In May 2013, around 30 political scientists from across Europe met in Lausanne, Switzerland, for a workshop to discuss various aspects of VAAs and their influence on parties and voters. One of the outcomes of this workshop was the Lausanne Declaration. This represents the first attempt to establish a framework for defining the quality of a VAA. In addition to the actual design (questionnaire, matching algorithm, usability, etc.), ethical aspects relating to organizational and operational areas were also considered.
Since then, the spread and use of VAAs have increased significantly, making a strong focus on their quality and ethical aspects even more important. At the same time, both the political environment (including democratic backsliding and political polarization) and the technical environment (such as AI, LLMs, and chatbots) have undergone significant changes. As a result, the Lausanne Declaration no longer meets today's requirements and needs to be modernized.
The research network on VAAs within the European Consortium for Political Research (ECPR) has therefore launched a process to develop a modernized version of the Lausanne Declaration.
Source: Garcia, Diego, and Stefan Marschall (eds.) (2014). Matching Voters with Parties and Candidates. Voting Advice Applications in a Comparative Perspective. ECPR Press: 227-228.
Elections are a central element of democracy. They legitimise the allocation and the use of political power. Elections have to be organised in a true and fair manner, allowing citizens to make their decisions based on their free will. Citizens have to be informed about the available electoral choices they have. Being convinced that Voting Aid Applications (VAAs) provide valuable information about candidates and parties running for elections, support citizens in the decision-making process in the course of elections, and allow for electoral choices which are closer to the political position of the voters, and considering that VAAs have become increasingly popular and potentially influential in the electoral process, we abstain from suggesting an ideal form of a VAA, but rather recommend certain standards and minimal requirements that should be respected by all the makers of VAAs.
In order to contribute sustainably to the good functioning of democracy, VAAs should be open, transparent, impartial and methodologically sound.
All institutions, organisations, associations, groups, private companies and individuals financially supporting a VAA have to be made visible. Funding has to be made transparent.
All intentions and purposes associated with these tools have to be revealed by the makers of VAAs
A VAA should be freely accessible to all citizens.
A VAA should aim at the inclusion of as many parties/candidates that are on the ballot as possible. The criteria for the exclusion of parties and candidates should be publicly available and justified.
Parties and candidates should not be excluded from the tool for ideological reasons.
VAAs should be designed in a simple and intuitively understandable manner.
VAA makers ought to carefully watch that the design does not favour a party/candidate in a systematic manner.
VAAs are based on the assumption that users’ proximity to parties and candidates can be measured by their degree of accordance on political issue positions. Ideally, VAAs make this presumption visible.
The issue statements included in a VAA should be relevant and reveal the different dimensions of competition in the political system for which the VAA is designed. If applicable, voters should be able to express their issue salience by weighting or deciding on which issues they want to be compared to parties and candidates.
Party and candidate positions on the statements can be coded on the basis of expert opinions, of documents and party manifestos, and of selfplacements. The method used to position parties and candidates should be made known to the users of the VAA.
Following the principle of transparency, the algorithm matching users to parties and candidates should be documented and clearly explained to users.
The results can be presented to users in the form of rankings, maps, spiders and graphs. Visualisations should be valid and instructive. Guidelines for understanding the results should be provided to the users.