Voting Advice Applications and the Challenges for Representative Democracies

The last two decades have seen Voting Advice Applications (VAAs) turning into a standard feature of election campaigns in most European countries (and beyond) with millions of users resorting to them. More recently, VAAs have turned into a subfield of political science in their own. In the last years, a progressive interest has arisen with respect to the consistency and reliability of the voting advice provided by these applications. Additionally, the impressive numbers of users visiting VAA-websites during election campaigns have led political scientists to research the effects of these tools on users’ electoral behaviour. Finally, VAAs have started to become a topic of democratic theory. At the turn of the decade, we deem VAA research ready to embark on a journey towards new research avenues. Existing studies provide relatively strong evidence about VAAs’ positive impact on turnout. Yet, much less is known about the underlying conversion effects. What are the parties and candidates that take advantage of the VAA-effect? Eventually, this question speaks to the notion of issue voting itself, to which the VAA literature has mechanically relied upon. Can VAA-based research illuminate the potential challenges stemming from fostering issue voting in this highly polarized age? Beyond a deepened understanding of their effects of (research on VAAs) we also encourage more analytical efforts using these tools as research devices (research with VAAs). What can we learn from VAAs about the socio-political transformations underlying the growing polari-zation of both mass publics and the elites in liberal democracies? Is there anything distinguishing about VAAs that makes them preferable to more established research tools and data pools? In addressing these pressing questions, the Panels of the Section are expected to bridge VAA research to central fields of political science such as electoral studies, party research, survey research and democratic theory. The Section will bring together scholars from different sub-fields of political science and aims to foster the linkages between these sub-disciplines and at the same time creating a more in-depth understanding of these increasingly relevant tools. Interdisciplinary analyses relying on insights from, e.g., AI/machine learning, big data analysis, online survey research, behavioural psychology, and normative theory, are especially welcome. These questions will be addressed concentratedly in three Panels covering the three main strands of VAA research. Due to the restricted number of Panels, this time the call for Papers will be highly competitive ensuring that only high-quality research will be presented. The Section is a follow-up of the successful General Conference Sections in Bordeaux (2013), Glasgow (2014), Montreal (2015), Prague (2016), Oslo (2017), Hamburg (2018) and Wroclaw (2019) dedicated to VAA research, each of which saw about 20 Papers presented and 30 authors attending. The section is endorsed by the ECPR Research Network on Voting Advice Applications. Panel 1: Analysing Effects of Voting Advice Applications on Users Chair: Marta Gallina Co-Chair and Discussant: Martin Rosema Abstract: Literature about VAAs has already shown that the usage of these tools can affect users in many different ways. First, VAAs can affect the attitudinal dimension of electoral politics, for example motivating voters to learn more about politics and party positions. Some scholars have pointed out that after doing such tests users exhibit increased levels of political knowledge, efficacy, or interest. Furthermore, VAAs can also affect the behavioural dimension of electoral politics, motivating users to turn out in higher numbers or affecting the candidate or party choice of their vote intentions. However, many questions around the VAA effects remain, such as what explains the differences in the size of these effects across contexts, the impact of self-selection of users on the impact, and what variables moderate the effects. The aim of this Panel is to bring together a set of papers that advance our insights into such attitudinal and behavioural effects of VAAs. Contributions focusing on the impact that VAAs have on different aspects of political attitudes and behaviour are most welcome, regardless of what methodology (quantitative, qualitative, or mixed), type of data (e.g. user statistics, survey data, experiments), or cases (single country, comparative study) are used in the analysis. Panel 2: The Design of Voting Advice Applications: Theoretical and Empirical Perspectives Chair: Kostas Gemenis Discussant: Fernando Mendez Abstract: Despite their prominence in election campaigns, Voting Advice Applications (VAAs) often prove to be particularly controversial. VAA designers often find themselves in conflict with political parties, while VAAs are sometimes forced offline due to complaints. Quite often, however, the concern for ethical issues about providing fair and unbiased “ad-vice” to VAA users hides deeper theoretical and methodological debates. The Panel therefore welcomes theoretical contributions that aim to link the design of VAAs to major voting theories, as well as empirical contributions illustrating particular advantages or pitfalls in VAA design. How voting theories inform the ways in which questions are formulated and information is presented to the users? How VAA design can ensure the fair representation of different parties and their views? How VAAs designers aim to serve citizens by using different dissemination strategies for their VAAs? How can we evaluate the efficacy of VAAs using self-selected samples of VAA users? The Panel welcomes theoretical and empirical research on these, and other related questions. Panel 3: The Construction of VAAs and Consequences for Party System Research Chair: Eric Linhart Abstract: The answers of the various competing parties to VAA questions can be compared to each other. This research strand has become of growing interest in the past years, and quite some analyses estimate programmatic distances between parties based on VAA data. While, by and large, the results of this research is promising, only few studies have expounded the connection between VAAs’ construction and party positions. To what extent are such estimations sensitive to single or few questions? How does the over-(or under-)representation of certain policy fields affect results? Do estimations over time indicate shifts of party positions or reflect changing importance of different topics? What can we learn by analyzing subgroups of VAA questions? The Panels welcomes Papers contributing these or similar questions.