"Coloured Power" or "Coloured Bourgeoisie"? Political Attitudes among South African Coloureds

Why have Coloureds in South Africa—people of mixed racialdescent—voiced less opposition to white minority rulethan have black Africans? A theoretical analysis suggests thatColoureds feel both "relatively deprived" in comparison withwhites and "relatively gratified" in comparison with Africans,that they believe they may eventually be accepted by whites,and that individually they have difficulty identifying withthe broader Coloured community; in short, seemingly ideal conditionsfor the emergence of a "Coloured Bourgeoisie" rather than a"Coloured Power" ideology.     

"Some of my best friends are black...": Interracial Friendship and Whites' Racial Attitudes

We examine the major tenets and assumptions of the well-knowncontact theory of prejudice, and we compare them with the morecynical reasoning implied by the infamous "Some of my best friendsare black, but..." expression. After assessing the extant evidencefor the contact theory, we use a unique set of national surveydata to address the central postulates of that theory. We examinethe racial beliefs, feelings, social dispositions, and policyviews of whites who have contact with blacks as friends, acquaintances,or neighbors. Our results suggest that personal interracialcontact is selective in its effects on whites' racial attitudes,that intimacy is less important than variety of contacts, andthat any effects are contingent on the relative socioeconomicstatus of black contacts. On the basis of our analysis, we reassessthe contact theory and propose a more political conception ofthe attitudes of dominant groups toward subordinates. We arguethat the message contained in the relationship between personalcontact with subordinates and intergroup attitudes is less benignthan is suggested by the contact theory.     

On More Reliably Employing the Concept of "Reliability"

Confusion has resulted from using the term "reliability" tomean both measurement-error variance and sampling-error variance.This article discusses the conceptual similarities and differencesbetween the various types of variance subsumed under "reliability"in order to distinguish among them.     

The "Positivity Bias" inEvaluations of Public Figures:Evidence Against Instrument Artifacts

The "positivity bias" is a term used to describe the consistentfavorable evaluation of public figures found in surveys overthe past 40 years. This paper explored several possible artifactualexplanations for this bias,focusing on the survey instrumentitself. Two experiments varied the labeling and ordering ofscale endpoints, the affective value of the initial contextevaluated, and the presence or absence of a prestigious jobtitle associated with the nameof the public figure. None ofthe variations produced significantly different levels of positivitythan the standard control condition used in each experiment.RichardR. Lau is a graduate student in social psychology at the Universityof California, Los Angeles. David O. Sears is Professor of Psychologyand Political Science at UCLA. Richard Centers is Professorof Psychology at UCLA. This research was supported in part byGrant #SOC73-09153 A03 from the National Science Foundationto David O. Sears. The authors wish to expess their thanks toMark Williams, who did most of the coding and checking involvedin the survey.     

CHANGING RACIAL LABELS: FROM "COLORED" TO "NEGRO" TO "BLACK" TO "AFRICAN AMERICAN"

Labels play an important role in defining groups and individualswho belong to the groups. This has been especially true forracial and ethnic groups in general and for Blacks in particular.Over the past century the standard term for Blacks has shiftedfrom "Colored" to "Negro" to "Black" and now perhaps to "AfricanAmerican." The changes can be seen as attempts by Blacks toredefine themselves and to gain respect and standing in a societythat has held them to be subordinate and inferior.     

Correction to "The Polls: Environmental Protection"

In "The Polls: Environmental Protection" (POQ 50 (2):273) thefollowing figures are the correct ones: NORC-GSS, 1985, Toomuch: 8%, Dk: 4%; Roper, 1983, Too far: 14%.     

Reporting "The Polls" in 2004

Media reports of polls indicate how well public opinion pollshave been integrated into campaign coverage. This article examineshow polls were used in 2004. Although there were relativelylimited methodological changes in how polls were conducted in2004, there were changes in how the polls were treated in themedia. Americans in 2004 were subjected to intense debates aboutpolls and to as much reporting about "the polls" as there wasof the polls themselves. The discussion of "the polls" in 2004included claims of electability during the Democratic nominatingprocess, increased reporting about methodological issues, andheightened political criticisms of "the polls." The articleconcludes with a discussion of the current state and the futureof news polling.     

Between Ideology and Economy: The "Time Politics" of Child Care and Public Education in the Two Germanys

The article compares the discourses and policies of the "timepolitics" of public, that is, institutionalized, extra-familialcare and education of pre- and elementary school children inEast and West Germany. The FRG and the GDR represented two highlydistinct welfare and education systems, which referred to eachother in a complex relationship of "distancing and interconnection."Proceeding from the concept of the "path dependency" of societaland political developments, the article analyzes which factorsworked together when, how, and in which concrete historicalcontexts to form specific discourses and policies of the "timepolitics" in both German states.     

"Walking the Walk" of Public Service Motivation: Public Employees and Charitable Gifts of Time, Blood, and Money

Research generally indicates that public employees "talk thetalk," but do they also "walk the walk" of the public servicemotive (PSM)? Are public service employees more likely thanothers to engage in public service activities? The behavioralimplications of PSM are addressed by studying the involvementin charitable activities of public, nonprofit, and private workers.Using data from the 2002 General Social Survey, multivariatelogistic regression models are estimated to examine self-reportedgifts of time, blood, and money to charitable organizations.It is found that government employees are more likely to volunteerfor charity and to donate blood than for-profit employees are.Additionally, nonprofit workers are also more likely than theirfor-profit counterparts to volunteer. However, no differenceis found among public service and private employees in termsof individual philanthropy. These findings generally lend supportfor the hypothesis that PSM is more prominent in public servicethan in private organizations, especially as it pertains togovernment personnel.     

Following the "Mail Hawks": Alternative Measures of Public Opinion on Vietnam in the Johnson White House

Scholars argue that public opinion grew in importance duringthe Vietnam conflict, yet most find President Johnson was notresponsive to public opinion during the War. We amplify thesetheories by demonstrating the practical value of public opinionmail sent to the White House on Vietnam, reshaping theoriesabout the constraining role of public opinion in foreign policy.We find that the White House mail, but not opinion polling,favoring escalation of the War had a significant and positiveimpact on President Johnson's policy rhetoric. From these andsimilar findings, we conclude that the Johnson Administrationfollowed core "hawkish" political allies (those individualsdesiring a rapid escalation and quick end to the war) ratherthan those approving of a withdrawal, suggesting mail-gaugedopinion from electoral partners (and core political allies morebroadly) has value in foreign policy making.     

The Abortion Attitudes Paradox: Model Specification and Gender Differences

Rather unexpectedly, prior work has failed to find consistent gender differences in public support for legal abortion. Given that gender differences in public opinion emerge for a wide range of other issue areas, it seems paradoxical that there is no consistent gender difference on the issue of abortion. I propose that this failure to find a consistent gender difference is due to how abortion attitudes are modeled. Controlling for religiosity, which research has shown women to score higher on, results in a small and consistent gender gap in support for legal abortion with women more likely than men to support.     

"Little" and "Big" Pictures in Our Heads: Race, Local Context, and Innumeracy About Racial Groups in the United States

Americans do not know what percentage of the nation's residentsare whites, blacks, Hispanics, Asians, and American Indians.Using the 2000 General Social Survey, I find that respondentsof all races underestimate the percentages of whites and overestimatethe percentages of racial/ethnic minority groups and multiracialAmericans in the United States; however, they perceive theirlocal communities quite differently. As a first step towardunderstanding this discrepancy, I test whether individuals’local surroundings serve as a source of information for theirpictures of the United States. I examine the relationship between"objective" data and respondents’ subjective perceptionsof where they live, and compare their respective effects onAmericans’ perceptions of the nation. Multivariate multilevelanalyses show that respondents’ perceptions of differentracial group sizes in their communities are the strongest predictorsof innumeracy at the national level, while "objective" racialcontext measured at the local level has less of an effect. Thesefindings have important implications for research on racialcontext, which assumes that census numbers for respondents’locales are good proxies for their perceptions of the size ofracial/ethnic groups in their communities. Furthermore, thesefindings suggest that scholars need to start thinking aboutwhy whites and non-whites have similar "big pictures" of thenation, why their "little pictures" vary a great deal, and whythe motivations for over- and underestimation may differ byracial/ethnic group.