Vincent Hutchings, University of Michigan
In this chapter, the task force focuses on the beliefs and opinions that Americans have about inequality. Specifically, we examine the commitment to egalitarianism as a norm and how this commitment varies across groups defined by race, class, ideology, and political party. Additionally, we examine support for governmental efforts to reduce income inequality across racial groups, and the extent to which attitudes about equality in the abstract are shaped by attitudes about equality for different class and racial groups. Lastly, we focus on levels of racial group identity, how this concept can be measured, and whether it varies over time and across different groups.
Since 1984, the American National Election Studies (ANES) has sought to measure support for the value of egalitarianism. The battery, originally designed by Stanley Feldman, consists of six items incorporating the concepts of equal opportunity, concerns about the pace of equal rights, whether the failure to provide equal opportunity is a big problem, whether equality should be a societal goal, and whether the pursuit of equality would lead to fewer problems in this country.1 One of the virtues of this scale is that it is not designed to capture egalitarianism on any specific dimension, such as gender, sexual orientation, class, race, or ethnicity. Instead, it refers to equality more broadly, and hence ostensibly measures egalitarianism as an abstract value. The Cronbach’s alpha on this six-item scale approaches or exceeds conventional thresholds for coherent attitudinal scales. For example, the alpha on the egalitarianism scale was .78 in the 2012 ANES survey, which had 5,914 respondents (2,054 in the face-to-face component of the study and 3,860 in the Internet component). This statistic varied considerably across the two survey modes, with an alpha of .82 in the Internet survey but an alpha of only .68 in the face-to-face survey.2 It is not clear why this discrepancy appears but the answer may partly lie in the fact that the two samples are not equivalent, with the face-toface sample having a higher response rate (38% versus 2%) and fewer, but a more representative number of, college graduates.