Professor uses psychology to help predict elections, analyze policy

December 1, 2020

By Andrea Klick, student reporter

While most pollsters expected Joe Biden to have a runaway win from President Donald Trump, Wändi Bruine de Bruin, Provost Professor of Public Policy, Psychology and Behavioral Science at USC Price, predicted a much closer outcome. Her method had been accurate before; in 2016, when experts expected Hillary Clinton to win with a landslide vote, Bruine de Bruin’s poll question found Trump to be the clear winner.

How did these contrary results come to be? During the Los Angeles Times USC Dornsife Presidential Poll for the 2016, 2018 and 2020 election cycles, Bruine de Bruin and her team asked participants not just who they’d be voting for, but also the percentage of their social circle who would be supporting each candidate.

The technique, called social circle polling, has helped more accurately predict election results in recent US elections as well as elections in the Netherlands, Sweden and France. In a way, social circle polling helps researchers gain a larger sample size by learning the reported opinions of their respondent as well as that person’s perceived opinions about their social circles.

Bruine de Bruin said that respondents may feel more comfortable sharing how their friends and family will vote than to admit how they themselves will vote. And while respondents may feel compelled to say they’d vote for one candidate over another, they’re usually more likely to vote for the same candidate as the people they surround themselves with.

“Because these are people we’re close with, we know what they’re doing – but presumably these are also people that we respect and like,” Bruine de Bruin said. “So, if they’re doing something, we’re more likely to do it also. It’s not necessarily bad social influence because we’re all busy – I don’t want to read all the guidance and all the decisions I have to make! For some of my decisions, I may just ask someone in my social circle what they’re doing and work from there.”

Even though the USC Dornsife Presidential Poll predicted Trump as the narrow winner of the recent presidential election this year, its projection illustrated the closeness between the two candidates better than many national polls that anticipated a runaway win for Biden.

During this pandemic and election focused year, Bruine de Bruin has also researched ways to successfully implement policies such as face mask requirements. Her work currently involves surveying people to learn if they wear masks, in what situations they wear them, and why they may or may not, so that policy and public health interventions can be better planned. If policymakers try to create messaging around mask wearing or any policy without first talking to community members, they likely won’t be successful, Bruine de Bruin said.

Even with virus prevention information being widely available, Bruine de Bruin said people may not respond to government messaging or be able to easily implement recommendations without support or resources. In terms of mask wearing, Bruine de Bruin said that despite media coverage of protests against coronavirus restrictions, surveys she has conducted with USC’s Understanding Coronavirus in America Study have found that a majority of Republicans do wear masks. “Behavioral science can help policymakers understand why people may not be wearing masks and what resources and communication strategies they could implement to help more people participate,” she said.

“[Policymakers] communicate about their findings with their colleagues, and they’re good at that. That’s what they’re trained in. But that doesn’t necessarily mean that they’re good at communicating with non-expert audiences to whom they want to give advice,” Bruine de Bruin said. ”That means they use words that are not understood in the same way or understood at all by their non-expert audiences. They think that people have motivations that people don’t necessarily have, and then they get frustrated and think people are stupid.”

Bruine de Bruin’s work focuses on behavioral science and the intersections of policy effectiveness and psychology, and she will start teaching at USC Price next semester. She hopes to teach future policymakers how to create effective and impactful policies and campaigns by researching community members and learning the best ways to help them implement a policy in their lives.

At USC Price, her courses will focus on how to use insights and methods from behavioral science to implement more effective policies — and to track the impacts of these policies.

“The easiest intervention is one in a domain where people want to achieve a certain goal,” she said. “They’re currently not achieving it because they have some issues implementing the recommendations. Find out what those issues are and then help them with it, either by giving them information or by changing the options that they have or by changing the policy.”

Related faculty