Suchergebnisse

Analysts of social science data often need the flexibility & power of interval-level statistics, even though strictly interval measurement has not been achieved. In such cases, analysts should attempt to assign category values that minimize distortion of the underlying variables. Various methods for category value estimation are described, including estimation from the observed frequencies based on an assumption about the underlying distribution, estimation from single or multiple criteria variables, & estimation from item text. These techniques can be easily used, & their sensible application can avoid needless measurement error & resulting statistical bias. 1 Table, 4 Figures. HA.

1. R-2341/2. VIII,157 S.,Lit.,Tab.,Diagr. ISBN 0-8330-0163-9; 2. Technical Appendixes. R-2341/3- V,122 S.ISBN 0-8330-0205-8

This article contrasts the "self-interest" and "symbolic politics" explanations for the formation of mass policy preferences and voting behavior. Self-interested attitudes are defined as those supporting policies that would maximize benefits and minimize costs to the individual's private material well-being. The "symbolic politics" model emphasizes pressures to make adulthood attitudes consistent with the residues of preadult socialization. We compare the two models in terms of their ability to account for whites' opposition to busing school children for racial integration of the public schools, and the role of the busing issue in presidential voting decisions, using the 1972 Center for Political Studies election study. Regression analysis shows strong effects of symbolic attitudes (racial intolerance and political conservatism) on opposition to busing, and of the busing issue on presidential voting decisions. Self-interest (e.g., having children susceptible to busing) had no significant effect upon either. It is concluded that self-interest is often overestimated as a determinant of public opinion and voting behavior because it is too rarely directly assessed empirically.