Altruism, risk aversion and health concerns do not appear to affect whether college students comply with stay-at-home orders during the coronavirus pandemic, a new study shows, calling into question the efficacy of relying on individual motivations to enforce public policy.

Research from the University of California Merced showed that students were not necessarily affected by protecting the safety of others or of avoiding risks to themselves in choosing whether to abide by stay-at-home demands. 

Nine days after California issued its stay-at-home order and while the state was reporting over 1,000 new COVID-19 cases every day, 25% of college students surveyed said they violated the order and socialized with others for non-essential, nonwork purposes.

Ketki Sheth and Greg Wright, co-authors of the research published in Review of Economics of the Household, surveyed 338 undergraduate students between March 26 and April 7 to determine compliance with California’s stay-at-home order and individual tendencies toward risk and altruism.

When asked about their behavior over the past 24 hours, 51% of students said they left their home and 35% said they had social interactions other than for the purposes of obtaining food, health care or banking services. About a week later, 52% of students said they left home and 33% had social interactions, only eight percentage points of which were related to paid employment.

Self-reported risk and altruism preferences have been shown in other studies to be accurate predictors of risky behaviors, such as smoking and holding stocks, and altruistic behaviors, such as helping strangers and volunteering. In this study, though, no connection was found between a person’s preferences and their likelihood of complying with stay-at-home orders.

People with a higher risk tolerance were more likely to leave their home and interact socially overall, but when focusing only on social interactions unrelated to essential services or employment, risk tolerance was no longer associated with noncompliance.

Additionally, unlike previous research that has shown people are more likely to follow public health policies if they’re old, have asthma, have diabetes or are otherwise more sensitive to the health crisis, Sheth and Wright found no correlation in their research.

“Again, our population is kind of special,” Wright said. “I mean, we’re looking at college students.”

Even if the findings aren’t applicable to the population at large, though, Wright said tracking the behavior of college students can help mitigate infection rates.

“This is actually a very specific and important policy question that university administrators are facing and that the whole country is facing in terms of how college students behave,” Wright said. “I mean, a lot of what people are finding is that the outbreak and the spread of this is much more significant in college towns, particularly when the students come back."

Responsiveness to the pandemic may have also been more politicized in the U.S., the researchers said, with risk perception of COVID-19 and social distancing compliance shown by other studies to be affected by partisan leanings in the U.S.

“If you watch Fox News, you’re more likely to be a Republican and you’re less likely to wear a mask, comply with stay-at-home and believe that COVID is a big deal,” Wright said. “If you watch CNN and MSCNBC, you’re likely to believe kind of the opposite. So it seems like this has become, unlike previous health crises, a partisan thing.”

In an effort to publish their research quickly, Wright said he and Sheth had to make some tradeoffs, one of which was the ability to identify causal impacts. 

“Usually, a paper like this might not fly as well in the field of economics because at the end of the day, we can say things about correlations,” Wright said. Economists usually prefer the ability to observe people before and after some kind of shock in order to identify changed behaviors.

“So the tradeoff was that maybe we can’t quite get the research design, the empirical method, that would be ideal, but it’s kind of shifting gears toward, I think, the public service element of the job in a way,” he said.

Though the paper was written quickly and isn’t particularly long — including references, it fills only 12 pages — Wright said he believes the researchers asked an important question.

“And hopefully, now it’s out there with thousands of other papers, maybe it helps inform policy,” he said. “That’s the goal.”

The study “The usual suspects: do risk tolerance, altruism, and health predict the response to COVID-19?,” published Oct. 24, 2020, in Review of Economics of the Household, was authored by Ketki Sheth and Greg Wright, University of California Merced.