TIT STARTS of the school year normally brings a sense of relief to most parents. But with cases of covid-19 in America reaching levels last seen in February and the highly infectious Delta variant sweeping the country, the start of the term has been met with fear instead. While the benefits of in-person schooling are clear – children learn best at their desks, vulnerable people are less likely to be left behind, and parents are able to focus on their own work – there were concerns about mixing in playgrounds and crowded classrooms. would increase the spread of the virus.
Enjoy more audio and podcasts on ios Where Android.
Across the country, back-to-school dates vary enormously. Children from some school districts in Texas returned to class in mid-July, while New York’s 1 million students only returned last week. This variation creates an opportunity to test how the start of the trimester influenced the cases of covid-19. So The Economist builds a statistical model to do just that.
First, we calculated the average case rates for each county for each week from mid-June, before school returned, until now. We compared the case rate in each county with the average within the state. This allows the model to monitor overall trends in the number of cases as well as other factors, such as mask warrants or super-spreading events, that could make the virus more or less prevalent in a state. We then looked at the effect of different return-to-school dates, to see if counties that returned earlier had higher case rates than the rest of the state.
The results are clear. In the weeks after the start of the quarter, there were more cases of covid-19 in a given county than expected. Even taking into account demographic factors such as age, race, income, education and politics, going back to school increased the case rate (see graph). On average, for each additional week in office, the increase in the number of cases was about the same as the effect of a one percentage point increase in Donald Trump’s vote share in 2020 (counties pro-Trump tend to have higher covid-19 rates).
The effect was not the same everywhere, however. According to our model, in addition to reducing the overall number of cases, the county’s vaccination rate played an important role in influencing what happened after schools returned. In counties where many people were bitten, the start of the quarter had little effect on the spread of the virus. In counties with the lowest vaccination rates, cases increased after returning from schools.
It should be noted that across America very few children have been vaccinated, so the bites themselves likely have little impact in preventing the spread in classrooms. In counties where more people are vaccinated, infections originating from inside school doors may not escape as easily into the community. In addition, schools in the most affected areas may also take more precautions. They could, for example, impose more strictly the wearing of a mask or social distancing.
States set their own rules for controlling covid-19 in schools. More than a dozen states are mandating face covering in public schools, and nine require teachers and staff to be vaccinated or undergo weekly tests, including Washington state and Oregon which have made the compulsory vaccination for teachers. On September 9, the Los Angeles School Board voted unanimously to require vaccines for students aged 12 and older. On the other end of the spectrum, several conservative states have tried to ban schools from enforcing the wearing of masks.
Our model cannot detect the effects of these state-level interventions. However, numerous other research studies conducted across Europe and America have found that wearing a mask and social distancing helps prevent the spread of covid-19 in schools. With these measures in place, schools were not the most likely sources of infections for children. And if a child caught the virus, it tended not to spread to his classmates. However, the situation looks quite different when those measures are relaxed as the virus spreads, as has been the case across much of the country.
With proper classroom precautions and high levels of community immunization, the start of the new term shouldn’t have been so scary. Our results add to the growing consensus that in-person schooling does not necessarily have to be accompanied by increasing cases of covid-19. But as some states continue to put politics ahead of public safety, it likely will. ■
All of our pandemic and vaccine related stories can be found on our coronavirus hub. You can also find trackers showing the global vaccine rollout, excess deaths by country, and the spread of the virus across Europe.
This article appeared in the United States section of the print edition under the headline “Safety by the Numbers”