Research question/goal:

This project investigated the short- and long-term dynamics of voting behaviour in German federal elections. In addition to studying these dynamics separately, it attempted to link short-term dynamics within election campaigns to long-term dynamics during a legislative period and beyond. In theoretical terms, the project was based on the idea that short- and long-term dynamics can be accounted for by the interplay of voter characteristics and situation-specific factors. 

The complex and multi-faceted research design of the GLES was successfully applied for the first time during the 2009 Bundestag elections. Shortly after the election, the project provided all interested researchers with an unprecedented pool of high-quality data sets. In addition, the 2009 federal election and electoral change in Germany were analysed from various perspectives in numerous publications using the data collected. In the second project phase, the proven design was continued and applied to the 2013 federal election.
Against this backdrop, German citizens eligible to vote were surveyed repeatedly again during the legislative period and the election campaign preceding the 2017 election as well as after this election and government formation. The sample comprised both citizens who had already been surveyed during the election campaign preceding the 2013 federal election and citizens who were surveyed in fall 2016 for the first time.

The results of our research indicate that when making up their minds in the run-up to federal elections, voters respond to campaign communications, though not all in the same way. Whereas for some voters, campaigns primarily serve as a means to confirm their political preferences, others respond to the campaign by changing their partisan preferences and make their decision late. Looking at the decision-making in subsequent elections, voters who had made up their minds long before the first election were not necessarily among the early deciders in the next election. Likewise, late deciding in one election predicted late deciding to some extent in the following election, but far from perfectly. Linking short- and long-term dynamics, in some cases, short-term campaign dynamics caused inter-election vote switching. In others, changes of voting intentions between election campaigns were neutralised by campaign dynamics that lead voters back to the party they had voted for in the preceding election. In a sizable number of cases, however, inter-election switching was due to changes of voting intention between campaigns, whereas voters did not change their voting intention during the campaign anymore. Inter-election differences in the time and substance of decision-making as well as the interplay of short- and long-term dynamics appear to be linked to the exposure to elite communication and other contextual features whose effects are conditioned by voter characteristics.

The German Longitudinal Election Study (GLES) has been permanently anchored as an institutionalised election study at GESIS. All the latest news can be found at gles.eu