Do voting aid applications matter? A mobilization field experiment with disadvantaged citizens

By | on October 27, 2017 | 0 Comment


Valérie-Anne Mahéo
Department of Political Science
Université de Montréal
Supervised by André Blais
Grant received in 2013-2014

Voting Aid Applications

Voting Aid Applications have become increasingly popular in the past years, in many countries (like Australia, Canada, Sweden, Israel) and at different levels of politics (would it be municipal, regional or national elections). While there exist a variety of voting aid applications, they are often non-partisan web tools that aim at informing citizens and facilitating electoral decision-making. The common operating principle of Voting Aid Applications (VAA) is that: they compare the positions of parties with the position of the citizen on a variety of policy issues, and at the end, they calculate and illustrate the general proximity of the citizen to the political parties. While they seem to attract more and more interest and attention from the media, governments and citizens, we still don’t know if these Apps effectively inform and mobilize the public. Past studies have in fact been unable to assess accurately the potential effects of VAAs, due to causal inference problems.

The Experiment

This article is based on a unique experimental evaluation of a VAA in the context of an election. During the electoral campaign of the 2014 Quebec election, I implemented a randomized field experiment to test whether a VAA, called the Vote Compass, could stimulate the political knowledge, interest, information behavior and turnout of the individuals who use it. I recruited 400 citizens on the street and in various locations of a low-income neighborhood, to offer them to take part in a survey and visit websites of information on an electronic tablet. This study innovates by reaching a bigger proportion of individuals who tend to be non-voters and non-users of this App, and randomly assigning them to use the Voting Aid App (or another website for the control condition). Furthermore, I evaluate the effects of the App at different points in time, in the short-term as well as in the medium-term (after the election). So we can effectively evaluate whether using the Vote Compass helps citizens in times of elections.

The Results

The results of this study show that there are no general effects of the Vote Compass on any of the political attitudes or behaviors, would it be in short or medium-term. However, if we look at the differential effect of the App for less or more educated citizens, we see that the Vote Compass tend to stimulate information-seeking behavior and voting intentions only among the less educated citizens (and to a certain extent political interest as well). While this study is exploratory and based on a small sample, it highlights that VAAs might not benefit all citizens equally.



  • 2017 – Presentation given at the Chaire de recherche en psychologie politique de la solidarité sociale (UQAM) : “How does the Vote Compass Affect Citizens’ Political Behaviour?”.
  • 2016 – Presentation given at Université Laval, Department of Political Science: “Does the Vote Compass Have an Impact of Citizens’ Political Behaviour?”.
  • 2014 – Presentation given at Elections Canada Speaker Series: “Participation and Mobilization of Canadian Youth and Disadvantaged Citizens”.


Comments are closed.