New research suggests voters respond more to feelings than facts

A recent study published in Frontiers in Political Science provides evidence that emotions influence voting decisions more strongly than policy preferences. By analyzing data from five United States presidential elections between 2000 and 2016, the researchers found that voters are not only swayed by how closely a candidate’s views align with their own, but also by how the candidate makes them feel. Emotional reactions—especially positive feelings toward one candidate over another—tend to have a larger effect on vote choice than rational evaluations based on ideology.

Voting is often thought of as a rational act, one in which citizens weigh the pros and cons of candidates based on issues, policies, and competence. This perspective is shaped by long-standing theories such as rational choice theory, which suggest voters make calculated decisions to maximize their self-interest. In this view, individuals compare party platforms to their own policy preferences and vote for the option that promises the greatest benefit.

However, emotions have long been part of the political experience. Campaigns use music, imagery, slogans, and personal stories to connect with voters emotionally. Political scientists have increasingly recognized that affect (feelings like hope, pride, anger, or fear) plays a role in shaping political behavior. While emotions have been shown to mobilize voters and strengthen partisan identity, fewer studies have directly compared their influence with that of rational thinking in determining how people vote.

The new study set out to do just that. The researchers aimed to assess the extent to which emotional reactions and rational evaluations independently and jointly predict vote choice. More importantly, they wanted to understand whether emotional factors have a stronger effect than rational ones.

“I published a paper entitled ‘Why Do People Vote? Rationality or Emotion‘ in the International Political Science Review, where I examined how rationality and emotion shape voter turnout. Building on this work, Professor Costas Panagopoulos and I extended the analysis to explore how these two factors influence individuals’ vote choices. Because little scholarly attention has been devoted to examining the effects of rationality and emotion on vote choice simultaneously, our study seeks to address this gap in the literature,” explained study author Ching-Hsing Wang, an associate professor at National Cheng Kung University.

Free daily newsletter

For their study, Panagopoulos and Wang analyzed voter data from the American National Election Studies, which collects information from nationally representative surveys conducted around each presidential election. They focused on the five elections from 2000 to 2016, encompassing contests won by George W. Bush, Barack Obama, and Donald Trump.

The researchers built a statistical model that included two main components. The first was a “party differential,” a measure of how ideologically close a voter felt to each major political party. This represented the rational side of vote choice. The second was the “emotion differential,” which reflected how positively or negatively a voter felt about the two major party candidates. This served as the emotional side of the equation.

By using a method called binary probit regression, the authors estimated how much each factor contributed to a voter’s decision to choose one candidate over another. They also included other variables, such as presidential approval, views on the economy, party identification, education, income, age, race, and gender, to account for other known influences on voting.

The findings suggest that both rational evaluations and emotional responses independently affect how people vote. In every election examined, voters who felt ideologically closer to a candidate’s party were more likely to vote for that candidate. At the same time, those who felt more positive emotions toward a candidate—such as hope, pride, or affection—were also more likely to vote for them.

However, the researchers found that emotional reactions were typically more powerful predictors than rational assessments. In the combined analysis across all five elections, a one standard deviation increase in emotional preference led to a 9.2% increase in the likelihood of voting for a candidate. In contrast, a similar increase in policy agreement led to only a 3.1% increase. The emotional effect was nearly three times larger.

“Our study shows that emotions carry sufficient weight to influence voting behavior across the electorate, implying that campaigns that build emotional connections may be more persuasive than those relying solely on policy arguments,” Wang told PsyPost. “At the same time, because emotions are such powerful drivers, voters should make a conscious effort to reflect on their own reactions. By recognizing when their emotions are triggered and then assessing whether those feelings are supported by a candidate’s record or policy positions, voters can better balance emotional influence with rational judgment.”

This pattern held in every election year studied. For example, in the 2016 contest between Donald Trump and Hillary Clinton, the emotional differential had the largest influence of all. The study found that those who felt more positive emotions toward Trump were significantly more likely to vote for him, regardless of whether they aligned with his policy positions.

The data also suggest that emotions matter more in elections without an incumbent. In open-seat contests—when no sitting president is running—voters appeared even more emotionally driven. Emotional preferences had a stronger impact in 2000, 2008, and 2016 than in 2004 and 2012, when incumbents George W. Bush and Barack Obama sought reelection.

Another way the researchers evaluated the importance of emotions was by comparing how well different models fit the data. They found that models including both rational and emotional factors fit the data best. Models that excluded emotional considerations performed more poorly. This suggests that ignoring emotions in studies of voter behavior leaves out an essential part of the picture.

Finally, the researchers tested whether rationality and emotion interact—whether emotional responses shape the way people interpret policy positions, or vice versa. With one exception in the 2016 election, the data showed little support for this idea. Instead, emotion and rationality appeared to influence vote choice separately.

“Although both rational thinking and emotions shape how people decide their vote, emotions tend to carry greater weight than rational considerations,” Wang said. “Our findings suggest that voters should be mindful of whether a candidate is appealing more to their feelings than to their judgment and recognize how these emotions affect their choices.”

“When experiencing an emotional reaction, it is advisable to review the candidate’s track record, policies, or debate performance to see whether the feelings are grounded in substance. Voters should also remain cautious of emotional appeals, as campaigns often use music, imagery, and storytelling to influence public sentiment, and being aware of these tactics can lead to more balanced decision-making.”

The study provides robust evidence across multiple elections, but the authors acknowledge there are still some limitations. For instance, their analysis focused on direct effects and did not explore the many ways emotions might also influence vote choice indirectly. For example, emotions can shape how people seek out information, interpret news, or process campaign messages. These subtler dynamics were beyond the scope of the current study.

In addition, the measures used for rationality and emotion relied on self-reported survey responses, which may not capture all dimensions of how people form political judgments. Future research could explore alternative ways of measuring both cognitive and emotional inputs, including how they change over time or respond to specific campaign tactics.

“Our next steps are to explore whether rationality and emotion play significant roles in shaping individuals’ political behaviors, such as participation in protests, and to examine their effects in contexts beyond the United States,” Wang said.

The study, “Rationality, affect, and vote choice,” was published August 18, 2025.