95% accuracy can be misleading though.
Say you test 100 people, and 10 have COVID.
90 people test negitive and don't have COVID
5 people test positive and do have COVID
5 people test negitive and do have COVID
Amongst those that have it, half are testing negitive! Which makes the 95% accuracy very poor indeed!
In probability terms its a confusion between P(A|B) and P(A).
I'm not saying this is what is happening, but that we just have to be careful to not get excited by high numbers.
Originally Posted by: Quantum