Mediated Leader Effects: The Impact of Newspapers' Portrayal of Party Leadership on Electoral Support.

Conventional wisdom holds that party leaders matter in democratic elections. As very few voters have direct contact with party leaders, media are voters' primary source of information about these leaders and, thus, the likely origin of leader effects on party support. Our study focuses on these supposed electoral effects of the media coverage of party leaders. We examine the positive and negative effects of specific leadership images in Dutch newspapers on vote intentions. To this end, we combine an extensive automated content analysis of leadership images in the media with a panel data set, the Dutch 1Vandaag Opinion Panel (1VOP), consisting of more than fifty thousand unique respondents and 110 waves of interviews conducted between September 2006 and September 2012. The results confirm that media coverage of party leaders' character traits affects voters: Positive mediated leadership images increase support for the leader's party, while negative images decrease this support. However, this influence is not unconditional: During campaign periods, positive leadership images have a stronger effect, while negative images no longer have an impact on subsequent vote intentions.

Political leaders and the media. Can we measure political leadership images in newspapers using computer-assisted content analysis?

Despite the large amount of research into both media coverage of politics as well as political leadership, surprisingly little research has been devoted to the ways political leaders are discussed in the media. This paper studies whether computer-aided content analysis can be applied in examining political leadership images in Dutch newspaper articles. It, firstly, provides a conceptualization of political leader character traits that integrates different perspectives in the literature. Moreover, this paper measures twelve political leadership images in media coverage, based on a large-scale computer-assisted content analysis of Dutch media coverage (including almost 150.000 newspaper articles), and systematically tests the quality of the employed measurement instrument by assessing the relationship between the images, the variance in the measurement, the over-time development of images for two party leaders and by comparing the computer results with manual coding. We conclude that the computerized content analysis provides a valid measurement for the leadership images in Dutch newspapers. Moreover, we find that the dimensions political craftsmanship, vigorousness, integrity, communicative performances and consistency are regularly applied in discussing party leaders, but that portrayal of party leaders in terms of responsiveness is almost completely absent in Dutch newspapers.