Attitude Consistency

Intra-Attitudinal Consistency

An attitude is said to be consistent when all elements elicit similar evaluative judgments. Intra-attitudinal [Page 93]inconsistency arises when some of these elements are positive in nature, while others are negative. Within the cognitive component, accessible beliefs may conflict, cognitions and affects may also conflict, and the various affects may differ, simultaneously. Intra-attitude consistency can be examined by comparing how respondents feel, how they think, and what their summary evaluation is. The many studies that showed low correlations and consistency between the cognitive, affective, and conative (behavioral) components have raised doubts about the early theorists' multicomponent view of attitudes. Martin Fishbein and Icek Ajzen restricted the term attitude to a person's evaluation and drew a clear distinction between cognitions, attitudes, behavioral intentions, and behaviors. From their point of view, an attitude is a person's general feeling of favorableness or unfavorableness and an index of the degree to which a person likes or dislikes an object. This one-component view of attitude opened the way to the development of more sophisticated explanatory models in which the summary evaluation is at the core of an attitude, while behavior or the intention thereto is an effect, and cognition (knowledge and beliefs) and affect (general moods and specific emotions) are origins of the attitude. The affects underlying attitudes are more easily accessible in memory than the cognitions, and they tend to predominate among ambivalent respondents with different evaluative-cognitive and evaluative-affective consistencies. Individuals differ in their tendency to base their attitudes on cognition or affect, as Geoffrey Haddock and Mark Zanna have shown. Consistent attitudes were found to be stronger, more predictive of subsequent intentions and behavior, more resistant to persuasive communication and change, and thus more stable than inconsistent attitudes. Inconsistent attitudes seem to result in more information processing and knowledge acquisition.

Interattitudinal Consistency

Attitudes toward the same object and attitudes toward different objects are said to be consistent when they do not contradict each other. Interattitudinal consistency with respect to the same object (dual-attitude model) is empirically examined by the extent to which a respondent's explicit and implicit or habitual attitudes are related to one another. Interattitudinal consistency with respect to different objects is empirically examined by the extent to which these attitudes are related to one another and/or the level of correspondence between the respondent's attitudes and his or her ideological self-identification (e.g., consistent liberal or conservative attitudes), left–right self-placement, and respondent's value system and/or the number of issues on which a respondent adopts his or her (liberal or conservative) party's position. Interattitudinal consistency is considered desirable from the point of view of “correct” voting in representative democracies, requiring citizens to make well-founded voting decisions. Scholars vary in their views on the level of interattitudinal consistency. Philip Converse was one of the first who showed that the political attitudes that many people have are not consistent and referred to this aspect of public opinion as “nonattitudes.” However, the evidence presented by Max Kaase and Kenneth Newton, using Eurobarometer data, suggests that public opinion (in Western Europe) is relatively internally consistent (at both the aggregate and the individual levels). Mass opinion about the welfare state and the scope of government is structured into relatively few attitude publics. Mass attitudes about spending on government services and attitudes toward policy issues cluster into three groupings of related issues (welfare, security, and quality of life). There is also a match between saying that a particular issue has high priority as a public service and expressing a willingness to pay taxes for it. Finally, the statistical associations with left–right orientation and party identification are usually stronger than with any of the social, economic, and demographic variables. However, this left–right factor generally explains only a small proportion of the variance in issue positions. A factor that can also help us understand the structure of political attitudes is values; there is now a substantial amount of evidence that values are a major source of structure for political attitudes, according to Stanley Feldman.

...