Swiss Study Finds Referendum Polls May Shape How Citizens Cast Their Vote

New research from Switzerland is adding fresh evidence to a long-running debate about whether public opinion polls merely measure voter sentiment or actively influence it.

A study published in the academic journal Electoral Studies suggests that voters in Switzerland’s direct democracy system may adjust their support for referendum proposals based on expectations about whether those measures are likely to succeed or fail. The findings contribute to broader international discussions about the role polling plays in democratic decision-making and public participation.

The research was conducted by political scientist Oliver Strijbis of Franklin University Switzerland in Sorengo, Ticino. Through a series of survey-based experiments involving approximately 4,000 participants, the study examined how information about a proposal’s perceived chances of success affected voter attitudes toward national initiatives.

According to the findings, support for a proposal tended to weaken when participants were informed that the measure was expected to lose. Researchers described this pattern as a “Titanic effect,” referring to a tendency for some voters to distance themselves from causes perceived as heading toward defeat.

The experiments focused on several Swiss popular initiatives, including proposals related to broadcasting fees, fireworks restrictions, and corporate accountability. Participants were presented with different information regarding the expected outcomes of these votes, allowing researchers to measure whether anticipated results influenced individual voting intentions.

The study also looked for evidence of a so-called “bandwagon effect,” a phenomenon frequently discussed in political science in which support grows for an option perceived as likely to win. While researchers observed some indications of this behavior, the effect appeared limited and did not reach a level considered statistically significant in the experiments.

At the same time, the research found no meaningful evidence for an “underdog effect,” in which voters rally behind a proposal because it appears unlikely to succeed. The results suggest that perceptions of probable defeat may have a stronger influence than perceptions of potential victory.

The findings emerge in a country where direct democracy plays a central role in political life. Swiss citizens regularly vote on constitutional amendments, national initiatives, and major policy questions through referendums held several times each year. Because public votes are frequent and often highly visible, Switzerland has become an important setting for studying voter behavior and democratic participation.

Questions about the influence of polls are not new. Political scientists have debated for decades whether surveys simply reflect public opinion or whether they can help shape it. Earlier Swiss and international research has pointed to several ways polling information may affect political behavior, including turnout levels and perceptions of electoral competitiveness.

Previous studies examining Swiss referendums found that polls can influence voter participation by affecting expectations about how close a vote will be. Research published by economists and political scientists studying Swiss ballot measures concluded that closely contested polls tend to encourage greater voter turnout, while perceptions of competitiveness can alter participation patterns across the electorate.

Those findings have helped fuel discussion about whether poll releases should be regulated more strictly before major votes. Several countries already impose temporary restrictions on publishing election surveys shortly before voting day, arguing that last-minute polling may affect voter decisions. Switzerland has generally favored transparency and broad access to political information.

Strijbis emphasized that the existence of polling effects should not automatically be interpreted as evidence of manipulation. The researcher argued that manipulation would require intentionally misleading or falsified survey results designed to influence outcomes. Instead, he highlighted transparency as the critical safeguard, including clear disclosure of who commissioned a poll and which organization conducted it.

The study also points to another challenge facing Swiss political observers: the relatively limited number of public polls available for many referendum campaigns. When only a small number of surveys are conducted, unusual or inaccurate results can have a larger impact because fewer competing measurements are available to provide context or correction.

For campaign organizations, the research offers practical implications. The findings suggest that emphasizing unfavorable polling numbers in an effort to energize supporters may not always produce the intended effect. Instead of motivating additional backing, highlighting likely defeat could further reduce support among undecided voters.

As Switzerland continues to rely heavily on direct democratic decision-making, understanding how voters process political information remains an important area of research. While the latest study does not suggest that opinion polls determine outcomes on their own, it adds evidence that expectations about success or failure can influence individual choices at the ballot box.

The findings are likely to remain relevant beyond Switzerland. Democracies around the world continue to grapple with questions about polling accuracy, media coverage, voter psychology, and the ways information can shape political participation. As polling technology evolves and access to real-time political data expands, researchers will continue examining how public expectations influence democratic outcomes.