Twice as likely to test positive if tested, but also nearly three times less to get tested. The way Axios presents these data is misleading. The meaning is not obvious; many COVID cases were never tested/confirmed, so I wouldn’t infer mask wearers *more* likely to get COVID.
https://twitter.com/peterbakernyt/status/1401870525740195849
This piece needs to be fixed, the text doesn’t match the chart. 
Of course there’s all sorts of other confounds. Maybe one reason people wore masks regularly outside the home is they had a higher baseline level of COVID exposure. Mask wearers also may have been more careful in general. You can’t really infer from observations like this.
Also, mask wearers getting tested 3x as often likely means they got a lot more tests when not symptomatic, so a lower positivity rate is to be expected even with equal risk of getting COVID. etc.
People seem still confused by this. Contrary to headline, Axios data show people who say they always wore masks were *more* likely to have gotten a positive COVID test result than those who never did, and much more likely to get tested. These facts are of course related. But...
... this doesn’t tell us anything about the underlying rate of COVID cases. Test positivity among those tested doesn’t tell us much because the tests are not administered at random and conscientious people are more likely to get tested when asymptomatic.
People whose reported mask behavior differs also differs in other ways. They may be at more risk of exposure, or they may be broadly more conscientious. Basically it’s a data point that says nothing either way, but the story draws a strong inference anyway.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
