Gender differences in political attitudes among whites arise from a variety of sources that may vary from issue to issue. Explanations based on gender-based social roles, basic value differences, socioeconomic status, and women's autonomy are tested in this study through an examination of both compositional and conditional effects. Compositional effects occur when men and women differ on an explanatory variable. Conditional effects occur when a variable has differential effects on the policy preferences of women and men. Using data from the 1996 National Election Study, OLS regression and logit results demonstrate the complex sources of gender gaps across issue areas. Some factors such as education have more of a liberalizing effect on women, while such factors as religiosity have more of a conservatizing effect on men. Overall, issue gender gaps arise both from women's cultural role and from women's increasing autonomy from men.Nearly 20 years after feminists and political observers first called attention to the "gender gap" in voting behavior, the existence and significance of gender differences in political attitudes and behavior remain controversial. Most acknowledge, however, that women today are somewhat more likely than men to identify with the Democratic party and to vote Democratic. There is, in addition, widespread agreement that issue attitudes are among the important variables explaining gender differences in voting behavior, especially attitudes on "compassion" issues like social welfare and redistribution and issues involving the use of force such as capital punishment and military intervention (Kaufmann and Petrocik 1999;Mueller 1991;Seltzer, Newman, and Leighton 1997).But what is the source of the gender differences on these and other issues? The most provocative question about the gender gap is: What aspects of the socialization of women and their role in society lie behind these issue differences? This is the subject of this analysis. We attempt to explain the gender gap on some of the issue areas that underlie the well established gender differences in partisan and electoral choice. These issues include social welfare, the role of government, use of force, feminism, racial attitudes, and liberalism0conservatism.We would like to thank Rosalind Blanco Cook, William McLean, Matthew Vile, Dennis Gleiber, and Virginia Haysley for their invaluable research assistance, and two anonymous reviewers for their helpful comments.

This research examines changes in the environmental coalition over the 1980s. During these years, concern over environmental problems has increased. How has the coalition favoring greater environmental protection changed? We use logistic regression to examine changing determinants of support for increased environmental spending. In brief, ideology, party identification, and age become less important as determinants, and education and urban residence become more important. The implications are discussed.

Much work has been done on the correlates of confidence in the United States Supreme Court. However, very little research has been undertaken to discern the correlates of confidence in state and local courts. Using survey data from Louisiana, we examine confidence in state and local courts. We focus on the role of experience, arguing that the opportunity for wide participation in these courts makes the confidence calculation different from that of a remote institution like the US Supreme Court. We find that, indeed, experience matters and further, that type of experience matters. Those with more stake in the outcome of the court case and less control over it (e.g., defendants) are least confident in state and local courts, while those with little stake and substantial control (e.g., jurors) are most confident in them. Procedural justice concerns also loom large in the confidence calculation for these lower courts. Timeliness, courtesy, and equal treatment all affect public confidence.

No abstract

A large body of research on the impact of economic conditions on elections has been amassed. Past research emphasized congressional and presidential elections. The present research extended the question of economic effects to the state level. Are governors held responsible for the economic conditions of their states? The existing studies at the state level all used aggregate data and concluded that states' economies do not affect the fortunes of the governors. The studies inferred that the national economy, not the state economy, affects governors. The present research was a direct test of the aggregate studies using micro-level data from statewide surveys. Strong state economic effects on gubernatorial approval were found in a multivariate model. Furthermore, evaluations of the national economy were almost insignificant in predicting gubernatorial approval.Over the past fifteen years, a large body of research on the impact of economic conditions on elections has been amassed. It is now well established that the electorate responds to economic downturns by punishing the incumbent and the incumbent's party. The earliest studies used aggregate data (Kramer 1971;Stigler 1973;Arcelus and Meltzer 1975) and focused on presidential and congressional elections. In later surveys, researchers began investigating the economic effects with micro-level data (Weatherford 1978;Fiorina 1978;Kinder and Kiewiet 1979;Kiewiet 1981).