Social diffusion of campaign effects: campaign-generated interpersonal communication as a polarizing mediator of anti-tobacco campaign effects.

Abstract

Traditionally, campaign effects have been modeled as a result of individuals' direct exposure to campaign messages. A person's indirect exposure to messages through interpersonal conversation about a campaign, however, can be another way that campaign effects occur. Based on the notion of social diffusion of campaign effects, this dissertation research examined the role of campaign conversation in the generation and diffusion of campaign effects. More specifically, this dissertation research tested whether anti-smoking campaign-generated conversation affected individuals’ smoking-related perceptions using the TruthSM campaign data (Study 1) and experiment data (Study 2). Study 1 tested whether campaign conversation, along with individual exposure, can mediate the impact of the TruthSM campaign using secondary analyses of the Legacy Media Tracking Survey (LMTS) data. LMTS is a random digit dialing telephone survey of a nationally representative sample of youth (LMTS II: n =10,357; LMTS III: n = 10,079). Multilevel modeling results of LMTS II data show that both encoded exposure and campaign conversation mediated the

relationship between TruthSM campaign availability and smoking beliefs, although the mediating roles of encoded exposure and campaign conversation were not found in LMTS III. The results of LMTS II data also show that campaign conversation widened the difference in smoking beliefs between non-smokers and hardcore smokers, indicating a possibility that campaign conversation can function as a polarizing mediator. Based on the concern of potential backlash effects of conversation among smokers, Study 2 tested whether a smoker’s campaign conversation with another smoker can produce elevated pro-smoking normative perceptions and behavioral intentions using a laboratory experiment (n =72). Results show that smokers’ conversation generated boomerang effects. Although campaign conversation did not affect participants’ injunctive norms, personal descriptive norms, or behavioral intention, it led to elevated pro-smoking societal descriptive norms (i.e., higher smoking prevalence estimates) among participants who had a low number of smoker friends. Altogether, this dissertation research provides general support for the social diffusion process of campaign effects and the important role of campaign conversation in this process. It shows that campaign-generated conversation can generate, amplify, and, sometimes, reduce campaign effects. Theoretical and practical implications for campaign planning and evaluation are discussed.