Loading…
29
TIL the CDC's mask study from 2022 actually held up when I checked the raw data
I spent most of last year calling that mask effectiveness paper junk. The headlines were a mess. Then last week I pulled the actual dataset from the CDC site, 14,000 participants, and ran a simple chi-square in Excel. The p-value was 0.003. Not great by modern standards but not the fake result I assumed. What got me was the control group had way more asymptomatic cases they didn't test for. Once I adjusted for that, the effect size doubled. I still think the media overhyped it, but the core finding is real. Has anyone else found a study they dismissed that survived a closer look?
1 comments
Log in to join the discussion
Log In1 Comment
leor861mo ago
Man, that's the exact kind of thing that makes me double check my own stubbornness. I once spent a whole week trashing a nutrition study about eggs and cholesterol, then finally looked at the data and realized they controlled for smoking and exercise, which I'd totally ignored. My bad. The effect was smaller than the headlines said, but it was real. Good on you for actually running the numbers though, most of us just yell at the summary. I'm with you on the media part though, they always make it sound way more dramatic than it is.
5