Please Spit in This Tube: An Election Experiment
Every day another article comes out about how voters are stressed by this election. But we wanted to know: what is the election doing to our biology?
The American Psychological Association recently found that more than half of all Americans — 52 percent — say this year’s presidential election is a “somewhat” or “very significant” source of stress in their lives. The survey was self-reported, meaning respondents answered a few questions online and the APA took their self-assessments at face value. Anecdotally, those assessments probably ring true for many of us, but it turns out there’s a way to measure the physiological effects of election stress.
Over the last few years, a group of neuroscientists and political scientists have pioneered a new field called biopolitics, the study of biology and political behavior. Professor Kevin Smith is a political scientist at the University of Nebraska-Lincoln and a co-author of the book, "Predisposed: Liberals, Conservatives, and the Biology of Political Differences.” He often collaborates with Dr. Jeffrey French, who runs a lab at the University of Nebraska-Omaha and studies cortisol, a hormone we release when we’re stressed.
One of Smith and French’s recent studies looked at stress and voting. They wanted to know if cortisol levels influence whether people vote. The easiest way to test cortisol is through saliva, so they collected spit samples from a bunch of participants and got their official voting records for the past six elections.
The researchers found that people with higher cortisol levels vote less. And that finding correlates with another one of their studies, which found that people who voted absentee experienced less stress than people who went to the polls.
So we asked French and Smith to help us design an experiment of sorts. We’d use the presidential debates as a proxy for the election. Our team would go to debate watch parties and collect saliva samples from viewers to measure their cortisol levels. We’d also ask the participants to fill out a survey about themselves: their party affiliation, age and self-reported stress level. And we’d see who had the biggest changes in their cortisol over the course of the debate.
During the first two presidential debates, we went to watch parties in Times Square, Midtown Manhattan and Northern New Jersey. Participants spat three times into tiny tubes: before the debate, to get a baseline sample, midway through the debate and after the debate.
We over-nighted the samples to Omaha, where Dr. French processed them in his lab. A few weeks later, he had the results.
We all agreed that the debate watch parties seemed stressful. At a bar in Times Square, we talked to young Republicans unhappy with their nominee and worried about their party’s future. Others were terrified at the prospect of a Clinton presidency. In Midtown, a group of Democrats had gathered to watch at the Roosevelt Institute, a left-leaning think tank. A few of them brought their own alcohol, to temper their anxiety (French and Smith took alcohol and caffeine intake into account in their analysis) and a number of them worried about Trump’s popularity.
But the results surprised us: cortisol levels stayed close to normal levels throughout the debates. Clinton supporters had a small spike at the midway point, but not by much. Overall, the stress levels for liberals and conservatives didn’t really change — with one exception.
The researchers looked at cortisol levels based on whether participants had someone close to them who planned to vote for the opposing candidate. And for Trump supporters who had a conflict with a person close to them — a parent, a sibling, a spouse — cortisol levels actually went up after the debate. They probably found the debate more stressful.
French and Smith warned us that this wasn’t a pristine study. In fact, both professors laughed when we asked if they’d submit our work to a peer-reviewed journal. But they agreed that this finding was statistically significant. And they didn’t find it for Clinton supporters, or voters who supported a third party candidate.
The other significant finding related to baseline cortisol levels — the participants’ stress level before the debate. The researchers found that Trump supporters had much higher baseline levels compared to Clinton voters.
Smith, the political scientist, couldn’t tell us why Trump voters had two times as much cortisol in their saliva compared to Clinton supporters. But he did say that our experiment served as an interesting pilot study — one that made him think differently about what he hopes to study next: tolerance.
Here, Smith made a comparison to same-sex marriage. Opposition to it shifted when researchers found some biological or genetic basis for being gay — when it started to be considered innate. Smith wonders if the same is true for political difference. As he told one of our reporters, “If you're a liberal and I'm a conservative and I believe you're a liberal because you're genetically predisposed to be, then am I more tolerant of you or less tolerant of you?”
In other words, if political difference is related to our biology, maybe we’ll be more tolerant of each other. And therefore less stressed. And therefore more likely to vote. At least, that’s the hope.
We've been on hiatus, working on some new stories. If you're joining us for the first time, here are some of our favorite past episodes: