How Americans Understand Racial and Religious Differences: A Test of Parallel Items from a National Survey

Abstract

How can we better conceptualize attitudes about difference in an increasingly diverse, multicultural United States? This article uses data from a recent, nationally representative telephone survey with oversamples of African Americans and Hispanics to analyze attitudes about two prominent sources of distinction in the United States. Race and religion were selected because they tend to be understood in very different ways—race as a social problem, religion as an individual choice and collective good. To assess the utility of these contrasting emphases built into common survey measures, we constructed a battery of questions that included parallel items for both. Our findings indicate that, with some notable exceptions, Americans' attitudes tend to be more similar than different, such that respondents see comparable (rather than contrasting) positive and negative aspects of race and religion in the United States. Based upon these results, we argue for a more multifaceted approach to the conceptualization, measurement, and analysis of race and religion, with implications for how we generally approach difference, diversity, and multiculturalism.

NOTES

Notes

This research was supported by the Edelstein Family Foundation as part of the American Mosaic Project in the College of Liberal Arts at the University of Minnesota.

1. These survey questions come from the seminal work Racial Attitudes in America (Schuman et al. 1997).

² The first two items come from the Religion and Public Life 2002 survey conducted by the Pew Center. The third item was adapted from multiple variations of an item that is popular on many different congregational surveys.

³ As the voluminous new literature on whiteness would suggest, the racial identification of white Americans is quite complicated and multifaceted—sometimes it is hidden or invisible (Doane 1997), sometimes it is expressed in virulent forms of white supremacy, and in other cases through the more benign re-adoption of “symbolic” white ethnic identities (Waters 1990). More on this below.

⁴ As of December 2010, the data and code books for the American Mosaic Project survey are housed at the Inter-university Consortium for Political and Social Research (ICPSR) at the University of Michigan (Hartmann, Gerteis, and Edgell 2003). This can be accessed online at http://dx.doi.org/10.3886/ICPSR28821.

⁵ This response rate compares favorably with that of most recent national RDD-based studies. For example, a recent American National Election Study (ANES) had a response rate of about 35 percent using a fairly generous definition, compared with the AMP response rate of 36 percent using the same definition. Here, it is also worth noting that the Council on Market and Opinion Research (CMOR) maintains an ongoing study of response rates, using calculation methods consistent with what we used, and their study shows that the current mean response rate for RDD telephone surveys is 10.16 percent.

⁶ We do not claim that these four are the only key conceptual dimensions for measuring attitudes about difference in U.S. society, but these four seem to be those most commonly addressed in the race and religion literatures and surveys.

⁷ For these analyses, “nonbelievers” are those claiming no religious identity (also called religious “nones” in some contemporary accounts). We chose this to parallel our other measures used in the religious subgroup analysis, which are based on self-identification.