A Multilevel Analysis of Antimarijuana Public Service Announcement Effectiveness

Abstract

In public-health campaign research, 3 prominent theories of persuasion and media effects—elaboration likelihood model (ELM), activation model of information exposure (AMIE), and limited capacity model of motivated mediated message processing (LC4MP)—have been used to predict message effectiveness. Although conceptually overlapping, these theories suggest contradictory predictions about individual-level and message-level factors on persuasion outcomes. In this study, we contrast and test competing predictions of antidrug message effectiveness from 3 recent publications that draw on ELM, AMIE, and LC4MP. We use televised antimarijuana messages, young-adult samples, and a multilevel modeling approach. Significant interactions between individual- and message-level factors were found predicting message effectiveness as theory dictates; these results replicate some, but not all of the findings from the aforementioned publications.

Keywords:

• Elaboration Likelihood Model
• Persuasion
• Health Communication
• Multilevel Analysis
• Antidrug Public Service Announcements

Acknowledgments

We thank Joseph N. Cappella (Annenberg School for Communication, University of Pennsylvania) for generously providing the original PSAs, extracted arguments, and content analytical data for this study. We also thank Chadd Funk for his help with our data collection.

Notes

1. We thank Joseph N. Cappella for generously providing the original PSAs, extracted arguments, and content analytical data for this study.

2. Due to space limitations, we have omitted nonsignificant results for lower-order interactions that were not addressed in this study's hypotheses.

3. Due to the limited space we do not provide another table with only slightly different numbers. The result table for I-squared instead of MSV is available upon request from the paper's first author.

4. This calculation compares the top third of SS scores with the bottom third; comparing the upper half to the lower half would lead to less of a difference.

Additional information

Notes on contributors

René Weber

René Weber (PhD, MD) is Professor of the Department of Communication at the University of California Santa Barbara

Amber Westcott-Baker

Amber Westcott-Baker is an Assistant Professor in the Department of Communication at the University of Maryland

Grace Anderson

Grace Anderson is an Assistant Professor in the Department of Communication Studies at Samford University