Yes, the Campaign Matters: Issue Voting in U.S. Presidential Elections

• In Event: The Campaign, Issues and Salience

Abstract

In the United States, forecasting models based on structural factors like the economy and party identification often predict the winner months before the campaign has even started. The predictability of electoral campaigns has fed the view that campaigns “don't matter”. This view was somewhat reinforced by the growing polarization of the party system and the resulting sorting out of voters into partisan camps over the last decades. This has led some to argue that American voters live in “echo chambers” where they are only exposed to information that reinforce their partisan predispositions. This calls into question the role that issues and the policy information delivered by campaigns play in the decision-making process of voters.

My study uses a cognitive theory of directional voting to assess the role of policy information in voter decision-making in the United States. This theory foregrounds campaigns as the main vehicle through which voters get information about the policy positions of candidates. This move also has the advantage of avoiding the causal inference problems linked to using voters’ subjective placements of the candidates on issues, which are confounded by voters’ self-placement and feelings toward the candidates. To examine the effect of policy information on voter decision-making and its mediation by affective predispositions, I use survey data in combination with original data on the issue statements made by candidates on Twitter during the 2012 and 2016 presidential campaigns.

My findings challenge the echo chamber thesis by showing that voters respond to policy information that is congruent with their preferences from both the congenial candidate and the candidate they are prejudiced against. Hence the results show persuasion effects, whereby voters were convinced by the campaign to vote for a candidate they were prejudiced against, and activation effects, whereby ambivalent voters were convinced by the campaign to vote for the candidate they were predisposed toward. Overall, my findings show that campaigns matter and that policy information still plays an important democratic function, even in a highly polarized political system.

Author

• Sarah Lachance, University of Toronto