On-line processing in candidate evaluation: The effects of issue order,issue importance,and sophistication

This article reports the results of a study that replicates and extends the impression-driven model of candidate evaluation reported in Lodge, McGraw, and Stroh (1989). This model holds that evaluations are formed and updated on-line as information is encountered, and that as a result, citizens need not rely on specific information available from memory to form their candidate evaluations. In the present work we explore whether the order in which information is encountered, as well as whether information that is personally important, influences the weight accorded to evidence in on-line processing. In addition, differences in information-processing strategies due to political sophistication are examined. The results indicate that important information receives more weight than unimportant information. In addition, the evidence suggests that political sophisticates are more efficient on-line processors than are less sophisticated individuals. The implications of these results for models of candidate evaluation are discussed.     

An examination of how voters form impressions of candidates' issue positions during the nomination campaign

There is increasing evidence that issues influence voter preferences during the nomination campaign (Bartels, 1985; Bartels, 1988); however, only Bartels (1988) and Conover and Feldman (1986, 1989) have examined how partisans forge perceptions of candidates' positions on issues prior to the general election campaign. The goal of this paper, then, is to examine how individuals develop perceptions of candidates' issue positions during the crucial months leading to the nominating conventions. Relying on theories developed in social-psychology, I tested five competing hypotheses known to influence individuals' perceptions of candidates' issue positions. An examination of the findings revealed that there is strong support for one of the hypotheses and modest support for three additional hypotheses. In summary, it appears that voters are quite ingenious in forming impressions of where candidates stand on the issues. They rely on stored information about politics, they actually adjust candidates' true positions to relieve cognitive inconsistencies, they evoke their own issue positions to assume candidates they like agree with them and candidates they dislike disagree with them, and finally they evoke their own issue positions to assume candidates agree with them even when they hold no sentiment toward the candidate.     

Privatization,bureaucracy, and risk aversion

The role of governmental risk aversion in the decision to privatize the production of goods and services has not been examined closely. Using a model of a risk-averse, single-service Niskanen bureaucrat, we determine the conditions under which a bureaucrat will prefer to privatize rather than produce in-house. If the private-sector firm is risk neutral, the result will be a fixed-fee contract with complete insurance. If the private-sector firm is risk averse, the result will be a cost-plus contract with the degree of cost sharing determined by the bureaucrat's share of total risk aversion. In both cases, the bureaucrat's sponsor may affect the likelihood of privatization by manipulating the rewards and penalties imposed on the bureaucrat.     

Rational Choice and Community Power Structures1

The community power debate concluded with each side believing they had won. Political theorists have generalized power, making empirical investigation very difficult; urban scholars have turned their attention to more manageable empirical problems. Rational choice advances the debate, exposing the errors of all sides and facilitating a new approach which transcends structural versus individualist methods. By separating various aspects of power in urban contexts, complementary techniques such as network analysis in a bargaining framework, semi-structured interviewing and the use of text databases permits a comprehensive investigation of agenda-setting and the mobilization of bias. The paper demonstrates the utility of this approach by comparing it to 'regime theory', the latest paradigm of urban research.