Author Description

This dissertation examines the effects of political scandals on political behaviors at three different analytical levels across Western European countries (mainly on Germany and the United Kingdom). Recent developments in the literature regarding political scandals have led to debates on whether and how political scandals impact political behavior. However, there are limitations in the literature, which include a narrow-focus on specific types of scandals as well as issues regarding research design that restrict our comprehension of the effects of scandals. Due to these limitations our understanding of whether political scandals have any impacts on political behavior or how they impact political actors is still unclear. Building on recent work with regards to developments on causal inference and experimental design, this dissertation explores the causal effects of political scandals on parties' support rates, voters' evaluation of individual politicians, including candidates, the prime ministers, and the ministers, and finally, interactions among cabinet members and the members of parliament (MPs). The first chapter examines whether the involvement of a minister in a scandal affects his/her own party's support rate. To test the hypothesis, this chapter uses interrupted time-series analysis to the data constructed by Jennings & Wlezien (2016), which brings together daily polling of parties across a significant period of time in Western European countries. Together with the results derived from the techniques of breakpoint detection along with the analytical results from three cases across different time periods, this chapter finds that supporters usually do not withdraw their support due to a single scandal that the minister is involved in, and furthermore, a minister's scandal does not hurt the support rates of his/her coalition parties too. Secondly, the results suggest that researchers need to take techniques associated with break-point detection, such as Double M