The proposition that consistency among people's knowledge (K), attitudes (A), and behaviors (B) is contingent on high levels of involvement was tested in a long-term health intervention campaign. Support varied depending upon the indicator of involvement used. Cognitive involvement with the health topic, as indicated by number of responses to an open-ended question about heart disease, produced the clearest support. Using LISREL procedures, high K-A, K-B, and A-B correlations were found only in the high cognitive response group. Affective indicators of involvement produced mixed support. Perceived risk of heart disease was related to K-A-B consistency in a direction opposite to the hypothesis, but extremity of affective response was associated with the predicted high K-B correlation. A behavioral indicator of involvement based on reading the campaign literature produced results consistent with the hypothesis, although less so as the campaign progressed. The experimental health education campaign produced no discernible effect on K-A-B consistency, despite gains in knowledge itself.

Conflicting results in the involvement literature arise from differences in the definitions and operationalizations applied to the term. This study argues that the construct should be treated as encompassing a number of related concepts and should be broken into its component forms. Two of these components, attention and perceptions of message relevance, were manipulated in an experiment designed to assess the impact of a message on heart disease risk reduction. Results indicate that attention to the message had primarily cognitive effects, leading to greater message recall, whereas perceptions of relevance influenced attitudes and behavior.

This study examines the process through which a fear appeal may transform low-involvement audiences into active publics. Cognitive and emotional responses of uninvolved viewers to a film on environmental Contamination are analyzed, together with the coping strategies used to deal with the threat. The research integrates Grunig 's situational theo y ofpublics with Rogers' protection-motivation theo y to expand the predictive ability of the situational theory. The data indicate that post-test public membenh@, cognitive activity and emotional arousal during viewing are sign ficantly related to viewers' preferences for coping through message resistance, seeking additional information, or taking action on the film's topic. Results suggest that both cognition and affect mediate viewers' responses to a fearful message, thereby contributing to the creation of active publics who arepersuaded to take action on aproblem.The family of constructs that indicate heightened audience engagement with media content has repeatedly been shown to predict more powerful and enduring exposure effects. This family encompasses audience activity, interest, personal relevance, problem recognition, and involvement, a5 well as a number of other related concepts.Of all these constructs, involvement has received the most attention. High involvement has been found to lead to deeper message processing and more enduring attitude change (Petty & Cacioppo, 1986); a more "rational" hierarchy of effects (Ray et al., 1973); higher knowledge-attitude-behavior consistency

The authors examined comparative effects of jive health campaign channels used in the Stanford Five-City Multifactor Risk Reduction Project. Three distinct criterion variables were evaluated: (a) reach, measured as the number of messages intervention community residents remembered; (b) specificity, assessed by examining whether the campaign dzjkrentially reached those w/w were already knowledgeable and practicing cardiovascular disease risk reduction; and (c) impact, defined as the amount of knowledge gained during the 5-year campaign. Path analyses revealed that reach was highest for tip sheets. Spec$city was highestfor booklets and then television programs. Newspaper messages appeared to have the most impact, followed by booklets and television public seruice announcements, tip sheets, and finally, television programs. Channels varied according to reach, specificity, and impact, andeachof these criteria were distinct. No channel was optimal for all three of the outcome measures, suggesting that channel selection involves tradeofi among dz&%ent types of effects. ealth communication campaigns that use multiple channels to deliver messages are more effective in changing behavior than H those that rely on a single modality ( B l o b et al.

No abstract