The U.S. has launched the largest experiment in children’s education in our history. Will we learn from it?

Across the country, the 130,000 K-12 schools educating more than 50 million children are embarking on totally new ways of teaching. While a small percentage have returned to full-time in-person school, the vast majority have not. Kindergarten and first grade students are trying to learn the foundational skill of reading by sitting in front of a screen displaying their teacher and the other 15 to 25 children in their class. Others are enrolled in “hybrid” programs where they are on-site for some fraction of the week and remote for the rest.

We are pediatricians and scientists, deeply versed in the practice of “evidence-based medicine,” which maintains that how we diagnose and treat patients should be based on scientific evidence. There are, of course, large gaps in our knowledge about what works and what doesn’t, but new treatments are not given at scale until they have been shown to be effective in large, controlled studies that elucidate the benefits and the risks.

The consequences of failing to deploy an evidence-based medicine approach have been evinced during the pandemic. Claims that chloroquine and hydroxychloroquine were effective treatments for COVID-19 were subsequently debunked with carefully done randomized controlled trials. As a result, the FDA revoked its emergency use authorization for these drugs to treat the infection. And while we desperately await a vaccine against SARS-CoV-2, it appropriately won’t be given to the millions seeking it until large-scale Phase 3 randomized controlled trials have been completed and carefully vetted.

To accomplish this, the federal government and industry are pouring tens of billions of dollars into “Operation Warp Speed” to develop a demonstrably effective and safe vaccine as soon as possible. The same type of massive scientific effort is needed to determine the best way to school our children in the face of the pandemic. 

Most experts agree that children will not be able to attend school in a normal fashion until the beginning of the 2021-2022 school year at the earliest. The vast number of variations in how schools are approaching the education of the nation’s children represents an ideal opportunity to study which approaches are effective, which are worthless and which are actually harmful. The consequences of inadequate education in terms of health are well described. Failure to graduate high school is associated with 10 years less life. And the single best predictor of high school graduation is third-grade reading level. Can we afford to deploy untested approaches and casually await to assess their impact? Are we prepared to do the equivalent of giving our children hydroxychloroquine without considering the consequences? We owe it to our children to figure out the best way to teach them and do it as soon as possible.

We believe that the federal government should immediately embark on an evaluation of this massive experiment currently underway. While randomizing children to different methods of instruction would be ideal, as it is for testing of drugs and vaccines, it is both infeasible and unnecessary in this case. Using sophisticated statistical methods that have been developed, we can take advantage of the tremendous variation that currently exists to determine in short order which teaching models are best and deploy them widely. However, the evaluation must start today to prospectively and robustly capture what teachers and schools are doing, how many hours of instruction and in what mode children are engaging it, how much children are attending online sessions and which ones they attend, and most important, how much they are learning. Enrolling families will be relatively easy. How many parents would not want to know what the best way to teach their child in these unprecedented times? 

We cannot afford nor do we need to wait until children return to school in-person to see what the effects of distance learning will be. Learning is a continuous process that can be assessed far more rapidly than waiting until the end of the school year so that we should take corrective actions as soon as possible. The stakes are simply too high to conduct this experiment on our nation’s children without finding out how we can best serve them. The cost of this evaluation is minor, compared not only to the trillions of dollars spent to support the country since the pandemic began, but more importantly to the minds of the 50 million children at risk.