| My dissertation builds upon Donald Stokes' critique of spatial models, and the lack of attention paid to valence issues in the determining of electoral outcomes. While most valence-related studies define valence in terms of economic performance or incumbency, I create a new cross-national measure of valence for parties in nine West European countries over a 22-year period based upon content analysis of news reports taken from Keesing's Record of World Events. This new measure defines valence in terms of specific events that would affect voters' evaluations of parties along three dimensions: competence, integrity, and unity. Examples of these "valence events" would be political scandals, intra-party disagreements, and acts/decisions that would suggest party incompetence. The major finding of my research is that parties' valence images significantly influence their electoral support, and that the electoral effects are of a similar magnitude to the effects of their announced policy positions. To date, no study has systematically examined whether valence events such as political scandals affect parties' vote shares, nor estimated by how much. |
