A new national effort asks K-12 schools to voluntarily — and anonymously — report their confirmed and suspected coronavirus cases, along with the safety strategies they’re using.

Opening schools safely in person is seen as key to restarting the economy and recovering the learning loss that has fallen most heavily on marginalized groups of students. There are also many fears associated with reopening — of severe illness among vulnerable staff and family members, and of stoking broader outbreaks, as seems to have happened where colleges have reopened in person.

The COVID-19 School Response Dashboard, which NPR is reporting on exclusively, was created with the help of several national education organizations. Right now it shows an average of 230 cases per 100,000 students, and 490 per 100,000 staff members, in the first two weeks of September. The responses come from public, private and charter schools in 47 states, serving roughly 200,000 students both in person and online, as of Tuesday, Sept. 22.

As of right now, this sample is a tiny fraction of the more than 56 million K-12 students in the United States. But the dashboard will be continuously updated, and the number of schools participating is expected to grow.

Emily Oster, an economist at Brown University, is spearheading the effort. She is known for her popular writing on data and science literacy for parents. She said she got involved in independent data collection on coronavirus cases in schools because “other people weren’t doing it.” Eventually, the School Superintendents Association, the National Association of Elementary School Principals, the National Association of Secondary School Principals and groups representing charter and independent schools joined forces to urge school leaders to contribute.

Oster says the key difference with this dashboard, as opposed to other data collection efforts and some media reports, is that it highlights not just raw case numbers but also infection rates — cases as a percentage of total in-person attendance for students and staff members. Plus, it provides some of the broader context, for example: Just how are schools opening? What kinds of hybrid policies do schools have? What kinds of mitigation factors are schools undertaking?

Oster hopes that the number of schools participating will grow significantly, from hundreds to thousands of schools, and that the data will give insights into trends over time. Eventually, she hopes, the information will be able to “say something like, ‘Which of these mitigation factors are associated with lower infection rates in schools?’ — which I think will help schools that are still closed try to open safely.”

“I think it is a really worthwhile and good effort,” said Danielle Zerr, a pediatrician at Seattle Children’s Hospital and a professor of epidemiology at the University of Washington, who is not involved in the dashboard. But, she said, the project is useful only relatively speaking, in the context of the general lack of a robust national testing, tracing or data-collection effort for schools. “There’s nothing systematic, that I’m aware of anyway, to really evaluate how we bring children back safely.”

Here’s a closer look at some key metrics within the database:

Infection rate among students

Schools that have brought students back in person reported a total of 230 cases among just under 100,000 students who attended school in person during the two-week period covered by the dashboard.

According to guidance from the Centers for Disease Control and Prevention on school reopening, 230 cases per 100,000 over two weeks falls into the red or “highest risk” category.

If expressed as a daily case rate, that would be 16.4 cases per 100,000 people per day, which falls in the “orange” or next-to-highest risk level category set by the Harvard Global Health Institute and the Brown University School of Public Health.

It’s important to underline, though, that just 77 of these 230 cases were confirmed by coronavirus tests — the bulk are “suspected” cases, which could be run-of-the-mill colds or fevers.

The lack of readily available testing is a huge flaw in the myriad plans around the country for bringing kids back to school, says Zerr. “If this country was interested in putting the right kind of resource behind such an effort, we would have testing available, we would make it very easy. We would require testing before bringing everybody back to school, and there would be regular testing.”

Infection rate among staff

Schools in the database reported a raw total of approximately 130 cases among nearly 40,000 staff members over the two-week period covered by the dashboard.

That rate is in the orange, or “higher-risk” category, according to the CDC school guidelines.

As a daily case rate, that would be 49 cases per 100,000 people per day, which is in the “red,” or highest, risk level category set by the Harvard Global Health Institute and the Brown University School of Public Health.

Again, the case numbers include just 44 confirmed cases and twice as many suspected infections.

About half of schools responding to the survey were running in person at reduced capacity to allow for some social distancing. About 1 in 4 were running in person at full capacity. The rest were remote only.

Interestingly, schools with remote-only learning plans reported an infection rate among staff similar to that of schools running at full capacity. This could indicate cases brought in from the community rather than contracted at school, Zerr suggested. On the other hand, Oster noted, in some places teachers are being asked to teach remotely from their classrooms and infections may be passed among adults in a building.

Mitigation strategies

The data collection asks schools that have brought students back in person which policies they have put in place to stop the virus’s spread. One of the most common policies, wearing masks, is considered very effective by experts. Yet fewer than half of schools say they’re doing social distancing, which is another key strategy touted by experts.

By tracking the reported use of these strategies along with reported infections, says Zerr, over time you could potentially learn which strategies are most effective.

However, Zerr says that self-reporting isn’t the best way to find out what schools and students are actually doing. Are masks worn by all students and staff members all day, for example, or do they droop to the chin when kids are changing classes? “If you were to do things robustly, you’d have boots on the ground going in there and assessing to what degree any of these mitigation strategies are actually happening.”

And there’s an additional problem: Schools that are most zealous about mitigation might also have more resources to devote to testing, which might turn up more cases, making them look worse in the database.

Traditional public schools report less use of safety strategies

No matter what mitigation strategy you look at — masking, distancing, temperature checks — private and charter schools reported following those measures at much higher levels than traditional public schools.

This may be a function of having more resources, especially when it comes to options like holding classes outdoors. Or it may be that schools of choice have more leeway when it comes to imposing these requirements on students.