27 percent of blacks, 32 percent of Latinos, and 45 percent of Asian voters, supported the California Civil Rights Initiative (Proposition 209). In the case of Propositions 63 (English-only), 187 (illegal immigration), and 227 (bilingual education), however, the exit polls showed that majorities of white and Asian voters were favorable, whereas the yes vote among 47 Table 4.1 Percentage Voting “Yes” on “Ethnic” Initiatives, by Race/Ethnicity Voting Group Proposition 63a Proposition 187b Proposition 209c English Illegal Affirmative Only Immigration Action Total 68 59 55 Whites 72 83 62 Blacks 67 55 27 Latinos 39 31 30 Asians 58 55 45 aSOURCE: California Exit Poll, November 1986. bSOURCE: California Exit Poll, November 1994. cSOURCE: California Exit Poll, November 1996. dSOURCE: California Exit Poll, November 1998. Proposition 227d Bilingual Education 61 67 48 37 57 Latinos always was less than 40 percent. A majority of black voters supported Propositions 63 and 187, but not 209 or 227. In sum, there is a consistent tendency of whites to be most favorable toward initiatives that eliminate benefits targeted toward ethnic minorities, but the pattern of voting among blacks, Latinos, and Asians varies from issue to issue. Racial Threat as an Explanation One interpretation of ethnic differences in voting emphasizes the role of group competition for jobs, power, and status. As spelled out in Chapter 1, this theory holds that the dominant ethnic group in a community experiences the influx of minorities as a visible threat to its established position and responds with defensive actions designed to shore up its economic and cultural advantages (Quillian, 1995; Tolbert and Hero, 1996). The relative size of minority groups in one’s community is assumed to measure the intensity of group competition and, therefore, the extent of the perceived collective threat to the interests of whites, the numerical majority and culturally dominant ethnic group. The presumed causal chain, then, is from the objective ethnic mix of one’s environment to the subjective experience of threat and anxiety to 48 political polarization. The alternative contact theory proposes a different path—from residential diversity to racial tolerance to compromise and consensus. Our survey was conducted just a week before the vote on Proposition 209. The stated voting intentions anticipated the pattern of ethnic differences in voting that ultimately emerged (see Table 4.2). A majority of whites said they would vote for the measure and a majority of minority respondents were opposed. At first glance, however, the racial threat prediction that whites in neighborhoods with large minority populations would be more likely to vote for Proposition 209 was not confirmed. Of the whites living in districts where they were the majority, 61 percent said they would vote for Proposition 209, compared to 56 percent of those living in the ethnically heterogeneous majority Table 4.2 Objective Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity Voting Group Percentagea Whites in majority minority precincts 56 Whites in majority white precincts 61 Total whites 59 Blacks in majority black precincts 26 Blacks in majority minority precincts 14 Total blacks 22 Latinos in majority Latino precincts 35 Latinos in majority minority precincts 28 Latinos in majority white precincts 36 Total Latinos 33 Asians in majority minority precincts 34 Asians in majority white precincts 40 Total Asians 37 a“Don’t know” responses are excluded from this analysis. 49 minority districts.1 One interpretation of these findings is that whites, as a group, did perceive affirmative action as a threat to their collective interests, but that this belief did not vary across residential environments. An alternative view is that whites living in majority minority areas have adapted to their environment and become more accepting of liberal racial policies. Latinos and Asians in majority minority precincts, although not blacks, were more likely to oppose Proposition 209 than those in majority white precincts. This finding, as others reported above, suggests that the numerical dominance of one’s own ethnic group in a neighborhood may facilitate the development of a sense of group consciousness and identity. Living as a small minority within the mainstream, by contrast, can foster more conformity to the dominant group’s norms. When we consider subjective reactions to ethnic diversity more directly, the contextual differences among white voters predicted by the threat hypothesis do emerge. Those perceiving their neighborhood as becoming more diverse were 10 percent more likely to favor Proposition 209 than those perceiving no change. Similar differences appear when one compares whites with negative and favorable or neutral opinions about the effect of minorities or immigrants on the quality of life in their neighborhoods (see Table 4.3). This held true for whites in both majority white and majority minority precincts, indicating a potential rise in political tensions in the aftermath of rapid ethnic change (Green et al., 1999). The two sides of the campaign over Proposition 209 disagreed about whether affirmative action programs still were needed in California and ____________ 1When queried in the survey, 15 percent of the respondents said they were still undecided about their vote. For clarity of interpretation, these respondents are excluded from the models reported here. One possibility is that these respondents were not genuinely undecided but rather pro-209 voters unwilling to express an anti-minority opinion to interviewers. Examining the social background and political views of these “don’t know” respondents, however, showed that in terms of age, education, party affiliation, ideology, and general racial attitudes, the aggregate position of this group fell in between respondents with definite pro and anti voting intentions, respectively. In addition, we repeated the analyses with these respondents included, thereby creating a three-point dependent variable, using a multinomial logit model. There were no significant differences in the results and no change in the nature of the observed contextual effects. 50 Table 4.3 Perceived Context and Percentage Voting “Yes” on Proposition 209, by Race/Ethnicity Question Do you think your neighborhood is becoming more ethnically diverse, less ethnically diverse, or is about the same? Whitesa Less More Diverse or Diverse Same Blacks Less More Diverse Diverse or Same Latinos Less More Diverse Diverse or Same Asians Less More Diverse Diverse or Same 64b 54 14 27 34 30 30 32 51 Please tell me whether (blacks/Latinos/Asians/ whites) have improved, made worse, or had no effect on the quality of life in your neighbohood. Made Worse Improved or No Effect Improved Improved Made or No Made or No Worse Effect Worse Effect Made Worse Improved or No Effect 65 56 26 22 30 32 39 31 Please tell me whether (illegal immigrants/legal immigrants) have improved, made worse, or had no effect on the quality of life in your neighborhood. 70 54 24 22 42 29 33 31 aFor whites, n = 516; for blacks, n = 180; for Latinos, n = 347; and for Asians, n = 229. b64 percent refers to the percentage of whites who believe their neighborhood is becoming more ethnically diverse and who voted for Proposition 209. whether such programs resulted in qualified applicants for jobs being passed over in favor of undeserving minorities. Our survey posed these questions to respondents. Predictably, supporters of the initiative were less likely to say that affirmative action programs were still needed and more likely to report knowing someone who had personally suffered as a result of group preferences. Here, too, differences in racial and ethnic background, but not in one’s ethnic context, affected these beliefs about affirmative action (see Table 4.4). Whites in majority white neighborhoods were a little less inclined to say that blacks still needed Table 4.4 Percentage Answering “Yes” to Affirmative Action Questions, by Race/Ethnicity and Ethnic Context Group Whites in majority minority precincts Whites in majority white precincts Total whites Blacks in majority black precincts Blacks in majority minority precincts Total blacks Latinos in majority Latino precincts Latinos in majority minority precincts Latinos in majority white precincts Total Latinos Asians in majority minority precincts Asians in majority white precincts Total Asians Is Affirmative Action Still Necessary For For For For Blacks? Latinos? Asians? Women? 58 54 40 59 48 48 35 51 52 51 37 54 92 90 71 89 86 83 70 84 89 87 70 87 69 74 54 76 70 75 63 80 73 75 61 80 70 74 59 79 70 76 67 76 57 62 46 56 64 69 57 67 Has Affirmative Action Helped Someone Undeserving? 31 37 35 15 14 14 15 16 25 18 28 24 26 52 affirmative action and were more likely to say that they knew of a case in which affirmative action had helped someone undeserving. What is more striking about these data, however, is the fact that a majority of all four ethnic groups believed that affirmative action was still needed to help blacks, Latinos, and women. In all four groups, too, just a minority reported knowing someone who suffered from reverse discrimination. If anything, this pattern of opinion points to a sense of empathy rather than threat. It also confirms that voting for Proposition 209 was strongly influenced by the proponents’ success in framing the initiative as a vote against rigid group preferences rather than against all forms of assisting minorities to overcome discrimination and disadvantage. Accounting for the Vote on Proposition 209 The findings presented above failed to show strong or consistent differences in support for Proposition 209 between people of the same ethnicity living in different racial environments. Beliefs about the consequences of increased ethnic diversity mattered more than where one lived. With these uneven results as background, the next step is to provide a more complex account that incorporates the influences of both individual-level and neighborhood-level factors. Such an analysis helps to identify the social and political sources of voting on Proposition 209 and to estimate the residual effects of neighborhood context on support for or opposition to affirmative action. The statistical models discussed below incorporate the same set of predictors used in Chapter 3, with voting intentions on Proposition 209 as the dependent variable. The results address these main issues: 1. Whether the respondent’s ethnicity and strength of racial or ethnic identification had similar effects on voting on Proposition 209 in all types of neighborhood contexts; 2. Which other individual-level characteristics predicted voting on Proposition 209 and whether these same attributes mattered in every ethnic group; and 3. Whether contextual effects in voting for Proposition 209 emerge when one adjusts for the influences of the respondent’s 53 individual characteristics, and hence for compositional differences among neighborhoods. In describing the underpinnings of the vote on Proposition 209, we again report findings first for the sample as a whole and then for each ethnic group separately. With regard to the basic test for the racial threat hypothesis, the multivariate model confirmed the negative results presented above. Ethnic context had no direct effect on how Californians voted on affirmative action. The significant influences we detected were individual-level characteristics: ethnicity, general political orientations, and subjective perceptions of the effect of ethnic change. Of course, there could be relationships between objective neighborhood conditions and these attributes of residents; as mentioned above, minority residents of predominantly white neighborhoods were more likely to have a relatively high socioeconomic profile. Table 4.5 provides an overview of the model estimating the vote on Proposition 209 for the sample as a whole. This table indicates whether the respondent’s ethnicity, racial or ethnic identification, and residential context had statistically significant influences on voting intention. The table also lists the other variables included in the model with significant effects. The more complete summary of the equations for the sample as a whole and for each ethnic subgroup in Table 4.6 gives the numerical value of the logistic regression coefficients for each predictor. Looking first at the results for the entire sample (first column in Table 4.5), we find strong relationships between ethnic background and voting intentions. Even after adjusting for other background characteristics, political orientation, and neighborhood context, minority Table 4.5 Effects of Race/Ethnicity, Racial Context, and Racial and Ethnic Identification on Proposition 209 Vote Race/Ethnicity, Racial Context, and Racial and Ethnic Identification Race/ethnicity: significant Racial context: insignificant Racial and ethnic identification: significant Other Significant Coefficients Age Party identification Ideological self-identification State economic situation 54 Table 4.6 Explaining Voting Intentions on Proposition 209 (logit analyses) 55 % minority in precinct Black Hispanic Asian Gender Education Age Party identification Ideology Neighborhood diversity Neighborhood effect of blacks Neighborhood effect of Latinos Neighborhood effect of Asians Neighborhood effect of illegal immigrants Neighborhood effect of legal immigrants Neighborhood effect of whites Personal financial situation State economic situation Racial identifiication Constant Total Sample B SE 0.01 0.36 –1.15** 0.34 –0.98** 0.26 –1.31** 0.28 –0.21 0.19 –0.14 0.09 0.36** 0.12 0.47** 0.12 0.44** 0.13 –0.13 0.18 –0.07 0.36 0.37 0.34 –0.22 0.42 0.28 0.25 –0.26 0.39 –0.17 0.55 0.06 0.13 0.36** 0.12 0.19* 0.09 –2.40* 1.02 Adjusted R–squared n 0.17 640 NOTES: B is logit coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Whites B SE 0.49 0.62 –0.57# –0.01 0.70** 0.95** 0.37 0.02 0.18 1.00 –0.45 –0.41 –0.95 1.10 –0.15 1.10** 0.06 –5.52** 0.26 241 0.35 0.16 0.24 0.21 0.24 0.34 0.64 0.63 0.69 0.48 0.73 1.30 0.25 0.26 0.16 2.07 Blacks B SE 1.74 1.59 0.08 –0.62* 0.16 0.50 0.43 –0.72 –0.02 –0.90 –0.03 1.24 –0.58 0.58 0.29 0.35 0.61 0.41 0.60 1.13 1.03 1.51 0.89 1.26 –0.18 0.39 0.53# –2.35 0.19 91 0.45 0.38 0.27 3.13 Hispanics B SE –0.79 0.67 –0.18 –0.18 –0.11 0.21 0.20 –0.10 –1.13 0.43 0.00 0.78# 0.25 –0.44 0.14 0.08 0.07 –0.51 0.07 195 0.34 0.17 0.23 0.22 0.22 0.32 0.88 0.64 0.77 0.46 0.70 0.77 0.22 0.21 0.15 1.79 Asians B SE 0.89 1.00 –0.19 0.09 0.63# 0.03 0.51 0.03 1.43 0.82 0.53 0.26 0.35 0.35 0.37 0.52 1.10 1.14 –0.44 0.69 0.44 1.29 0.33 –0.05 0.46# –7.91** 0.13 104 0.39 0.33 0.26 3.09 respondents were less likely than whites to favor the anti-affirmative action initiative. As noted, ethnic context did not alter these results. Neither the proportion of minority residents of a neighborhood nor the perception of increasing ethnic diversity had an independent influence on voting intentions. Table 4.6 reveals that feelings of racial identity did affect voting decisions. Overall, respondents who said that their racial and ethnic background was important to them when they decided political and social questions were more likely to oppose Proposition 209. Because strong feelings of ethnic identity were much more prevalent among minority group voters than whites, this result suggests that the mobilization of feelings of ethnic consciousness may have made the connection between affirmative action policies and their group interests more visible and psychologically salient. If this is true, then there may have been an indirect influence of ethnic context upon voting on Proposition 209, captured by the effect of residential context on the strength of group consciousness. Table 4.6 reveals the strong effects of party affiliation and ideology on voting on Proposition 209. Even with multiple controls for personal background and local context, Republicans and self-defined conservatives were more likely to vote against affirmative action. So, too, were older voters and those who felt that the state economy had deteriorated in the last year, a finding that is consistent with the idea that feelings of economic vulnerability often heighten ethnic tension and competition. With differences in general political outlook controlled for, however, women respondents in our sample were as likely as men to say they would vote for Proposition 209. Despite the efforts of the opposition campaign to frame the initiative as dangerous to women’s interests, this argument may not have had a special resonance among female voters as a whole. This is not to say that gender consciousness did not play a similar role as feelings of racial or ethnic identity. The gender gap in party affiliation and ideology—with women more likely than men to be Democrats and liberals—suggests that the role of group consciousness among women is already captured by these control variables. The results for the entire sample blur some interesting differences between whites and minority voters. The pattern of effects among whites 56 generally approximated those reported for the overall sample: age, a Republican party affiliation, and the feeling that the state economy had deteriorated boosted support for Proposition 209 among white respondents. In general, however, party affiliation and ideology had stronger influences on the choices of whites than on those of minority group members. Among black respondents, a higher level of formal education and strong feelings of racial identification significantly increased opposition to Proposition 209. Ethnic context had no effect among this group. Among the Latino respondents in the sample, there were hints of an influence of ethnic context, as measured by the proportion of minority residents in one’s neighborhood; the coefficient for this predictor approached conventional levels of statistical significance. No other predictor in the model had a statistically significant effect for the Latino respondents. Finally, among Asians, only age and pessimistic views of the state’s economy had statistically significant effects on how they intended to vote on Proposition 209. Once again, residential context did not seem to matter. Reviewing the Role of Context To summarize, this chapter’s investigation of voting on Proposition 209 found that ethnic group interest and broad political identifications were the main determinants of how Californians regarded affirmative action. The ethnic composition of one’s neighborhood did not have a direct effect on how people voted, disconfirming the racial threat hypothesis that anti-minority reactions among whites are related to the size of the local minority group population. This latter result deviates from a recent national study that found a significant relationship between the size of the black population in an area and racial hostility among the local white residents (Taylor, 1998). One possible explanation for these divergent findings is that the links in the causal chain from neighborhood composition to perception of threat to group identification to political conduct are more complex than generally assumed by the racial threat theory. The concentration of minority populations is not a necessary condition for perceived threat. What may matter more is change in the ethnic composition of a 57 neighborhood, the nature of inter-group contacts, and the economic status, cultural distinctiveness, or political organization of minority group members. One might argue that a more differentiated treatment of context might have revealed effects undetected by the present study’s emphasis on the proportion of minority group residents. Limitations in the available data restricted our ability to pursue some interesting alternatives; for example, we lacked census information about change in the composition of the neighborhood units of analysis. Nevertheless, analyses using the proportion of black and proportion of Latino residents in a neighborhood—rather than the proportion of minority residents as a whole—as the measure of ethnic context did not change the results reported above. Another possibility is that economic, rather than ethnic, context mattered, with the reactions of white voters on racial issues conditioned by local economic circumstances. In addition, if neighborhoods with large minority populations also tended to be poor—as measured by the aggregate level of unemployment, the extent of home-ownership, and median income, for example—any observed effect of ethnic context might be a spurious reflection of local economic conditions. To address these possibilities, we also estimated models predicting voting intentions on Proposition 209 that incorporated the proportion of the households in a neighborhood with annual incomes under $10,000, between $10,000 and $20,000, and so on, as a measure of economic context. Including economic context as a control did not alter the estimated coefficient for ethnic context; nor did economic context have an independent effect on stated voting intentions. It also could be argued that using the precinct or neighborhood as the unit for measuring context may not capture the pattern of group interactions that actually foster a sense of threat. It may be that the metropolitan area or county is the focal point for political or economic competition between ethnic groups. In the present study, we were able to merge county-level contextual data with information about individual respondents. Analyses showed that the size of the minority population at the county level similarly had no effect on the racial attitudes or beliefs about Proposition 209 among white respondents. 58 The divergent results about contextual effects also may be related to differences in the attitudinal or behavioral responses to minority populations being studied. For example, Taylor (1998, pp. 522–523) found strong effects of the proportion of blacks in an area on the level of traditional anti-black prejudice and stereotyping of blacks. But her study found no statistically significant effect of ethnic context on hostility to race-targeting policies designed to achieve equal outcomes—a measure that is close in meaning to support for Proposition 209. More generally, beliefs about the relative costs and benefits of ethnic diversity are founded on vicarious learning as well as on actual proximity to members of other groups. Competition for admission into the University of California occurs statewide, so the proportion of minority group members in one’s own neighborhood or county is unlikely to be the only source of opinions about how whites are likely to be affected by affirmative action programs in higher education. In addition, the campaign over Proposition 209 centered on how to define affirmative action in the public’s mind, with references to group interest, party, ideology, and principles of fairness as the main cues to which voters in all contexts were exposed. Under these circumstances, generalized political orientations rather than localized concerns might be expected to influence voting behavior. The oft-assumed tendency of whites to react to ethnic change around them with increased opposition to policies, parties, or candidates perceived as pro-minority is not automatic. This may vary with the visible effect locally of large minority populations, as with the effect of immigration on the makeup of public schools, or with how issues are defined and justified.2 One result worth pursuing in further research is the influence of ethnic context, defined in both objective and subjective terms, on feelings of group consciousness among both whites and minorities. ____________ 2Several studies (e.g., Tolbert and Hero, 1998) suggest that the proportion of Latinos and immigrants in a county influenced the vote on Proposition 187. The results are problematic, since those analyses did not control for compositional effects. The present study asked respondents whether they approved of Proposition 187. However, this retrospective answer coming two years after the actual election is an unreliable indicator of support, so our failure to detect contextual effects on responses to this question cannot be regarded as definitive. 59 Identity politics—the idea that politics should be motivated by the drive to enhance the standing of one’s own ethnic group—may be more widespread in locales with a substantial number of minority group numbers. One reason for this increased likelihood is that greater numbers engender denser networks of intra-group communications and facilitate the mobilization of resources for political participation. If this is so, then the influence of context on racial attitudes and voting may be quite potent, but indirect. We found that broad political orientations such as party affiliation and ideological outlook were stronger predictors of voting on Proposition 209 among whites than among minority groups. More generally, our models did poorly when it came to accounting for the choice of nonwhite respondents. To improve our understanding of cleavages within minority groups, subsequent analyses should attempt to incorporate better measures of the social status, residential mobility, and immigrant generation among respondents from these groups. 60 5. Conclusion In many respects, the relations among ethnic groups are more a product of civil society than public policy. Group attitudes and stereotypes are formulated largely by daily experience and the cultural transmission of images—processes that are substantially beyond the realm of governmental action and leadership. In fact, there are inherent limitations on the capacity of democratic governments to modify ethnic group relations. In the United States, for instance, the First Amendment freedoms of speech and association hamper the ability of public officials to mold social attitudes in desired directions. Whether the issue is teenage violence, civic malaise, or hate speech, Americans seem to fear the potential remedies of thought control and censorship more than the problem per se. Yet, there are important reasons why public policies that improve the quality of inter-group relations in California are needed: The ethnic diversity of the state’s population is a given, and it seems axiomatic that hostility and conflict among the various ethnic groups can have debilitating consequences for society. Even when racial and ethnic tensions do not lead to overt unrest and violence, they can contribute to economic inefficiency and undermine the legitimacy of government. Prejudice may result in a failure to hire the best people or negotiate the best contracts because people refuse to work or deal with others from different racial and ethnic groups. Deep ethnic cleavages also can make bad politics. Decisions by a government that is controlled by one racial or ethnic group will have less moral authority if the members of the outgroups feel that they have not had a fair chance to affect political outcomes. Second, even if the pattern of inter-group relations is primarily formed in civil society, governmental actions can accelerate or moderate trends that emerge in the private sector. Even when a government does 61 not legally discriminate against particular groups, it can favor some groups over others in more subtle ways. Holding power gives some groups access to jobs and economic opportunities that are denied to others. The needs of some communities may receive more attention than the needs of others. The effects of these differences can exacerbate or lessen racial tensions. Also, elected officials can influence attitudes about race and ethnicity by legitimating some positions and attacking others. Their campaigns help frame public debate by choosing either to emphasize or downplay ethnic issues. These strategic decisions influence the salience of group interests and identifications, potentially evoking strong emotional responses from voters. The struggle over affirmative action is illustrative. As a policy, Proposition 209 altered the conditions under which people of different genders, races, and ethnicities competed for jobs and educational opportunities. But the fact that the debate over Proposition 209 was framed within the context of partisan presidential politics affected the perceptions that different groups had of the measure. This study examined public reactions to ethnic diversity in California in the context of political contention about the future of affirmative action in the state. One important purpose was to understand whether the prevailing pattern of group attitudes made harmonious relations among the state’s different racial and ethnic communities more difficult to achieve. A second goal was to examine whether the ethnic composition of residential areas affected inter-group relations. In particular, we tested the widely held theory that white prejudice and hostility are related to the size of minority group populations in an area. There are four major conclusions. The first concerns the pattern of conflict and consensus across ethnic groups. There were substantial ethnic group differences on a number of racial attitudes and policies. Specifically, whites were less likely than members of the three ethnic minority groups to say that their race or ethnicity was important for their political identity. Whites were more likely than voters in the other ethnic groups to support Propositions 209 and 187 and were more likely to regard affirmative action programs as unnecessary and unfair. These ethnic group differences cannot be simply explained by other factors such 62 as party identification, ideology, social background, or residential context (see Table 4.6). At the same time, there was consensus among the four ethnic groups on a number of questions related to the state’s ethnic composition. An important point is that a large majority in every group was either neutral or positive about the effect of people of a different ethnic or cultural group moving into their neighborhood. They also were united in expressing more negative views about illegal immigrants than about any other group. Despite their differences about the need for affirmative action, respondents from all four ethnic groups also tended to agree on the egalitarian principles of merit and nondiscrimination as the primary criteria for job promotion and educational opportunity. Even if a desire to say the socially acceptable thing was an influence on responses in our survey, the generally sanguine outlook on ethnic heterogeneity at the neighborhood level is comforting. Certainly we found little evidence that Californians, whatever their racial and ethnic background or residential context, viewed current demographic trends in the state as creating the “ethnic cauldron” depicted by some journalists and scholars. At several points we noted that a majority in every racial and ethnic group seemed to view the effect of their own ethnic group on the neighborhood most favorably and to agree that people tended to be happier living and socializing with others of the same background. It would be highly misleading, however, to conclude that Californians consider residential and social segregation as a desirable goal. As other studies have shown (Farley et al., 1994), people of all ethnic groups indicate that their preferred neighborhood is one in which people of their own background are a majority, but not their only neighbors. Whites seemed most worried about the effect on property values of a heavy influx of black residents; blacks were most concerned about the prevalence of hostile attitudes in a largely white community. In addition, realism, not prejudice, may underlie the belief that it is easier to socialize with people with a similar culture and background. This separatist opinion is not the same as saying that people would not value friendship with people from other ethnic groups. Indeed, the rising rates of intermarriage between whites and Latinos and between whites and Asians 63 in California belie such a claim. Although there was considerable doubt about the feasibility of achieving a color-blind society, these data do not gainsay the accumulated evidence of public acceptance of integration (or diversity) in the workplace, in higher education, and at public gatherings such as movies, sporting events, and the like. To be sure, some people do move to live in ethnically more homogeneous neighborhoods; the phenomenon of white flight is real. Doubtless, too, there also are people who seek to live in culturally diverse communities. As long as there is the freedom to make these residential choices, there will be ethnic clusterings of all sorts, and only some neighborhoods will be ethnically mixed. What this outcome means for the quality of group relations depends in part on how ethnic context affects inter-group relations. The critical question that deserves additional research is whether the tolerance that groups express toward one another is genuine, or whether it is simply the culturally acceptable response that masks deeper hostilities. If it is the latter, we could be lulled into underestimating the extent of the potential problems arising from the changing ethnic composition of the state. The second major conclusion of this study is that multiethnic neighborhoods are not the primary locus of ethnic political divisions in this state. A number of previous academic studies have argued for the threat hypothesis, suggesting that racial tension should be highest in contexts where different groups interact the most. The present data showed that in most instances there was very little variation in perceptions and attitudes between ethnically mixed and homogeneous areas. Indeed, where there were differences between whites and nonwhites, they often were in the direction of greater tolerance and agreement between whites and nonwhites in mixed areas than in homogeneous ones. Clearly, a fuller understanding of the conditions under which ethnic context has a distinctive influence on white attitudes and political behavior requires more direct measures of perceived threat, inter-group contact, and residential choices reflecting different tastes for diversity. As noted above, ethnic context may be a proxy for several, potentially offsetting, social processes. The implication of our findings about the lack of contextual effects is that there is no reason to assume that as California’s growing ethnic 64 diversity multiplies the number of mixed neighborhoods, the level of ethnic tensions necessarily will increase. The fact that whites in multiethnic neighborhoods were not less tolerant than those living in mainly white areas means that efforts to reduce prejudice cannot simply be focused on specific localities experiencing ethnic change, such as the large cities. California’s political climate on racial and cultural issues depends as much (perhaps even more) upon the perceptions of the more numerous whites who live in majority white areas than whites who live near sizable minority populations. Concentrating only on the ethnically diverse areas would therefore neglect important sources of political tension. Among all ethnic groups, living in an ethnically mixed neighborhood boosted feelings of ethnic and racial identification. Clearly, how group consciousness—what social psychologists call in-group favoritism— affects political conduct is a vital issue that deserves close attention. From the perspective of minority groups, a stronger sense of racial and ethnic identity may fuel efforts to advance collective goals and redress inequality. Obviously, one implication of the finding that racial and ethnic identification among minorities is stronger in areas with a higher density of minorities suggests that efforts at political organization may be more successful by targeting residents in those neighborhoods. However, raising the salience of ethnic identity also may foster a backlash and intensify group conflict as the collision of in-group favoritisms makes for more out-group hostility. The third major conclusion is that attitude formation on racial and ethnic issues in California politics seems to occur more on a global than a local basis. There were few differences across neighborhoods sorted by their ethnic makeup, and where these did emerge they seemed to be more compositional than contextual. Differences in the background characteristics, ideology, and party affiliation of respondents primarily account for the observed variations across neighborhoods. To use the example of Proposition 209, people’s voting intentions were not so much influenced by what people perceived in their local areas as by their general political orientations and what they learned from the respective pro- and anti-209 campaigns. 65 This suggests that how political leaders and the mass media frame issues may be more critical to racial and ethnic attitudes than what goes on in local contexts. Because most people get their political information from the newspapers and the electronic media, local experience may be swamped by information diffused on a much broader basis. Even as they strive to improve group relations, policymakers must be aware that media coverage they cannot control may have a stronger influence on opinion formation. The final main conclusion of this study concerns the relatively negative orientation of Californians toward undocumented immigrants. Respondents in our poll—conducted two years after Proposition 187— had distinctively more positive views about legal immigrants than illegal ones. This finding held for all of the major ethnic groups. An implication of this is that policymakers should be careful not to lump legal and illegal immigrants together in assessing the public support for programs that affect these groups. Educating the public that the legal Latino population is much larger than the undocumented group also might reduce resentment about immigration and increase the public’s willingness to support policies that aid the larger Latino community. In sum, there is no evidence that ethnic group relations are particularly troubled in highly diverse areas or that these neighborhoods will serve as cauldrons for future problems. Racial attitudes seem to be more influenced by a person’s political and social background than by local context. Resistance to ethnic diversity is not more likely among those who directly experience it near where they live. In some ways, this is a hopeful note on which to conclude, because it tells us that ethnic tensions in California are not preordained to increase as the state continues to cope with the effect of its changing demography. 66 Appendix A Survey Sample and Data Sources Our data are drawn from a telephone survey of 1,500 registered voters conducted between October 26 and November 2, 1996, just before the general election. The sample was designed by Dr. Ken McCue. Using California’s Statewide Database, formerly known as the Assembly’s Redistricting Database, the entire state was divided into “consolidated precincts.” The consolidated precinct level is a unit of analysis whose creation becomes necessary when merging census data with election results or registration data. For the purpose of this sample design, the consolidated precinct level is used to represent a “neighborhood” or context. Census data merged with registration data had to be used to draw the sample because the objective was to find registered voters of certain racial categories who lived in the same neighborhood. The Census does not report registration data, however, and the statement of registration for voting does not report race or ethnicity. Census data are block-level data that are, in most cases, reported by the block group. Election data are collected on the electoral precinct level and registration data on the registration precinct level. Both sets of precincts encompass census blocks as the smallest unit of analysis, but they do not share all boundaries in most cases. One important objective of this sample design was to locate minority groups in varying racial contexts to enable comparisons of, for example, blacks living in majority black neighborhoods with blacks in majority minority neighborhoods. The sample was constructed by dividing registered voters into 10 categories and then drawing a random sample from each stratum, listed below as distinctive “objective” contexts. Thus, our purpose was not to represent the state’s electorate; rather, we wanted to include sufficient numbers of minority respondents to enable systematic comparisons of all ethnic groups across contexts. Given this research design, it is not surprising that the lower level of 67 support for Proposition 209 within our sample was lower than the actual vote. The first question in our survey summarized Proposition 209 and asked respondents: “If the election were being held today, would you vote yes or no on this initiative?” Those in favor constituted 41 percent of the sample, and another 15 percent said they were still undecided. As noted above, the contextual variable we employ as an indicator of racial threat is the size of the minority population within a consolidated precinct. As also noted above, 1990 Census data about the ethnic “mix” within a precinct is our “objective” measure of context. Respondents were grouped according to whether they lived in majority white precincts; majority minority precincts, in which the total of black, Latino, and Asian residents rather than a single ethnic group constituted a majority; majority black precincts; or majority Latino districts. Clearly, though, in an ethnically diverse society such as California, census data enable a number of alternative categorizations of ethnic context. For example, one might measure the size of each minority group separately or employ different combinations depending on the particular issue (e.g., language policy or affirmative action). Another possibility is to define context by the relative economic status of particular ethnic groups rather than their relative size. Similarly, the specific nature of the hypothesized group threat might dictate one’s choice of contextual variable. For example, if one believed that ethnic change posed a “cultural” threat to the majority group, then language use rather than ethnic origin might be the appropriate grouplevel variable to use. By the same token, if the hypothesized collective threat were economic, then economic conditions within a jurisdiction, such as mean per capita income or unemployment level, might be considered potential contextual sources of voting. 68 Appendix B Multivariate Analyses from Chapters 2 and 3 69 Table B.1 Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 (OLS regression) 70 % minority in precinct Black Hispanic Asian Gender Education Age PID Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Effect of Blacks B SE 0.03 0.06 –0.15** 0.05 –0.05 0.04 –0.05 0.04 0.06* 0.03 –0.01 0.01 0.01 0.02 0.03 0.02 0.00 0.02 –0.01 0.03 0.02 0.02 0.02 0.02 0.01 0.01 1.82** 0.12 Effect of Latinos B SE –0.08 0.06 0.01 0.06 –0.15** 0.05 0.00 0.05 0.05# 0.03 –0.02 0.02 0.01 0.02 0.05* 0.02 –0.02 0.02 0.03 0.03 0.06** 0.02 0.05* 0.02 0.03* 0.02 1.60** 0.14 Effect of Asians B SE –0.11# 0.06 0.03 0.06 –0.01 0.04 –0.14** 0.05 0.07* 0.03 –0.04** 0.01 0.00 0.02 0.00 0.02 –0.01 0.02 0.01 0.03 0.03 0.02 0.03 0.02 0.02 0.01 1.82** 0.13 Effect of Whites B SE –0.07 0.05 0.05 0.05 0.03 0.04 –0.05 0.04 0.03 0.03 0.01 0.01 –0.02 0.02 –0.02 0.02 0.03 0.02 –0.05* 0.03 0.06** 0.02 0.00 0.02 0.02 0.01 1.81** 0.12 Effect of Legal Immigrants B SE –0.04 0.07 –0.01 0.06 –0.09# 0.05 –0.10* 0.05 –0.01 0.04 –0.03# 0.02 –0.04 0.02 –0.01 0.02 –0.01 0.02 0.00 0.03 0.04# 0.03 0.04 0.02 0.02 0.02 1.93** 0.15 Adjusted R–squared n 0.02 1,036 0.05 1,041 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. 0.03 1,010 0.01 1,043 0.01 922 Effect of Illegal Immigrants B SE 0.20** 0.08 –0.08 0.07 –0.10# 0.05 –0.02 0.06 –0.01 0.04 0.00 0.02 –0.03 0.03 0.06* 0.03 0.00 0.03 0.09** 0.04 0.04 0.03 0.01 0.03 0.03 0.02 1.68** 0.16 0.02 829 Table B.2 Group Effects on the Quality of Life in Respondents’ Neighborhoods, from Chapter 2 (OLS regression) 71 Blacks in majority black precincts Blacks in majority minority precincts Blacks in majority white precincts Latinos in majority Latino precincts Latinos in majority minority precincts Latinos in majority white precincts Asians in majority minority precincts Asians in majority white precincts Whites in majority minority precincts Gender Education Age PID Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Effect of Blacks B SE –0.19** 0.06 –0.08 0.08 0.11 0.12 –0.03 0.06 0.01 0.06 –0.05 0.06 –0.02 0.06 –0.06 0.06 0.04 0.05 0.07* 0.03 –0.02 0.01 0.01 0.02 0.03 0.02 0.00 0.02 –0.01 0.03 0.02 0.02 0.02 0.02 0.01 0.01 1.82** 0.12 Effect of Latinos B SE –0.04 0.07 –0.10 0.09 0.11 0.13 –0.26** 0.06 –0.19** 0.07 –0.06 0.07 –0.02 0.06 –0.02 0.07 –0.02 0.05 0.06# 0.03 –0.02 0.02 0.01 0.02 0.05* 0.02 –0.01 0.02 0.03 0.03 0.06** 0.02 0.05* 0.02 0.03* 0.02 1.58** 0.13 Effect of Asians B SE 0.02 0.06 –0.11 0.09 –0.04 0.12 –0.11# 0.06 0.02 0.06 0.01 0.07 –0.23** 0.06 –0.08 0.06 –0.01 0.05 0.07* 0.03 –0.04** 0.01 –0.01 0.02 0.01 0.02 –0.01 0.02 0.01 0.03 0.04# 0.02 0.03 0.02 0.03# 0.01 1.76** 0.13 Adjusted R-squared n 0.02 1,036 0.05 1,041 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. 0.04 1,010 Effect of Whites B SE –0.06 0.06 0.09 0.08 0.06 0.11 –0.07 0.05 0.03 0.06 –0.01 0.06 –0.13** 0.05 –0.04 0.06 –0.07 0.05 0.03 0.03 0.01 0.01 –0.02 0.02 –0.02 0.02 0.03 0.02 –0.06* 0.03 0.06** 0.02 0.01 0.02 0.02 0.01 1.82** 0.11 0.02 1,043 Effect of Legal Immigrants B SE –0.12# 0.07 0.01 0.10 –0.07 0.14 –0.23** 0.07 –0.10 0.07 –0.11 0.07 –0.19** 0.07 –0.13# 0.07 –0.13* 0.06 –0.01 0.04 –0.03# 0.02 –0.04 0.02 –0.01 0.02 –0.02 0.02 –0.01 0.03 0.04# 0.03 0.04# 0.02 0.02 0.02 1.98** 0.14 0.02 922 Effect of Illegal Immigrants B SE 0.09 0.08 –0.09 0.11 0.05 0.15 0.04 0.08 –0.01 0.08 –0.07 0.08 0.09 0.08 –0.01 0.08 0.11# 0.07 –0.01 0.04 –0.01 0.02 –0.03 0.03 0.05* 0.03 0.01 0.03 0.10** 0.04 0.04 0.03 0.01 0.03 0.03 0.02 1.73** 0.16 0.02 829 Table B.3 Racial Identification and Racial Attitudes—Black Respondents, from Chapter 3 (OLS regression) 72 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –1.27** 0.36 0.19 0.15 0.01 0.07 0.01 0.10 2.71** 0.45 Color-Blind Society B SE –0.38 0.58 –0.47# 0.26 –0.34** 0.12 –0.05 0.16 0.04 0.25 0.00 0.17 –0.31 0.22 –0.19 0.18 0.11 0.16 0.04 0.12 4.65** 1.20 Adjusted R–squared 0.05 0.03 n 190 159 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE 0.30 0.49 –0.08 0.22 –0.12 0.10 –0.08 0.13 –0.10 0.21 –0.07 0.14 0.01 0.19 0.19 0.15 –0.07 0.13 0.13 0.10 1.81# 1.02 0.02 161 Happier with Others of Same Background B SE 0.04 0.54 –0.04 0.24 0.11 0.11 –0.13 0.15 –0.34 0.25 0.09 0.16 –0.06 0.21 0.39* 0.17 0.07 0.15 0.12 0.11 1.74 1.14 Minorities Gain at Expense of Others B SE 1.08# 0.56 –0.19 0.25 –0.18 0.12 –0.27# 0.15 –0.06 0.24 –0.33* 0.17 –0.59** 0.22 0.24 0.17 0.03 0.15 0.20# 0.12 4.14** 1.18 Minorities Get More Than They Deserve B SE –0.08 0.57 –0.04 0.25 –0.39** 0.12 –0.10 0.16 –0.05 0.26 0.04 0.17 0.16 0.23 0.11 0.17 0.16 0.16 0.03 0.12 3.15** 1.20 0.01 0.07 0.04 156 156 158 Table B.4 Racial Identification and Racial Attitudes—Latino Respondents, from Chapter 3 (OLS regression) 73 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –0.42# 0.24 0.06 0.12 0.11* 0.06 0.06 0.08 1.96** 0.29 Color-Blind Society B SE –0.09 0.38 0.03 0.19 –0.23** 0.09 0.21 0.13 –0.02 0.13 –0.29* 0.12 –0.44* 0.18 –0.15 0.13 0.14 0.12 –0.03 0.08 4.64** 0.76 Adjusted R-squared 0.01 0.03 n 367 303 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE –0.03 0.33 –0.07 0.17 –0.21** 0.08 –0.12 0.11 0.00 0.12 0.15 0.11 0.07 0.16 –0.01 0.12 0.21* 0.11 0.23** 0.07 1.58* 0.67 Happier with Others of Same Background B SE –0.52 0.33 0.08 0.17 –0.10 0.08 0.05 0.11 0.14 0.12 0.23* 0.11 –0.06 0.16 –0.18 0.12 0.03 0.11 0.20** 0.07 2.14** 0.67 Minorities Gain at Expense of Others B SE –0.32 0.35 –0.29 0.18 –0.21** 0.08 0.05 0.12 0.05 0.12 –0.11 0.12 0.01 0.17 –0.04 0.12 –0.05 0.11 –0.15* 0.08 4.22** 0.70 Minorities Get More Than They Deserve B SE 0.59 0.38 0.24 0.19 –0.12 0.09 –0.10 0.13 0.02 0.13 –0.12 0.12 0.04 0.18 –0.04 0.13 –0.34** 0.12 0.05 0.08 3.76** 0.76 0.05 0.04 307 301 0.02 0.02 300 303 Table B.5 Racial Identification and Racial Attitudes—Asian Respondents, from Chapter 3 (OLS regression) 74 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –0.09 0.27 –0.01 0.13 0.06 0.06 0.06 0.09 1.90** 0.35 Color-Blind Society B SE 0.53 0.44 0.05 0.21 –0.26* 0.10 0.38** 0.15 0.10 0.14 –0.17 0.15 –0.36# 0.20 –0.32# 0.17 0.00 0.14 0.12 0.10 3.92** 0.84 Adjusted R–squared 0.01 0.08 n 248 196 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE –0.44 0.41 0.14 0.20 –0.13 0.09 0.10 0.13 –0.09 0.13 0.17 0.14 –0.19 0.18 –0.01 0.16 0.34** 0.13 0.22* 0.10 1.74* 0.77 Happier with Others of Same Background B SE –0.39 0.42 0.32 0.20 –0.03 0.10 0.02 0.14 –0.02 0.13 –0.17 0.14 0.13 0.18 0.01 0.16 0.14 0.13 0.03 0.10 2.04** 0.79 Minorities Gain at Expense of Others B SE –0.13 0.42 –0.49* 0.21 –0.09 0.10 0.36* 0.14 0.04 0.14 0.30* 0.15 0.12 0.19 –0.11 0.17 0.14 0.13 –0.12 0.10 2.20** 0.82 Minorities Get More Than They Deserve B SE 0.05 0.44 –0.31 0.22 –0.31** 0.10 0.04 0.15 0.12 0.14 0.19 0.16 0.29 0.20 –0.11 0.18 –0.05 0.14 0.02 0.10 3.47** 0.85 0.05 0.02 195 194 0.07 191 0.04 185 Table B.6 Racial Identification and Racial Attitudes—White Respondents, from Chapter 3 (OLS regression) 75 % minority in precinct Gender Education Age Party identification Ideology Neighborhood diversity Personal financial situation State economic situation Racial identification Constant Racial Identification B SE –0.47** 0.17 0.08 0.09 0.06 0.04 0.02 0.06 2.86** 0.23 Color–Blind Society B SE 0.05 0.27 –0.22 0.15 –0.12# 0.07 –0.05 0.10 0.19* 0.09 0.11 0.11 –0.08 0.13 –0.07 0.10 0.04 0.10 0.03 0.07 3.00** 0.58 Adjusted R-squared n 0.01 0.01 516 438 NOTES: B is regression coefficient; SE is standard error. *p<0.05; **p<0.01; #p<0.10. Minorities Have Less Opportunity B SE –0.36 0.26 –0.10 0.14 –0.10 0.06 0.08 0.09 0.22** 0.08 0.13 0.11 –0.20 0.13 –0.04 0.10 0.39** 0.10 0.00 0.07 2.01** 0.56 0.09 439 Happier with Others of Same Background B SE –0.06 0.23 0.22# 0.13 0.23** 0.06 –0.06 0.08 –0.10 0.08 –0.06 0.09 –0.10 0.12 –0.02 0.09 –0.04 0.09 0.12* 0.06 1.93** 0.50 Minorities Gain at Expense of Others B SE –0.11 0.25 0.01 0.14 –0.15** 0.06 –0.04 0.09 0.24** 0.08 0.37** 0.10 0.02 0.12 0.17# 0.10 –0.05 0.10 –0.06 0.07 2.46** 0.54 Minorities Get More Than They Deserve B SE 0.10 0.26 –0.16 0.14 –0.32** 0.06 0.06 0.09 0.21** 0.08 0.10 0.10 0.16 0.13 0.05 0.10 0.14 0.10 0.08 0.07 2.78** 0.56 0.05 0.09 426 429 0.10 431 Appendix C Survey Question Wording Prop 209 Proposition 209, referred to as the California Civil Right Initiative, sometimes known as CCRI, would prohibit state and local governments from using race, sex, color, ethnicity or national origin as criteria for either discriminating against or granting preferential treatment to individuals or groups in hiring, promoting, granting admissions to college or selecting public contractors. If the election were being held today, would you vote yes or no on this initiative? Vote for President Who do you think you will vote for in the election for president? Neighborhood Diversity Do you think your neighborhood is becoming more ethnically diverse, less ethnically diverse, or is about the same? Neighborhood Effect of Groups Now I am going to ask you what effect different groups of people have had on the quality of life in the neighborhood where you live. After I give you the name of a group, please tell me whether that group has improved, made worse, or had no effect on the quality of life in your neighborhood. Blacks Latinos Asians Illegal Immigrants 77 Legal Immigrants Whites Who Still Needs Affirmative Action? As you know, Proposition 209 will eliminate all state and local government actions which in the past have provided special programs intended to benefit certain groups of people. Which of the following groups do you think still need the programs eliminated by Proposition 209? Blacks Latinos Asians Women Meaning of 209 Vote When you vote on initiatives like Prop 209, what do you most want to accomplish, change or preserve the law, or send a message to government officials? Will Private Companies Continue AA? Proposition 209 does not prohibit companies and businesses from providing special programs which benefit certain groups of people. Do you think that private companies and businesses would continue to provide these programs if Proposition 209 passes? At what level? If so, would they provide them at the same level as in the past or at a lesser level? Racial Attitudes Battery I am going to read you some statements which some people agree with, while others do not. Please tell me if you agree strongly, agree somewhat, are neutral, disagree somewhat, or disagree strongly. 78 1. The more jobs and places in college provided to minorities, the fewer there are for people who are not members of those groups. 2. In our lifetime, America can become a color-blind society. 3. Members of certain ethnic or racial groups still have less opportunities to get ahead than other people. 4. Members of particular ethnic or racial groups use special programs to get more benefits than they deserve. 5. People of different ethnic and racial groups are generally happier when they live and socialize with others of the same background. Diversity or Merit? Which of these statements come closer to the way you feel? 1. Diversity benefits our country economically and socially, so race, ethnicity, and gender should be a factor in determining the type of person who is hired, promoted or admitted to college. 2. Hiring, promotion and college admission should be based solely on merit and qualifications and not on characteristics of race, ethnicity or gender. Racial Identity in Politics When it comes to social and political matters, some people think of themselves mainly as black, white, Latino, Asian or Jewish, and that is very important to how they think of themselves. Other people don’t give as much thought to these things. When it comes to social and political matters, how important is your race and ethnicity to how you think of yourself? Personal Knowledge of Undeserved Job Do you have personal knowledge of a situation where a woman or a member of a racial minority got a job or promotion that he or she did not deserve as a result of affirmative action programs, or do you have no such knowledge? 79 Vote in 1994? Did you vote in the 1994 November election? Prop 187 Vote [If yes on 14] In 1994, there was a proposition on the ballot in California to deny governmental services to illegal immigrants called Proposition 187. How did you vote on Proposition 187? Ideological Self-Identification Do you consider yourself a liberal, moderate, or conservative? 209 Effect on Presidential Vote Would you change your vote for president based on whether or not that candidate had endorsed or opposed Proposition 209? Personal Financial Situation Would you say over the past year your personal financial situation has gotten better, worse, or stayed the same? California Economic Situation What about the economy of California? Would you say over the past year it has gotten better, worse, or stayed the same? Main Effect of 209 If Proposition 209 passes, do you think that its main effect will be to change California law or to send a message to government officials? Education What was the last grade of school you completed? Race What race do you consider yourself? (White, Black, Asian, or other?) 80 Asian Nationality [If Asian on question 22] What nationality do you consider yourself? (Japanese, Chinese, Korean, Vietnamese, Cambodian, Other?) Latino [If not Asian on question 22] What ethnicity do you consider yourself? (Latino or Other?) 81 Bibliography Allport, Gordon W., The Nature of Prejudice, Doubleday Anchor Books, Garden City, NJ, 1954. Blalock, Hubert M., Toward a Theory of Minority-Group Relations, Capricorn Books, New York, 1967. Bowers, Jake, “Black Threat and Black Context: New Evidence from the 1997 NES Pilot Study,” paper presented at the Annual Meeting of the International Society of Political Psychology, Montreal, Canada, 1998. Brown, Roger, Social Psychology (2nd ed.), Basic Books, New York, 1986. Cain, Bruce E., D. Roderick Kiewiet, and Carole J. Uhlaner, “The Acquisition of Partisanship by Latinos and Asian Americans,” American Journal of Political Science, Vol. 35, 1991, pp. 390–422. Citrin, Jack, Donald P. Green, Christopher Muste, and Cara Wong, “Public Opinion Toward Immigration Reform: The Role of Economic Motivations,” Journal of Politics, Vol. 59, 1997, pp. 858– 881. Clark, William A. V., “Residential Patterns: Avoidance, Assimilation, and Succession,” in Roger Waldinger and Mehdi Bozorgmehr, eds., Ethnic Los Angeles, Russell Sage Foundation, New York, 1996. Dawson, Michael C., Behind the Mule: Race and Class in AfricanAmerican Politics, Princeton University Press, Princeton, NJ, 1994. Farley, Reynolds, Charlotte Steeh, Maria Krysan, Tara Jackson, and Keith Reeves, “Stereotypes and Segregation: Neighborhoods in the Detroit Area,” American Journal of Sociology, Vol. 100, 1994, pp. 750– 780. Giles, Michael W., and Melanie A. Buckner, “David Duke and Black Threat: An Old Hypothesis Revisited,” The Journal of Politics, Vol. 55, 1993, pp. 702–713. 83 Glaser, James, “Back to the Black Belt: Racial Environment and White Racial Attitudes in the South,” Journal of Politics, Vol. 56, 1994, pp. 21–41. Green, Donald P., Robert P. Abelson, and Margaret Garnett, “The Distinctive Political Views of Hate-Crime Perpetrators and White Supremacists,” in Deborah A. Prentice and Dale T. Miller, eds., Cultural Divides: Understanding and Overcoming Group Conflict, Russell Sage Foundation, New York, 1999. Hood, M.V., III, and Irwin L. Morris, “Amigo o Enemigo? Context, Attitudes, and Anglo Public Opinion Toward Immigration,” Social Science Quarterly, Vol. 78, 1997, pp. 309–323. Key, V.O., Southern Politics in State and Nation, University of Tennessee Press, Knoxville, 1949. Levine, Robert, and Donald Campbell, Ethnocentrism, John Wiley & Sons, New York, 1972. Liska, Allen E., Mitchell B. Chamlin, and Mark D. Reed, “Testing the Economic Production and Conflict Models of Crime Control,” Social Forces, Vol. 64, 1985, pp. 119–138. Olzak, Susan, The Dynamics of Ethnic Competition and Conflict, Stanford University Press, 1992. Prentice, Deborah A., and Dale T. Miller, Cultural Divides: Understanding and Overcoming Group Conflict, Russell Sage Foundation, New York, 1999. Quillian, Lincoln, “Prejudice as a Response to Group Threat: Population Composition and Anti-Immigrant and Racial Prejudice in Europe,” American Sociological Review, Vol. 60, 1995, pp. 586–611. Quillian, Lincoln, “Group Threat and Regional Change in Attitudes Toward African Americans,” American Journal of Sociology, Vol. 102, 1996, pp. 816–860. Schuman, Howard, Charlotte Steeh, Lawrence Bobo, and Maria Krysan, Racial Attitudes in America: Trends and Interpretations (rev. ed.), Harvard University Press, Cambridge, MA, 1997. Stephan, Walter, “The Contact Hypothesis in Intergroup Relations,” in C. Hendrick, ed., Review of Personality and Social Psychology, Vol. 9, Sage Publications, Newbury Park, CA, 1987, pp. 13–40. 84 Tajfel, Henri, “Social Categorization, Social Identity, and Social Comparison,” in H Tajfel, ed., Differentiation Between Social Groups, Academic Press, London, 1978. Tate, Katherine, From Protest to Politics: The New Black Voters in American Elections, Harvard University Press, Cambridge, MA, 1993. Taylor, Marylee C., “How White Attitudes Vary with the Racial Composition of Local Populations: Numbers Count,” American Sociological Review, Vol. 63, 1998, pp. 512–535. Tolbert, Caroline, and Rodney Hero, “Race/Ethnicity and Direct Democracy: An Analysis of California’s Illegal Immigration Initiative,” Journal of Politics, Vol. 58, 1996, pp. 806–818. Uhlaner, Carole J., “Perceived Discrimination and Prejudice and the Coalition Prospects of Blacks, Latinos, and Asian Americans,” in Byran O. Jackson and Michael B. Preston, eds., Racial and Ethnic Politics in California, IGS Press, Institute of Governmental Studies, University of California at Berkeley, Berkeley, CA, 1991. Valenty, Linda O., Ron Sylvia, and Peter Haas, “Thresholds for Tolerance: The Impact of Racial and Ethnic Composition on California Propositions 209 and 187,” paper presented at the Annual Meeting of the International Society of Political Psychology, Montreal, Canada, 1998. Voss, D. Stephen, “Beyond Racial Threat: Failure of an Old Hypothesis in the New South,” Journal of Politics, Vol. 58, 1996, pp. 1156–1170. Wong, Cara, “Who Is This ‘We’ We Keep Talking About: Conceptualization and Measurement of Community,” paper presented at the Annual Meeting of the American Political Science Association, Boston, 1998. Wright, Gerald C., “Contextual Models of Electoral Behavior: The Southern Wallace Vote,” American Political Science Review, Vol. 71, 1977, pp. 497–508. 85 About the Authors BRUCE CAIN Bruce Cain is the Director of the Institute of Governmental Studies and Robson Professor of Political Science at the University of California, Berkeley. A graduate of Bowdoin College in 1970, he was a Rhodes Scholar at Trinity College, Oxford, and received his Ph.D. in political science from Harvard University in 1976. From 1976 to 1989, he taught California politics, political theory, and comparative government at the California Institute of Technology. His writings include The Reapportionment Puzzle, The Personal Vote (with John Ferejohn and Morris Fiorina), Congressional Redistricting (with David Butler), several co-edited volumes, and numerous articles for professional journals. He has served as a polling consultant for several state senate races and as a redistricting consultant to the California State Assembly, Los Angeles City Council, and Attorney General of the State of Massachusetts. He has been a consultant to the Los Angeles Times and a frequent commentator for radio and television stations in Los Angeles and the Bay Area. JACK CITRIN Jack Citrin is Professor of Political Science at the University of California, Berkeley, where he has taught since 1990. He received his B.A. from McGill University and his Ph.D from Berkeley. His research interests are American national politics, comparative politics in advanced industrial societies and California politics, with a focus on public opinion, political trust, and ethnic conflict. He is the author of The Politics of Disaffection among British and American Youth (1975), Tax Revolt: Something for Nothing in California (1982), and numerous articles and book chapters. With Professor David Sears of UCLA, he is presently completing a book entitled The Politics of Multiculturalism: Crisis of American Indentities. CARA WONG Cara Wong is a doctoral student at the University of California, Berkeley. Beginning in the Fall of 2000, she will be an assistant professor of political science at the University of Michigan. Her research interests include racial and ethnic politics in the United States, public opinion and voting behavior, and research methodology. 87" ["post_date_gmt"]=> string(19) "2017-05-20 09:35:04" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_800bcr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:35:04" ["post_modified_gmt"]=> string(19) "2017-05-20 09:35:04" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_800BCR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }