The Influence of Communication Forms at Different Points in Time during the 2004 U.S. Presidential Campaign

by J. Brian Houston

Understanding the relationship between people’s use of communication forms (i.e. ads, debates, the Internet, political talk radio, etc.) and various political outcomes (i.e., perceptions of candidates, participatory outcomes, etc.) is increasingly difficult in today’s communication environment because of the sheer number of diverse media forms. Most individuals rely on multiple communication forms, from which they derive both information and entertainment. Thus, a person’s political knowledge, for example, is likely the product of their use of multiple communication forms. That is, what an individual knows about President Bush’s campaign positions can be the result of information an individual has attained from reading newspapers, surfing the Internet, watching not only one’s local television news regularly, but also Saturday Night Live and Late Night with David Letterman, as well as seeing numerous print and television candidate ads and each of the presidential debates, and then discussing those debates the following morning with co-workers.

The complexity of the modern media environment is why it is becoming increasingly important for researchers to look at political information disseminated from all media sources, that is, to look at the totality of political information. Research that focuses on a single communication form does not account for media influence as it exists in the real world, a world in which numerous communication forms likely have simultaneous influence.

For this reason, Michael Pfau [1] has pursued a program of research that examines the relative influence of all communication forms. The work supporting this program of research has spanned several Presidential elections, and the work discussed here is the 2004 installment of this program of research. In order to investigate the relationship between all communication forms and political outcomes, Pfau conducted two national surveys in the months of September and October of 2004. Working with Pfau on this project were University of Oklahoma doctoral students J. Brian Houston and Shane M. Semmler, and administering the surveys was Dr. Mary Outwater of the University of Oklahoma Public Opinion Learning Laboratory.

The first national survey for this project was conducted in September 2004 following the Republican National Convention but before the first presidential debate, and the second survey was conducted in the middle of October and spanned the last two weeks of the general election campaign. Two surveys were conducted so that the relative influence of all communication forms could be measured both early and late in the election cycle. Both surveys were identical, varying only in their date of delivery.

Seventeen different communication forms were included in this research, and level of use of each of these forms (along with various socio-demographic variables) served as the independent variables. Examples of the types of communication forms and examples measured include: political talk radio, television comedy (e.g., Saturday Night Live), network television news, movies/DVDs, World Wide Web, televised political debates, newspapers, and others. The research attempted to assess the relative influence of each communication form while simultaneously controlling for the influence of all others forms, and after extracting the influence of relevant socio-demographic variables.

Two groups of dependent variables were measured in this research: normative political outcomes and political candidate evaluations. The first groups of measures, normative political outcomes, included the following measures: attitude about the democratic process, political expertise, and likelihood of participating (i.e., voting, contributing time or money, etc.). A second group of measures focused on candidate evaluations and included: attitude toward candidate, intention to vote for candidate, candidate character evaluation, positive and negative feelings about candidate, and thoughts about candidate.

Preliminary analysis of the data collected has resulted in several interesting findings. A finding that spans all specific measures is the result that the influence of communication forms on all outcomes increases as the election progresses, so that, for example, the influence of communication forms on political outcomes during the second phase of this study—the last two weeks of the election—is greater than those reported during the first phase—in September.

This makes sense in that early on in the election cycle, communication use and its influence is less because many have not yet focused on the campaign. So, because less attention is being paid to the campaign, information disseminated by the various communication forms about the campaign exerts less impact on political outcomes early in the campaign compared to later.

The results indicated that communication use affected normative outcomes (i.e., political expertise, attitude about democratic process, and likelihood of participating). In the first phase of the study, conversations with others and the anticipation of debates were positively associated with both level of expertise and likelihood of participating in the democratic process. These are the only two communication forms that are associated with more than one of the normative outcome measures. In the second phase, conversations with others and debate viewing continued to be positively associated with political expertise and likelihood of participating. However, political talk radio use was positively associated with all three normative measures. A number of other communication forms exerted more modest influence on normative measures and variance accounted for by all communication forms was greater than at Phase 1.

The second aspect of this project examines the influence of communication forms on candidate evaluations. Candidate evaluations are broken into three groups of measures:
perceptions of candidates, feelings about candidate, and thoughts about a candidate.

Again, for these measures just as with the normative measures, political talk radio was the most influential of all communication forms. During both phases of the study, political talk radio was positively associated with perceptions, feelings, and thoughts about Bush and negatively associated with perceptions, feelings, and thoughts about Kerry. The only other communication form approaching political talk radio’s level of influence was movies/DVDs. Movie/DVD use was positively associated with perceptions, feelings, and thoughts about Kerry during both phases of the study, and was negatively associated with perceptions, feelings, and thoughts about Bush during the second phase of the survey. These findings, and others not reported here, suggested that in 2004 communication use had become increasingly partisan, with Democratic and Republican identifiers seeking out communication venues that mirrored their partisan dispositions.

Other communication forms also exerted influence on at least two dependent measures. During Phase 1, use of ads was associated with positive perceptions of Bush, whereas the anticipation of debates was related to negative perceptions. During Phase 2, radio news and debate viewing were associated with negative perceptions of Bush. During Phase 1, anticipation of debates was related to positive perceptions of Kerry, and during Phase 2, debate viewing and newspaper use was associated with positive perceptions of Kerry.

Ultimately, research that looks at communication forms in unison is able to discern the relative power of the individual communication form. Comparatively, research that examines communication forms in isolation is unable to understand the relative influence of a single communication form as it operates in the natural media environment of several different communication forms.

J Brian Houston is a Ph.D. student at the University of Oklahoma

Return to the Roundtable

1. Pfau, M., Cho, J., & Chong, K. (2001). Communication forms in U.S. presidential campaigns: Influences on candidate perceptions and the democratic process. Harvard Journal of Press/Politics, 6, 88-105. Pfau, M., Kendall, K.E., Reichert, T., Hellwig, S.A., Lee, W., Tusing, K.J., & Proise, T. (1997). Influence of communication during the distant phase of the 1996 Republican presidential primary campaign. Journal of Communication, 47, 6-26. [return]