Public opinion surveys since 1965 find that black respondents are less in favor of legal abortion than white respondents. Using the 1982 NORC General Social Survey, we replicate and expand one of the few studies (Combs and Welch, 1982) that examined the structure and determinants of prochoice attitudes of blacks and whites. Our major findings are (1) the racial difference in prochoice attitude is as great in 1982 as in the 1970s, (2) contrary to the suggestion of Combs and Welch, the demographic and attitudinal determinants of abortion attitudes differ for blacks and for whites, and (3) for those respondents who differentiate their acceptance of legal abortion, the pattern of prochoice attitudes also differs by race. Elaine J. Hall is a doctoral student in the Department of Sociology, University of Connecticut, Storrs, and Myra Marx Ferree is an Associate Professor of Sociology in that department. The authors wish to thank Mark Abrahamson, Nancy Andes, and the anonymous reviewers for this journal for their constructive criticism of an earlier draft. The article was originally presented as a paper at the Eastern Sociological Society Annual Meetings, March 1985. Public Opinion Quarterly Vol. 50:193-207 ? 1986 by the American Association for Public Opinion Research Published by The University of Chicago Press 0033-362X/86/0050-193/$2.50 This content downloaded from 157.55.39.128 on Mon, 18 Jul 2016 05:14:36 UTC All use subject to http://about.jstor.org/terms 194 ELAINE J. HALL AND MYRA MARX FERREE fundamentalism), and the respondent's self-identified strength of affiliation, and (3) southern culture, measured by present region of residence and at age 16, and by present size of community and at age 16. Each of these variables has been shown to have a negative zeroorder correlation with abortion support. Each can also be assumed to be more prevalent among blacks than whites. Therefore, Combs and Welch predict that controlling for these three antecedents would reduce or eliminate the racial difference in abortion attitudes. In fact, Combs and Welch find that controlling for demographic variables (SES, southern culture, age and sex) reduces the difference between blacks and whites by only about one-third. When religious practice variables are also controlled, the black-white difference declines further. However, even when controlling for all 11 indicators, a small but statistically significant racial difference in abortion attitudes re-