Study title
Personalization of Politics between Television and the Internet
Ref study 12616
Study language English
Contributing institutions
  • Public opinion
  • Political attitudes
  • Political behaviour
  • Political communication
  • Political leadership
Geographical space
EFTA Countries
Over the last decades, the "personalization of politics" has turned into one of the defining elements of the democratic process. However, the common wisdom that sees popular political leaders as a fundamental electoral asset for their own parties finds only limited support in the scant voting literature. So far, comparative electoral research has proven reluctant in systematically addressing the impact of leaders on voting across time and space. Equally crucial aspects such as the role played by television exposure as a driver of personalization in voting behaviour, and the relationship between the rise of Internet-based political communication and the personalization trend have so far been under-researched.

The major aim of this project is to empirically assess the extent to which political leaders have come to affect voters' choice, as well as the role played by the media in driving this development across time. This project substantially extends the existing state-of-the-art on personalization of politics and electoral behaviour through its innovative methodological approach and its wider geographical/longitudinal scope. It involves the longitudinal harmonization and analysis of over a hundred existing cross-sectional datasets from European countries (1960-2015). Analysis of available panel data will complement the findings of the longitudinal analysis by focusing more in depth on causal dynamics. With regard to the research environment in which to place the study, the current state of the literature suggests that European democracies represent the locus where more research is in need. European democracies highlight many of the crucial variations in the structure of democratic politics and thus provide the ideal framework for such a thoroughly comparative analysis. The European-wide dimension of the study will allow for an extensive testing of the institutional, contextual and technological factors mediating leader effects across time and space.
Methods (description)
The project will target all publicly available National Election Study (NES) datasets collected in European countries, including all EU member states as well as those non-member states with established tradition in the field, such as Iceland, Norway and Switzerland.

Available cross-sectional datasets will be pooled in order to test the relationships between key variables across countries and time. The dependent variable of the analysis is respondents’ party choice in a given election. The measurement of leader evaluations will rely on “the most frequently included type of question about leaders in election studies” (Bittner, 2011: 16) – namely: feeling thermometers. In order to assess the relative importance of leader evaluations on voting, the statistical analysis will proceed with fully specified models of voting, so as to control for the wide range of individual-level variables that previous scholarship has shown to affect voter choice (e.g., socio-demographic characteristics, socio-economic status, party identification, ideological and issue proximity, economic evaluations). The analysis will rely heavily on multilevel modelling techniques, in order to illuminate on contextual/institutional factors as well as individual-level patterns of media exposure as mediators of leader effects on voting.
Methods (instruments)
Replicated study No
Financed by
Ethical approval No
Study type
Data availability
Source (Updates) Web
Date created 05.12.2016
Date modified 06.12.2016
Start - End date 01.01.2017 - 31.12.2019