Research

Information and Voter Decision-making

In my dissertation, Democracy in an Uncertain World: Campaign Information and Voter Decision-making, I compare the influence of the policy information provided by campaigns on vote choice in the United States, Canada and Germany. I argue that the view that “campaigns don’t matter” is a by-product of the focus on the United States, whose highly polarized two-party system represents a unique set of conditions that reduce the value of campaign information. Hence my dissertation moves beyond assumptions of universality in voter cognition that result from an overreliance on the U.S. case to analyze how voters use campaign information in their vote decision in different contexts. To do so, I use original data on campaign policy statements made on Twitter, electoral surveys, and experiments.

In a study on the U.S. 2012 and 2016 presidential elections, I assess how feelings toward parties and candidates—which have become highly polarized—moderate the effect of policy information on vote choice. I use an empirical strategy that can identify the effect of policy information on vote choice without the upward bias that comes from voters following their favoured candidate on issues. My findings challenge the “partisan intoxication” thesis, which claims that partisan cues trump policy. I find that voters can be persuaded to vote for a candidate they were prejudiced against when the candidate emphasizes issue positions that are congruent with their own positions. I also find that the candidates’ communication on issues during the campaign is necessary to convince some voters to vote for the candidate they were predisposed toward. Given the high baseline probability of voting for the favoured candidate, issue voting only plays a limited role in explaining electoral outcomes, yet one that can make a difference.

In contrast, in another study on Germany I find that campaign information plays an important role in explaining electoral outcomes. In multiparty systems, voters who care about certain issues must consider how the policies of the government will be affected by compromises between coalition partners. In this context, voters use the policy statements of minor parties during the campaign to target a government coalition whose policies would be in line with their policy preferences. My study clarifies the theoretical mechanisms that underpin different modes of strategic voting for minor parties—compensatory voting and coalition-insurance voting—and proposes a hybrid strategy. My results show that a substantial share of the votes for the Green party and the Free Democratic Party in the 2013 federal election were driven by a hybrid strategy, and votes for the Green party in 2017 by a compensatory strategy.

Finally, my third study addresses a puzzle that arises from the rational choice literature, which assumes that voters are risk-averse: if voters dislike ambiguity, why do electoral candidates routinely take ambiguous policy positions by making vague or contradictory statements? Using an ecological rationality theoretical framework, I argue that voters use the ambiguity of policy statements by major party candidates as a heuristic for assessing their positions relative to clear centrist candidates. I hypothesize that they take ambiguity as signalling higher variance in potential policy positions, hence a higher likelihood of holding non-centrist policy positions compared to clear centrist candidates. Using an experiment fielded in Canada, the United States and Germany, I find that policy ambiguity attracts voters who sit between the center and the extremes, yet it does so without repelling centrist and extremist voters. Hence it offers a solution to the puzzle and explains why policy ambiguity is an attractive strategy for electoral candidates.

Campaign Effects

In the United States, forecasting models often predict electoral outcomes long before the electoral campaign starts. This begs the question: do campaigns matter? Some argue that campaigns only have minimal effects: they merely activate “fundamentals”, understood as long-term structural forces, like the state of the economy and party identification. The counterpoint is that structural forces are only part of the story: the strategies of the candidates determine whether they are in fact activated, and can even induce new fundamentals that produce a durable shift in public opinion.

In The Predictable Campaign: Theory and Evidence (with Richard Johnston), we investigate how the institutional context conditions campaign effects.

Political Behaviour

My other main research area is political behaviour defined broadly, inside and outside the electoral context.

In Group-based Affect and the Canadian Party System (with Edana Beauvais), we analyze whether the relationship between group-based affect and the vote in Canada reflects a polarized pluralism system across seven elections from 1993 to 2019.

In Policy Design, Cost Information and Support for Guaranteed Income (with Richard Johnston and Alan Jacobs) we analyze how different information frames affect support for basic income in British Columbia.