ABSTRACT
In recent years, journalists, political elites, and the public have used Twitter as an indicator of political trends. Given this usage, what effect do campaign activities have on Twitter discourse? What effect does that discourse have on electoral outcomes? We posit that Twitter can be understood as a tool for and an object of political communication, especially during elections. This study positions Twitter volume as an outcome of other electoral antecedents and then assesses its relevance in election campaigns. Using a data set of more than 3 million tweets about 2014 U.S. Senate candidates from three distinct groups—news media, political actors, and the public—we find that competitiveness and money spent in the race were the main predictors of volume of Twitter discourse, and the impact of competitiveness of the race was stronger for tweets coming from the media when compared to the other groups. Twitter volume did not predict vote share for any of the 35 races studied. Our findings suggest that Twitter is better understood as a tool for political communication, and its usage may be predicted by money spent and race characteristics. As an object, Twitter use has limited power to predict electoral outcomes.
Notes
1. We acknowledge that this percentage is likely to be lower during a mid-term Senate election, as is the setting for this study.
2. They also both make criticisms of previous work, as well as suggestions for future studies, which we address later in this section.
3. This is more than 3.6 times the amount raised per candidate for seats in the U.S. House of Representatives. Furthermore, donations from individuals to Senate candidates outstripped those to House candidates by a factor of 4.7. These figures are presented as a straightforward way to compare nationwide interest in Senate and House races.
4. The data set includes one independent candidate, Greg Orman, who was the primary contender against the incumbent Republican Pat Roberts in Kansas.
5. The authors are currently underway on an update of this study, Normalizing 2.0. In service of this, the authors repeated the collection of the top-500 most influential U.S. journalists on Twitter, as calculated by Muckrack, for the year 2014. It is from this 2014 list that we add to our media list.
6. Before performing statistical analyses, assumption tests detected skewness and kurtosis slightly beyond acceptable scores. To address this issue, log transformations were performed on the dependent variable following the recommendation of Tabachnick and Fidell (Citation2007).
7. For each race, a candidate’s vote share is exactly 100% minus his/her opponent’s share. Because all the races in this sample were between a Republican and a Democratic candidate, we addressed this issue by splitting the sample into the two groups and conducting separate regressions for each of them.
8. VIF tests for media and public volume as independent variables on the political elites’ model detected multicollinearity, and only the coefficients for public volume are reported in the table (media volume dropped from analysis).
Additional information
Notes on contributors
Shannon C. McGregor
Shannon C. McGregor, MA is a PhD candidate in the School of Journalism at the University of Texas at Austin, and she will start as an assistant professor in the Department of Communication at the University of Utah in fall 2017. Her research interests are political communication, social media, gender, and public opinion.
Rachel R. Mourão
Rachel R. Mourão, PhD is an assistant professor in the School of Journalism at Michigan State University. Her research interests are journalism studies, political communication, new media, and Latin American studies.
Logan Molyneux
Logan Molyneux, PhD is an assistant professor of journalism at Temple University. His research focuses on journalism, social media, and technology.