Polarization in the Media During an Election Campaign: A Dynamic Network Model Predicting Support and Attack Among Political Actors

In multiparty election campaigns, many political parties and candidates compete for media attention, voters, and a government majority. Negative campaigning, which is often newsworthy, is an attractive strategy in the competition for media attention. However, political support for another party offers an alternative strategy because it signals preferred government coalitions. The research question to be addressed here is: What is prompting a political party to either support or attack another party in a news medium on a specific day in an election campaign? We analyze statements of support and attack among politicians and political parties reported in the news media during the 2006 Dutch national election campaign. Combining hypotheses from research in negative campaigning and the mediatization of politics with concepts from social network analysis, we predict whether a reported statement expresses support or an attack among political actors. Party size, party ideology, and incumbency of the political actors are important static predictors. Dynamic predictors, which indicate how the campaign has evolved thus far, include agreement or disagreement on issues recently raised in the media as well as recent attacks and support statements. The dynamic predictors are needed to explain polarization among political actors in the media during the campaign. The results show that reported statements affect the course of the election campaign.