Being a statistician on twitter in COVID times is seeing people forget about things like variable spread, correlation is not causation, hidden factors, and selection bias, among other things
It's painful to see numbers waved about & really dodgy looking conclusions reached
It's painful to see numbers waved about & really dodgy looking conclusions reached
It's not because numbers increase at same time as event that the event is responsible!
Increase in cases means nothing if increase in tests is not being factored in!
Don't compare 2 sets of numbers from 2 places without checking they are indeed the same!
Increase in cases means nothing if increase in tests is not being factored in!
Don't compare 2 sets of numbers from 2 places without checking they are indeed the same!
(these are just off the top of my head by staring at the timeline. I can probably find more but I have a book to write)
Very gentle reminder: I am perfectly willing to have technical conversations, but I am not here to be explained statistics 101, which I already know.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
