Matching Theory and Data: Why Combining Media Content with Survey Data Matters

Understanding media effects is a core challenge for scholars and students of electoral behavior and campaigns. In an era characterized by volatile electorates, a rapidly changing media landscape and increased mediatization, it is of central importance to disentangle the role of the media in these developments. This challenge has obvious relevance for both science and society, and it is one reason that the field of political communication is burgeoning and prominent in journals in both communication science and political science. A key question is how the media affect political behavior. This is not a new question, but one that has received heightened attention in recent years. In our article ‘Who’s Afraid of Conflict?’ we test whether individual exposure to conflictual news coverage of European Parliament elections mobilizes citizens, contingent on general polity evaluations. This article is based on an original two-wave panel survey and a media content analysis of television news and newspapers, which enables us to test the dynamics of such questions. The content analysis was conducted within the framework of PIREDEU, where the data are also publicly available. Our article has prompted a response by Fazekas and Larsen (F&L). In ‘Media Content and Political Behavior in Observational Research: A Critical Assessment’ they raise a number of issues with regards to our article and to combining survey and content data in general. We agree that this topic is important, and we are pleased with the opportunity to respond to their Note. Our response is organized as follows. We first address some of their observations about our article. We then discuss the idea of combining survey and content data more generally and the virtue of this approach. Finally, we list a number of issues that may help us looking forward.