A Better Way of Mapping Political Landscapes? Estimating Party Positions by Using Voting Advice Application Data

In party research, there is an ongoing debate on the most suitable method for positioning political parties within the political landscape. Methodological approaches vary, from drawing on expert or mass surveys, via observing legislative behavior of parliamentary party members to analyzing the content of party programs. A relatively recent trend in party positioning is the use of data provided by Voting Advice Applications (VAAs). VAAs present an innovative way to capture party positions and have augmented the repertoire of approaches for estimating party positions in the policy space. However, they offer more than just another way of mapping the political landscape: we argue that they can even improve the process of positioning parties. Using a deductive approach by assuming that there is a left-right-economic dimension on the one hand and a social-cultural dimension on the other enables us to map the political landscape two-dimensionally on the basis of party positions provided by an VAA. The coding of categories is done using the renowned codebook from the "Manifesto Project Dataset" (Volkens et al. 2020). While several papers argue that topics like Environmental protection or European Union are not applicable for the positioning process on the aforementioned dimensions, VAAs as the German Wahl-O-Mat pinpoint political issues on seemingly non-controversial aspects rather concretely. Using them, we show that it is possible to transform formerly non-confronting categories into controversial issues. As a result, our findings for the political landscape in Germany show a familiar but - compared to the usual approaches - slightly different pattern. The debate about the valid positioning of parties in the political space can thus be reopened, with VAAs as potential game changer.