Persuasive Communication of Risk Information: A Test of Cue Versus Message Processing Effects in a Field Experiment

Portions of the elaboration likelihood model of persuasion were tested in a natural selling with adults of various ages and educational backgrounds as subjects. The persuasive communication concerned risks of largl-scale use of coal for electricity production. The dependent variable was attitude change. Elaboration likelihood (EL) was operationalized through involvement and need for cognition. Highly credible sources were found to be more persuasive than less credible sources for low-EL but not for high-EL subjects. More persistence of attitude change (measured 14 months later) and stronger beliefs-attitude-intention consistency were found for high- than for low-EL subjects. Analyses of separate effects of involvement and need for cognition indicated that the two factors interacted concerning effects of source credibility. Need for cognition was related to attitude persistence. Involvement moderated beliefs-attitude-intention consistency, and need for cognition moderated attitude-intention consistency. Both factors were related to the route to persuasion.