Master of Science / Department of Journalism and Mass Communications / Joye C. Gordon / This experimental research adopts a typology of frames by Levin, Gaeth, and Schneider (1998) and seeks to a) determine what combination of attribute and goal frames produces the strongest effect on vaccination behavior; b) ascertain to what extent personal relevance of vaccination moderates this framing effect; and c) explore how individual pre-existing characteristics, such as recent vaccination history, vaccine risk perception, vaccine dread, and general attitude toward vaccination influence the persuasive power of framed messages. The study, designed as field experiment 2 (+/- attribute frame) x 2 (+/- goal frame) x 2(involvement), recruited 476 adult female participants that were exposed online to four experimental framing manipulations and a control condition. The main effect is consistent with the typology of frames — the combination of the positive attribute and the negative goal frame was the only condition that was significantly more persuasive than the control condition. Participants who had children or were pregnant, for whom vaccination was more relevant and meaningful, have not reacted to message framing differently. However, general pre-existing attitudes towards vaccines, perception of vaccine safety, perception of vaccine efficacy, vaccine dread, and vicarious experience with vaccine side effects, appear to be associated with antecedents of vaccination behavior. Overall, this study has focused on ecological validity,aiming at the applicability of framing theory in the context of health communication.