Image repair in the aftermath of inaccurate polling: How the news media responded to getting it wrong in 1948 and 2016

The US presidential elections of 1948 and 2016 produced surprise outcomes when the predicted winners ended up losing the election. Using image repair theory, this article explains the strategies the media used to repair their image in light of predicting the wrong winner. Using a qualitative analysis of news coverage that immediately followed the 1948 and 2016 presidential elections, this study finds that the media utilized similar image repair strategies of offering explanations for poor information, highlighting the media’s good reporting, diminishing the harm caused by the inaccurate predictions, and justifying the inaccurate predictions of both elections. However, the media responses in 1948 and 2016 differed greatly in tone and in the utilization of a new attack strategy to deflect criticism of the media itself. These strategies suggest that media use of image restoration is limited because of the unique societal expectations placed on the press, and that the media’s inaccurate 2016 predictions and subsequent attack strategies may have been contributed to the heightened criticism of mainstream news.     

The leadership dilemma: Examining the impact of strong leaders on parties

Party leaders are often regarded as crucial to a party's success. Successful leaders tend to be big personalities who dominate their party's organisation, policy development and electoral campaigns. But does that control come with a price? We test to see if such leaders damage their parties in the medium term. This happens because strong leaders might be ceded too much control of the party organisation, policy and electoral strategy. We specifically hypothesise that political parties will go through a period of leadership instability and electoral decline after strong leaders step down. Using a dataset with elections under party leaders in nine countries over a 25-year period, and a qualitative case study, we find some evidence for the theory, which should prompt further research of the question.     

Political support through representation by the government. Evidence from Dutch panel data

Research on political support demonstrates that satisfaction with democracy is higher among electoral winners than losers, and that it is higher for citizens who are ideologically more congruent with the government. In this paper, I analyze how support for the political system is affected by representation by the government. Expanding on previous studies, I leverage long-run panel data from the Dutch LISS panel spanning over several electoral cycles. Drawing on various measures that go beyond the distinction between election winners and losers and also measure how close citizens are to the government coalition as a whole, I show that being well represented by the government has a wide-ranging positive relationship with satisfaction with democracy, external efficacy and trust in political institutions. While this relationship is mostly short-run, political support can decline substantially if non-representation persists in the long-run. This highlights the relevance of long-run panel data for studying the consequences of representation.     

Are European Elections Second-Order Elections for Everyone?

Second-order election (SOE) theory has been used to explain voting behaviour in European elections. Voters believe that less is at stake in some elections and some voters hence tend to cast a protest vote. However, most studies on the topic have focused on the demand side of SOEs – i.e. on the voters – and have ignored or only partially tackled the supply side – i.e. the strategic behaviour of parties – and excluded small and marginal parties (SMPs). However, SMPs may have greater incentive to seriously compete in SOEs. The 2019 European Parliament elections in Germany were particularly interesting to SMPs as there was no voting threshold, thereby increasing the chances of SMPs. To capture whether SMPs view European elections as first-order elections and therefore expend more resources than parliamentary parties on these elections, I analysed campaign expenditures and manifestos. Results indicate that SMPs act within a rational actor framework, though not unequivocally.