Based on what gold standard?
Right. Sampling a lower acuity population, like the ONS random sample, would give you a lower PPV, because the population prevalence is low. But the pillar I/II samples have a ~higher~ acuity, so the PPV would be HIGHER (i.e. lower ratio of false/true positives
-
-
Also, that blog doesn't correct for test characteristics properly. The Rogen-Gladen estimator is a better way to do this (although of course still imperfect - Bayesian inference much better if you've got the time)
-
This is the point: Sample 340062: false-positive rates of 1.4% if no virus is present Sample 340060 HCoV OC43: false-positive rates of 2.2% (harmless cold corona virus) Sample 340065 HCoV 229E: false-positive rates of 7.6% if present (harmless cold corona virus)
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.