Evaluating expert judgments

Abstract.  Although expert surveys have gained a prominent place in comparative studies of party positions on issues, their validity has been called into question. In this article, some of the validity concerns are evaluated in the context of the authors' own expert survey on national party positions vis-à-vis European integration. One goal of the article is to demonstrate that this expert survey produces valid measures of party positions. An equally important goal, however, is to suggest some methods that can help in assessing the quality of expert survey data. These methods, which are rooted in psychometric theory, are applicable in a variety of contexts and are easily implemented.     

Political judgments,perceptions of facts,and partisan effects

We test two competing hypotheses about the impact of partisanship and information on people's political judgments and perceptions of facts using Canadians' reactions to a major scandal. Our findings with respect to subjective political judgments confirm the argument that partisan predispositions are crucial. But there is no evidence to support the argument that the polarizing effect of partisanship is most evident among the most informed. When it comes to perceptions of “objective” facts, the results are consistent with Zaller's reception axiom: the more informed people are, the more likely they are to correctly perceive objective facts. Partisanship does not appear to affect these perceptions.     

Bases of political judgments: The role of stereotypic and nonstereotypic information

Three experiments investigated the role of stereotypic and nonstereotypic criteria in judgments of political candidates. The effects of physical attractiveness, political party, and stands on specific issues on both absolute and comparative judgments of political candidates were examined to evaluate three hypotheses about stereotype and attribute use. In the absence of other information, candidates' physical attractiveness (conveyed through photographs) had a substantial influence on subjects' global evaluations of them and inferences of both their personal qualities and their political ideology. When other information about the candidates' party membership and stands on specific issues was available, however, the candidates' attractiveness had no effect on the evaluations of them. When subjects made judgments of only one candidate, subjects relied exclusively on the candidate's voting record. When subjects were asked to make comparative judgments of two candidates, however, they based their judgments on each candidate's party membership and not on their respective voting records. Implications of these results for the processes that underlie political judgments and decisions are evaluated.     

Collective and individual economic judgments and voting in West Germany, 1961–1984

Abstract: : This article attempts to provide individual-level support to findings of longitudinal macro-level studies of the relation between economic variables and electoral outcomes in the Federal Republic. The theoretical framework is one of incumbency-oriented vs. policy-oriented political responses to collective vs. individual economic judgments (Kiewiet, 1983). The data base consists of 25 surveys from 1961 through 1984. Empirical analysis comprises three steps: First, the relation between economic judgments and partisan sympathies is analyzed, since it appears likely that part of the covariations between voting and such judgments is spurious due to a common dependence upon partisan orientations. Second, the impact of both kinds of economic judgments upon voting over time is investigated with and without controls for partisan affiliation. Finally, the impact over time of macropolitical and macroeconomic variables on individual political response patterns to economic judgments is estimated at the longitudinal macro-level. The analysis demonstrates that collective economic judgments are generally a more potent predictor of electoral choice than individual economic judgments, and that there is clear empirical evidence both for the incumbency-oriented and the policy-oriented hypothesis linking such judgments to the vote.     

Determinants of political judgments: The joint influence of normative and heuristic rules of inference

Prior research on political judgment has been polarized, with one group emphasizing normative models of political inference and the other emphasizing heuristic models of political inference. In accordance with recent findings in psychology, this article is rooted in the assumption that both normative and heuristic criteria simultaneously influence political judgments. Furthermore, differences in the relative extent to which these two processes serve to determine political judgments are hypothesized to depend on the nature of the judgment task. Two kinds of political judgments are considered: judgments reflecting beliefs about the candidates' stands on the issues and judgments reflecting global evaluations of the candidates. The reported results confirm that (1) voters rely on both normative- and heuristic-based considerations when forming these judgments, and (2) the relative extent to which these criteria influence these judgments depends on the complexity of the judgment task.     

Guilty by disassociation (and innocent by association): The impact of relevant and irrelevant analogies on political judgments

We investigate how associative (X is similar to Y) and disassociative (X is different from Y) analogies may affect judgments of political target stimuli they are employed to embellish. In an experimental study it was hypothesized that the judgmental relevance of the analogy-target relation is an important determinant of the effect of the analogy on subsequent judgments. It was reasoned that for irrelevant analogies the (dis)associative relation would be less efficiently encoded and utilized than the evaluative tone of the analogy itself, resulting in relative assimilation to the connotation of irrelevant analogies, even where the analogy was actually disassociated from the target. Where the analogy-target relation was relevant to judgments of the target, these judgments were predicted to follow the specified analogy-target relation for both associative and disassociative analogies. The predictions were supported. The relation of these results to social judgment models and to the rhetorical devices used to influence political judgment is discussed.