Facebook Politics: Toward a Process Model for Achieving Political Source Credibility Through Social Media

Abstract

The last two election cycles have seen an exponential rise in the number of political campaigns integrating some form of social media into their communication plans. As candidates and campaigns increase their social media communications, political communication scholars have become increasingly interested in the process through which voters assess political candidates’ credibility through social networking sites. Using experimental data, this study examines the mediating role of attitude homophily in establishing political candidates’ source credibility among Facebook users. A multiple mediation model outlines a process wherein attitude homophily mediates the relationship between political cues and evaluations of source credibility. Theoretical and practical implications of the results for political social media campaigns are discussed.

KEYWORDS:

• Multiple mediation
• political campaigns
• social media
• source credibility

Notes

1. Although the sample included all students, there were some graduate students in the sample (16%) as well as freshman who had not completed their first semester of college and therefore answered this question with “high school graduate” (8%). Dropping these students did not substantively change any results reported hereafter. As such, they were left in the final models.

2. Competence: intelligent/unintelligent, untrained/trained, inexpert/expert, informed/uninformed, competent/incompetent, stupid/bright Goodwill: cares/does not care about me, has/does not have my interests at heart, self-centered/not self-centered, concerned/unconcerned with me, insensitive/sensitive, understanding/not understanding Character: honest/dishonest, untrustworthy/trustworthy, honorable/dishonorable, moral/immoral, unethical/ethical, phony/genuine

3. All OLS models were run with and without demographic and political controls. Strictly speaking, random assignment to condition should eliminate the need for these controls. The results reported here do not change with the inclusion of demographic and political controls in the model. As such, the more parsimonious results were reported here.

4. These numbers represent the upper and lower confidence intervals.

Additional information

Notes on contributors

Elizabeth E. Housholder

Elizabeth E. Housholder is a PhD candidate in the School of Journalism and Mass Communication at the University of Minnesota. Her research interests center around the intersection of political communication, new media, and strategic communication.

Heather L. LaMarre

Heather L. LaMarre is an assistant professor in the School of Media and Communication at Temple University. Her research interests include political communication, public opinion, and the psychology of strategic policy narratives.