INVOLVEMENT AND PERSUASION - TYPES, TRADITIONS, AND THE EVIDENCE

This article is a reply to Petty and Cacioppo's (1990) critique of our meta-analysis in which we concluded that research has established 3 different types of involvement with distinctly different effects on persuasion (Johnson & Eagly, 1989). We first correct their summary of our review. In response to their claim that outcome-relevant and value-relevant involvement are best reduced to a single construct, we assert that this proposal fails to account for existing research findings and provides only a highly speculative account of the processes that might mediate the impact of involvement on persuasion. We then reaffirm our earlier conclusion that the effects of outcome-relevant involvement are especially unstable when messages contain weak arguments. In fact, this conclusion is underscored by 4 studies that Johnson conducted after completing the meta-analysis. Finally, we explain how the methodological features of our review that Petty and Cacioppo fault are consistent with established principles of meta-analytic reviewing.