Conservative Weather in a Liberalizing Climate: Change in Selected NORC General Social Survey Items, 1972-78

This paper traces the changes in 12 especially volatile items in the NORC General Social Survey 1972-78 (11 national priority questions and reports of changes in one's finances) and an abortion item that showed an unexpected plateau pattern. Despite sociological predictions of a trend toward liberalism stemming from demographic changes (the Stouffer hypothesis) the set, as a whole, showed a conservative direction consistent with the claims of recent pop sociology. The paradox is resolved by a multivariate causal model that shows both patterns of change to be operating. The metaphor of a slow, long-tern trend toward liberalism in the opinion climate plus a sharp, shortterm shift toward conservativism in the opinion weather is introduced to interpret the results. The striking absence of interactions in the data casts doubt on the hypothesis that the young or the better educated tend to lead other groups as opinions shift. In the months of February and March of 1972 through 1978, the National Opinion Research Center (NORC), with heroic inhibition of its urge to improve their wording, asked identical questions of national samples of adult Americans in its General Social Survey (GSS). (For details, see Appendix.) This rigidity was, oddly enough, in the service of studying change. Because the questions are repeated, one may use the cumulative file (N = 10,652) to track trends for some 200 items, over six years within a variety of social groups, using repeated samples from the same statistical universe.