Mass media campaigns and conversation: testing short-term and long-term priming effects of topic-related conversation on conversational participants.

University of Minnesota Ph.D. dissertation. August 2009. Major: Mass Communication. Advisor: Brian G. Southwell. 1 computer file (PDF); xi, 230, appendices A-I.This dissertation investigates how participating in a topic-related conversation (i.e., a conversation focused on a specific event-related topic) may create short-term and long-term priming effects. In the case of topic-related conversations occurring prior to message exposure, it was expected that those conversations would function as a priming task and influence how subsequent mass media messages were evaluated. In the case of long-term engagement in topic-related conversation, it was expected that frequency of topic-related conversations and topic-related media use would predict current and future topic-related behavior. The dissertation presents the results of three studies that test hypotheses suggested by this line of reasoning. In Study 1, the basic hypothesis was that topic-related conversation would act as a priming task and influence how ensuing mass media messages would be evaluated. Consistent with what was predicted, prior topic-related conversation did influence how subsequent stimuli (i.e., two anti-binge drinking public service announcements) were evaluated. Study 2 was informed by the active-self account of prime-to-behavior effects (Wheeler, DeMarree & Petty, 2007, 2008). The study tested multiple hypotheses, including whether self-monitoring (Snyder, 1974) moderated the topic-related conversation--message evaluation relation and whether a measure of the active self-concept mediated the topic-related conversation--message evaluation relation. Consistent with what was predicted, self-monitoring did moderate the topic-related conversation--target stimulus evaluation relation. However, the hypothesis that scores from a measure of the active self-concept would mediate the topic-related conversation--target stimulus evaluation relation was not supported. Study 3 examined some longer-term effects of topic-related conversation. Influenced by prior research indicating that media use and conversation about the news can predict political participation (e.g., McLeod, Schuefele & Moy, 1999), the hypotheses for this study proposed that topic-related conversation about exercise and exercise-related media use would predict concurrent and future exercise-related behavior. Analyses indicated that at Time 1 increases in overall television use were associated with decreases in exercise, while increases in exercise-related conversation were associated with increases in exercise. A second analysis indicated that higher levels of sports television viewing and exercise-related conversation at Time 1 were associated with higher levels of exercise behavior at Time 2