self-outgroup differences in influence) added significantly to the prediction of 9 out of 10 behavioral measures, over and above the contributions of the experimental manipulations and perceptions of outgroup influence. Mediational analyses further indicated that, in the prediction of behavior (e.g., intention to censor the article), third-person perceptions fully mediated the effect of ingroup norm, and partially mediated the effect of article.
The results of this study provide important evidence not only for the behavioral component of third-person perceptions but also for the social-psychological basis of the effect. As noted earlier, evidence for the behavioral component of the third-person effect is limited by comparison to evidence for the perceptual component of the effect. Hence, this experimental test of the behavioral component provided additional support for Davison ( 1983) original formulation. Moreover, this research constitutes the first attempt to demonstrate the important mediational role of third-person perceptions in behavioral responses to persuasive messages (e.g., media censorship). Given a number of limitations with this experimental study, including the absence of significant effects involving the manipulation of social orientation, further work of this nature is required.
Our results highlight the need to re-emphasize the role of social groups, not only in understanding persuasion per se (see chapters in this volume by Mackie & Queller; van Knippenberg; and Fleming & Petty) but also in understanding perceptions of the persuasive impact of messages on self and other. Our research findings suggest that perceptions of persuasive impact on self and other are sensitive to the social psychological relationship between self and other in ways that are consistent with the theoretical assumptions of SIT. That is, third-perceptions are influenced by the subjectively salient social relationship between self and other and are governed by motivational needs, such as self-esteem, social identity, and differentiation from others (cf. Brewer, 1991; Hogg & Abrams, 1993). In this sense, third-person perceptions often represent not simply a self-serving perceived self-other difference in persuasive impact but also a group-serving us-them distinction.
It is important to note, however, that even when social identity is salient, respondents often perceive themselves as uniquely invulnerable to influence--that is, they perceive themselves as positively distinct from, rather than interchangeable with, ingroup members. Intergroup contrast in perceived persuasive impact is more readily observed than is self-ingroup assimilation. Perhaps as Simon ( 1993) suggested, "the cognitive differentiation between oneself and other people is much clearer than the