Research
11.03.2021

In a German "super election" year, can voting advice apps affect voter behaviour?

Simon Munzert and co-authors publish studies in Public Opinion Quarterly and Public Communication on how apps affect voter behaviour.

Germans in this “super election” year will face six state parliamentary elections and a federal election, amid a changing landscape of political parties. A growing number of popular Voting Advice Applications (VAAs) offer voters in many countries information on political parties to help them make informed decisions. The most popular German VAA, the Wahl-O-Mat, was used more than 15 million times in the 2017 election. This suggests that a considerable share of the electorate actually uses it. But do these apps actually have strong effects on voter participation or on their choices, as some studies have suggested?

Two papers recently co-authored by Simon Munzert, Hertie School Assistant Professor of Data Science and Public Policy and part of the Hertie School Data Science Lab, casts doubt on findings in other studies that such apps strongly influence voter behaviour. The studies do find that these apps make citizens more informed. 

In the first, “Do Online Voter Guides Empower Citizens? Evidence from a Field Experiment with Digital Trace Data,” published in January 2021 in Public Opinion Quarterly, Munzert, as well as Pablo Barberá of the University of Southern California, Andrew Guess of Princeton University and JungHwan Yang of the University of Illinois, provide evidence from a survey of over 1,000 German citizens in the 2017 federal election campaign.

The researchers assessed political behaviour, attitudes, media consumption, political knowledge, and social media activity. Unlike previous research on VAA usage that relies on people’s self-reports, the researchers used passive tracking software installed on desktop machines and smartphones to observe actual usage of the VAA. In addition, they drew on an experiment in which participants were randomly incentivized to use the tool. This allowed them to study the effects of VAAs in populations that are usually less likely to use it.

“Our findings reveal that the overwhelming consensus in favor of positive effects on turnout and vote choice should be treated with caution, as we find no such effects,” the researchers concluded. However, especially in today’s complex online information environment, they found that VAAs can help increase knowledge about parties’ positions on issues.

In a second paper, “Meta-Analysis of the Effects of Voting Advice Applications,” published in January 2021 in the journal Public Communication, Munzert and co-author Sebastian Ramirez Ruiz, a Research Associate and Hertie School Master of Public Policy graduate, present the most comprehensive review of VAA effects studies to date. They review the influence of voting advice applications (VAAs) on three core outcomes: turnout, vote choice, and issue knowledge. The meta-analysis includes 55 effects reported in 22 studies, and comprises 73,673 participants in 9 countries.

The authors report that they found “strong evidence for positive effects of VAA usage on reported turnout” and “modest evidence on knowledge increase.” However, the size of the observed effects vary considerably across studies. “Effects are substantively weaker in causally more rigorous experimental studies,” the authors write. This calls for better-designed experimental research and studies focusing on the acquisition of knowledge in VAA usage, they say.

Due to their popularity, VAAs look like they are here to stay, Munzert says. This is also mirrored through the fact that parties care a lot about how they are represented in the tool. Given the evidence in the study, Munzert says it is unlikely that they have a substantive impact on the vote outcome, but they can make people more informed voters, which is a key ingredient for a functioning representative democracy.

Read the paper in Public Opinion Quarterly here.

Read the paper in Public Communication here.

The Hertie School is not responsible for any content linked or referred to from these pages.
Views expressed by the author/interviewee may not necessarily reflect the views and values of the Hertie School.