The great experiment

The pandemic is tragic. It’s also an incredible chance to study human behavior.
Nadia Hafid for The Washington Post

As the pandemic spread in mid-March, psychologist Kristen Shockley got an email that would turn her life upside down: Her son’s day care was closing. Shockley and her husband, a start-up executive, both worked full-time, and their 15-month-old son required close supervision — they couldn’t exactly plop him in front of the TV all day. How on earth were they going to manage? But even as she was worrying, Shockley, an associate professor at the University of Georgia, saw an opportunity: “This would be a really interesting study,” she thought to herself.

Emily Anthes @EmilyAnthes, an award-winning science writer, is the author of "The Great Indoors: The Surprising Science of How Buildings Shape Our Behavior, Health, and Happiness."

Within days, Shockley, who studies how our work and family lives intersect, was sending out surveys to more than 300 dual-career couples whose child-care arrangements had been disrupted. She hoped to learn how the couples navigated the crisis (did they assign all the child care to one parent, switch off through the day, delegate to a grandparent?), how gender factored into these decisions, and whether any approach was associated with better family functioning, health and job performance. “I’m interested in a broader perspective on how couples decide on career compromise,” Shockley says. “There’s not a lot of good, solid research on the topic. So I thought, ‘Well this is just the perfect little microcosm, because people are having to make decisions — they’re being forced into it.’ ”

Over the past six months, the coronavirus pandemic has remade daily life, prompting widespread school closures, layoffs and home confinement. These changes have created social and economic chaos — but also unique research opportunities for social scientists, producing a “natural experiment” that could help answer questions about issues from family dynamics to how economic insecurity affects views of government policy.

Scientists are seizing the moment. By early September, the crowdsourced “COVID 19 Social Science Research Tracker” listed more than 300 projects — and that figure represents “just the tip of the iceberg,” says the tracker’s co-creator, J. Nathan Matias, an assistant professor of communication at Cornell University.

A student attends an online class at home in Miami on Tuesday. One team of researchers is studying whether school disruptions are leading to changes in children's habits such as exercise and bedtime. (Jayme Gershen/Bloomberg News)

How best to craft public health messages has been one irresistible topic. In two March studies of 2,200 people, researchers compared three different approaches to encouraging the public to take protective measures against the coronavirus, like washing their hands: One approach stressed protecting the individual, another emphasized safeguarding the community, and the third combined both ideas. The messages that mentioned community were the most effective, a result “consistent broadly with the idea that people are moral actors who don’t care just for themselves, they also care for others,” says Jillian Jordan, an assistant professor at Harvard Business School and a co-author of the study. (The paper is currently being reviewed for publication.)

Another research team is investigating whether gender bias plays a role in how people respond to coronavirus-related information from female governors. An early analysis of the data, which has not yet been published, suggests that male governors are viewed as better pandemic policy communicators and that residents of states with male governors are more likely to comply with public health recommendations, says lead investigator Sebawit Bishu, an assistant professor of public administration and management at the University of Colorado at Denver.

But the scope of the social disruption creates opportunities for experiments that go well beyond the public health crisis and efforts to combat it. For the past few years, for instance, researchers at the University of South Carolina have been following hundreds of schoolchildren to learn more about the connection between having a structured schedule — a preplanned daily routine, with specific required activities — and children’s health. They were probing whether children enrolled in school year-round maintained better dietary, sleep and activity habits than those who had long, potentially idle summer breaks. And then all the schools closed.

At first, the researchers were “really bummed out” that the pandemic had upset their carefully designed study, says Bridget Armstrong, an assistant professor of public health at the University of South Carolina. But then they decided to roll with it. Armstrong and her colleagues mailed Fitbit sleep-and-activity trackers to 74 of the children enrolled in their study, while also collecting data on the kids’ diets and daily schedules. What happened when families were forced to develop new routines on the fly? Did kids become more sedentary? Stay up later? How much did screen time increase? Can consistent mealtimes or regularly scheduled online classes mitigate those effects? When the kids do go back to school, would healthier behaviors return — and how quickly? “This gives us a really great opportunity to peek into: How are habits formed?” Armstrong says.

In her study of dual-career couples, also currently under review, Shockley found that delegating all the child care to one parent, who was also trying to work remotely, resulted in the lowest levels of family cohesion and the highest levels of relationship tension. Yet more than 30 percent of couples chose that option — and in 87 percent of these cases, it was the women who shouldered the parenting load. Strangely, this arrangement also seemed to lead to some of the worst self-reported job performance, not just for women but also for men. “Why were the men performing worse in this group when they weren’t doing any childcare?” Shockley says in an email. One possibility is that the arrangement creates a stressful dynamic at home that offsets the time men otherwise gain.

Meanwhile, researchers at Stanford are tracking whether the sharp economic downturn and the government’s response — which included expanded unemployment benefits and cash payments to millions of Americans — are changing views of a universal basic income, such as the $1,000-per-month “Freedom Dividend” proposed by former Democratic presidential candidate Andrew Yang. They’re focusing in particular on how people with different ideologies perceive such programs. Research is ongoing, but their findings suggest that support for the policy may have ticked up slightly in the early days of the pandemic — notably among self-identified conservatives, who are generally more skeptical of basic income — before falling back to baseline this summer.

Criminal justice reformers have long called for reducing the number of people behind bars. The pandemic, which has hit jails and prisons hard, may provide a chance to study the consequences of doing so. To stem the spread of the virus, a number of correctional facilities have released nonviolent or medically vulnerable inmates; New Jersey, for example, is on track to release about 20 percent of its prison population. Daniel Nagin, a professor of public policy and statistics at Carnegie Mellon University, is tracking changes in the size of America’s correctional populations alongside the rates of various types of crime. “I’ve heard a number of police chiefs say, ‘Oh, you’re letting all these people out of jail, that’s going to increase crime,’ ” Nagin says. “Well, is there any evidence that that’s correct?”

One study of crime rates during the pandemic in 27 cities, by the Council on Criminal Justice, did find that homicides and aggravated assaults rose in May and June — but other offenses, including burglaries and drug crimes, declined between February and June. It will take more time and analysis, however, to determine whether there’s any connection to release programs. After all, it has been a most unusual year, with widespread job loss, personal stress and social unrest, all of which could be affecting criminal activity.

A sign in Brooklyn this spring reminds commuters what they can do to stop the spread of the coronavirus. The pandemic has given some researchers a chance to study different approaches to public health messages. (Wong Maye-E/AP)

Indeed, that’s the question hanging over this research in general: whether so much has changed at once that it will be hard to draw useful, generalizable conclusions. “The principle of experimentation is that you change one thing, and you see what changing that one thing does, so you’re trying to hold everything else constant,” says Michael Sanders, an associate professor of public policy at King’s College London. But in this case, “that’s just not happening. Everything has changed.”

If children develop unhealthy habits, is that because their days are unstructured or because they’re stressed about the deadly pandemic? If productivity plummets, is that because remote work is inherently inefficient or because employees are simultaneously doing their jobs and helping their kids log on to Zoom classes?

What’s more, the pandemic is dynamic, as Harvard’s Jordan and her colleagues learned when they conducted a follow-up to their health-messaging study, a month after the first round. This time, they found that messages that focused on self-protection and those that highlighted community protection did equally well. The discrepancy could stem from the fact that the researchers tweaked the messages slightly (to reflect evolving understanding of the virus). Or, Jordan notes, maybe “people’s psychology has just changed over the course of the pandemic.” Then again, it could just be a sampling error.

The sheer number of papers, and the speed at which they are being churned out, makes some scientists uneasy. “There’s incredible pressure and demand to produce knowledge that’s visible now, and that can lead to errors,” says Matias, the Cornell communication professor. Indeed, many papers are being posted online before undergoing peer review.

Caution is warranted in general, many scientists say — but especially when the research is designed to shape public health policies. In a new paper, also still under review, an international group of psychologists questions whether the behavioral sciences “are mature enough” to guide our response to crises like the pandemic. Well before the arrival of the novel coronavirus, they point out, their field was wrestling with questions about the validity and reliability of its methods and findings.

Even the skeptics acknowledge that most scientists are trying to do the best they can under difficult circumstances. But they urge social scientists to acknowledge the limitations of their research and communicate their findings with nuance and humility. “It seems like every day I’m seeing some bold claim,” says Matias. “I think we’re going to be wrestling with how our field behaved in this moment of crisis for some time.”

Despite the challenges, it’s a moment of tremendous opportunity. New insights into public health communication could save lives not only in this pandemic but in the public health calamities yet to come. And while scholars would have preferred that this particular natural experiment had never happened, they are seizing the chance to learn what they can about human behavior, beliefs and relationships. The resulting knowledge can’t come close to making up for the illness and death we’ve seen in 2020 — but it could help improve our lives in an uncertain future.

We noticed you’re blocking ads!

Keep supporting great journalism by turning off your ad blocker. Or purchase a subscription for unlimited access to real news you can count on.
Unblock ads
Questions about why you are seeing this? Contact us