Secretary of Education Miguel Cardona recently attempted to make the case for masking in schools by citing a number of different studies that purported to show their positive effects in a series of tweets. However, his use of one study out of Wisconsin shows either a willful misrepresentation of the data, or an ineptitude in interpreting research that should be concerning to families impacted by U.S. Department of Education policy.
The study in question looked at transmission rates of COVID-19 in rural Wood County, Wisconsin. They follow 17 schools in the county, all of which required masking, and provide a description of the transmission rates that were found. Cardona claimed that the study found a “37% lower incidence of COVID than the surrounding community,” and claimed that this was scientific evidence that masking of students was effective. But this study makes no such claim. The 37% number found in the study is merely a comparison of COVID transmission in the school with transmission in the surrounding community—there was no control group. A control group is vital to a comprehensive study because it allows the scientists to exclude outside factors. There are plenty of other explanations for why transmission was lower in the school. The preponderance of the evidence suggests that kids are less likely to transmit COVID-19 than adults. But whatever the truth, the study cited is not evidence one way or the other.
The misrepresentation on the part of Cardona was so great that the author of the study in question took to Twitter to correct the record:
“Secretary Cardona, I was the senior author of this study. Our study is not able to give any information about the role masks played in the observed low in-school transmission rates. We had no control group so don’t know if the rate would have been different without masks.”
The tweet remains up as of this writing, despite attempts by the author of the study and many others to correct the record, and the Secretary has not explained the reasoning behind his misrepresentation. There are problems with the other studies he cites as well. None of the other studies cited has a control group. Even worse, as pointed out by American Federation for Children Director of Research, Corey DeAngelis, the fourth study cited is purely a simulation—meaning that it is guaranteed to find an effect of masking if an assumption that it works is included in the model.