Quote:
Originally Posted by 'Hamas' Jenkins
I can tell you why they're wrong:
They implemented selection bias to draw their conclusions. Those most likely to go to a coronavirus clinic are those most likely to test positive for COVID. Without random sampling of the public (which is done for other diseases, mind you), inferences I draw from that information are inherently flawed. They claimed that because 6.6% of their patients were positive for COVID, that 12% of the population in California was positive.
Let's say that I wanted to test for the prevalence of black lung in the country as a whole. Should I test people in the Pacific Northwest, or people in West Virginia? What are the flaws of doing one vs. the other? In reality, I should randomly sample enough of the population for a representative sample, not just one group.
If I test nothing but coal miners, what conclusions might I draw from the severity of black lung that are incorrect?
We also know from hard data in New York that the IFR is far above the lower bounds of their conclusions anyway, because more than 0.1% of the entire population in NYC has already died of COVID-19, and that's with far less than 100% of the city being infected.
I can tell you: they are 100% incorrect.
|
I follow, and thanks for the spelling it out like that.
So, following this line of thought, is it necessary to keep the entire country on lockdown or should we allow cities/states to open according to the numbers they've individually collected in regards to their regions? That might be an entirely too basic of a question and feel free to flame me if you think I need it. I'm just trying to find the light at the end of the tunnel, here.