Measures of Trends in Racial Attitudes of White Americans Towards Black Americans, 1972-1980.

Abstract

This dissertation is an investigation of attitudes of White Americans towards Black Americans using General Social Survey data collected by the National Opinion Research Center from 1972 through 1978 and 1980. Five items are used to represent different types of attitudinal measures. The measures are subjected to log linear multivariate contingency tables analysis. Hierarchical modeling techniques are used to determine "best fit" models. Logit analysis is used to determine the maximum likelihood statistics for liberal responses, given the full predictive model. The analyses show two trends emerging from the data. One is a trend of liberal change among items based on the desirability of racially mixed social contact. The literature predicts and these results confirm that through 1980, White Americans are becoming more liberal regarding issues of interracial contact. A second trend, supporting the theoretical model of rational choice/self interest, is one characterized by conservative change. Measures used ask the respondents' willingness to condone government spending for programs to alleviate conditions of inequality. This second trend suggests that White Americans are becoming less liberal on issues of public policy. Given the apparent discrepancy between these two trends, the issue of question validity is raised. It is the contention of this dissertation that attitudes exist at more than one level. The extant literature has typically examined only one level, that of interracial social contact, and concludes that conditions are improving. Equally valid are measures of racial attitudes which examine possible strategies to improve the plight of minorities and the economic implications for the promotion of racial equality. It is the conclusion of this dissertation that these latter items are more accurate measures of racial attitudes of White Americans in a society becoming generally more conservative. More meaningful measures should be utilized in future studies of racial attitudes of White Americans, measures which take into account respondents' willingness to accept the economic implications of equality.