Over 500 schools across the country have closed due to the new coronavirus, according to a running tally by Education Week. That’s a tiny fraction of America’s 130,000-plus K-12 schools, but the question of whether to close is one that more are likely to face — and that anxious families will be monitoring closely.

So should schools close?

Right now, the Centers for Disease Control and Prevention is not issuing a blanket directive. States and districts are taking it case by case: In New York, for instance, the governor said that schools would close for at least 24 hours if any student or staff tests positive.

Past experiences make clear that there are no easy answers. Research has found that such closures have helped contain outbreaks. But closing schools would also present significant challenges for families — especially low-income families. And the best decisions likely depend on both local context and facts about the virus, like its fatality rate, that aren’t yet fully known.

Here’s what we do know about the pros and cons of closing schools.

🔗Closing schools in the past slowed the spread of illness, and thus saved lives.

The case for school closures is straightforward.

“Schools are community gathering places where large numbers of people are in proximity to one another and respiratory infections can easily spread among young people and adults alike,” wrote Howard Markel, director of the Center for the History of Medicine at the University of Michigan, in the New York Times last week. “Shutting them down can be a key part of slowing the spread of easily transmissible viruses so that hospitals are not overrun with sick people.”

Markel cites his own study on the Spanish flu of 1918 and 1919. He and others found that cities that closed schools and banned public gatherings for about a month saw many fewer deaths. Doing so earlier and for longer periods seemed to help.

Similarly, a study of Japan’s response to the H1N1 flu in 2009 showed that closing schools limited its spread. And a 2013 overview of research found some evidence that school closures can help stop the spread of influenza, including to adults. Children appear to be less affected by COVID-19, the disease caused by the new coronavirus, but they may still be playing a role in spreading it.

But closing schools after a virus has already spread is unlikely to make much of a difference. Another study found that school closures in Michigan due to H1N1 were ineffective because they happened too late.

Markel says the need for closures will depend on how deadly the virus turns out to be. However, he points out, “this kind of data might not be available for analysis until Covid-19 has already spread widely. So for now, policies to contain it should be influenced by the adage ‘better safe than sorry.’”

🔗Closing schools comes with substantial economic and social costs

If schools close, students will not be in school learning. Their parents may have to take off work to care for them. And services that schools provide — like free meals or after-school programming — would be halted.

That has real costs, and would likely burden low-income families most. At Chicago’s Vaughn Occupational High School, which is now closed after a school aide tested positive, families are organizing a food pantry and scrambling for child care.

“Local officials must balance the potential public health benefits of closing schools with the other burdens and economic costs placed on a community,” John Bailey, who in 2005 worked on a White House team that helped plan for pandemics, wrote in Education Next. “As schools close, parents will have to find childcare or stay home from work, creating challenges for employers, including, most importantly, healthcare providers.”

Surveys in Pennsylvania and New York City showed that school closings of about a week due to H1N1 led to about one in five households having an adult miss work.

A 2009 study estimated that closing every American school for a month would lead to more missed work and reduce GDP by between 0.1% and 0.3% — or tens of billions of dollars. (This does not account for the likelihood that many more Americans today can effectively work from home.) Some of those forced to stay home with their kids are likely healthcare workers, the study also points out, which could make combating the virus even more difficult.

Indeed, a recent study estimated that 15% of health care providers have children but don’t have another family member in the same household who could provide childcare. That could keep some of them from going into work and helping patients, potentially counteracting the public health benefits of school closures.

🔗Short closures probably won’t be that harmful academically, but longer spells could create real problems.

Missing school could also likely affect student learning. But research has shown the effects are blunted when all students are off together.

Older research found evidence that bad weather, resulting in more snow days, led to lower test scores. But other research has questioned that conclusion. A Massachusetts study found that when individual students missed school due to snowy weather, they learned less, but when schools closed due to snow, there were no clear negative effects. That might be because it’s easy for teachers to put a lesson on hold but harder to catch up students who miss class time and fall behind.

These results suggest that a short spell of closed school would not substantially harm students’ academic progress — and in fact might even be preferable to keeping school open on days where many kids might be absent. (Keep in mind, though, that this research does not account for the other consequences of closures on students, like loss of subsidized breakfast or lunch.)

Longer closures are likely to come with more harmful consequences. That is backed up by research on lengthy school closures in other countries. Studies in Canada and Belgium, for instance, have found that weeks-long teachers’ strikes led to declines in student test scores.

In one unusual example, researchers found that missing an average of 88 days of school over the course of several years, due to frequent teachers’ strike in Argentina, hurt students in the long run: between ages 30 and 40, they earned 2 to 3 percent less and were more likely to be unemployed.

These studies also tend to show that students from lower-income families are the most harmed by missing school.

🔗It’s far from clear that virtual learning will blunt the impact.

How closures affect students will depend on many factors, including their length, whether the missed days are made up, and whether districts plan for alternative instruction.

Some districts — and tech companies — are optimistic that “remote learning” can take the place of regular instruction. But there are reasons to be skeptical that meaningful learning will occur if schools are closed.

For one, many districts may lack the technology infrastructure and the expertise to effectively deliver online instruction. Second, students may not be in a good position to receive such instruction. They may have responsibilities at home — like caring for younger siblings — or lack internet access. Even if virtual approaches are effective for some, they might also exacerbate existing inequities if they prove impractical for students with certain disabilities, for example.

And finally, research has repeatedly found that fully virtual classes are much less effective than traditional face-to-face instruction.

“If they have not already laid the groundwork to deliver courses online, the notion that they could outsource that — it’s fanciful,” Doug Levin, an education technology consultant previously told Chalkbeat. “To do it well, this is something that takes a good deal of time and training.”

Kalyn Belsha contributed reporting.

Correction: A previous version of this story described a study on the effects of missing school in Argentina as based on a single teachers’ strike. In fact, the study examined the effects of multiple strikes, which led students to miss an average of 88 days.