Discussion about this post

User's avatar
Dr Mike Yeadon's avatar

The specific issue here is that an inability to control for numerous variables impacting on reporting symptoms (being the endpoint), variables with likely large effects relative to the claimed effect of masking, makes the claim that the tiny change is due to masks completely unsupportable.

In other words, the uncorrected sources of noise must completely swamp the claimed signal. This the claimed signal isn’t established to exist, let alone that its a signal due to masking.

I find that, worse even than all the methodological errors that prevented any possible useful outcome right off the bat is this problem, and I see it t all the time:

You don’t invalidate negative results from well conducted empirical studies by performing another study, even if well conducted, which produces a result you favour.

There needs to be a reason to discard those prior studies which show conclusively that masks have no effect on respiratory virus transmission. If you don’t have such a reason, your behaviour is like throwing a dice over and again until you get a six, ignoring all the other times you didn’t get a six.

Expand full comment
Nuance Wire's avatar

1) They. Lie.

2) Always.

The end.

Expand full comment
145 more comments...

No posts