Answers are grounded in this paper and cite their sources.
Using large-sample data from the 1930s, this article examines the relationship between state and state policy in three issue areas. State is found to correlate strongly with state policy at the time of the survey and (in some cases) subsequent policy changes as well. These correlations survive standard controls for region and state socioeconomic conditions. Although other interpretations are also possible, a reasonable inference is that-at least on certain issues-public can exert a strong influence on state policy decisions. To what extent do policy differences between the American states reflect differences in the policy preferences of the different state publics? Despite the obvious relevance of this question, public opinion is largely a forgotten variable in contemporary research on state policy. Very little is known even about how strongly state policies on given issues correlate with state on those issues, let alone how much (if at all) actually influences state policy. An obvious reason for this gap in our knowledge is that we lack reliable survey data on state-to-state variation in opinion. Some researchers have attempted to close this gap by indirectly estimating (or simulating) the preferences of state electorates from the known demographic correlates of and the known demographic characteristics of state populations (Weber and Shaffer, 1972; Weber, Hopkins Mezer, and Munger, 1973). Although this research strategy holds some promise of success, the simulated opinions of artificial state electorates comprise a secondbest substitute for the real thing. This paper reports the relationship between state policy and on three issues for which directly measured state-by-state distributions are available. The data are not from a recent survey, but rather from early Gallup Polls in the thirties. That the results of some early Gallup Polls allow a seemingly reliable breakdown of at the state level rests with Gallup's early sampling technique. Survey researchers interview approximately 1,500 respondents for a national survey today, whereas in its earliest days the Gallup organization would typically obtain opinions from
No takes yet. Share an insight, caveat, or question.
Robert S. Erikson (1976) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: