dhsilv2 wrote:
And don't new yorkers on average walk WAY more than most americans AND aren't they screwed younger? NYC that is, not the state.That means in areas....say the south, we'd be looking at worse outcomes?
I'm not sure about how New Yorkers compare to the rest of the States. You could also argue that outcomes in New York City have been worse, because the hospital system was definitely at it's limit or argue that pollution in a big city increases the IFR and so on.
I wouldn't try to read too much into this result. It's in line with some of the estimates that we were working with for a while now. The stuff about 0.1% - 0.3% that some people were pushing recently looks even more questionable now. If someone was looking at case fatality rates, they can stop doing that now.
dhsilv2 wrote:
3000 is WAY in excess of what's needed in a simple random sample to have an EXTREMELY high accuracy. That said, if they got this from people in grocery stores, it might be skewed high as that might be a higher risk group on average. But none the less 3,000 is a MASSIVE sample that should be within 3% +/- assuming it's properly conducted.
In short, if 3k doesn't make you go "holy crap that's a huge sample size", you might need to go read up on stats.
Most people get groceries, right? Sure, you are oversampling people that go shopping more frequently, but people that go for groceries less frequently probably include doctors, nurses and first responders that work crazy hours and are a super high risk group. Don't think it's clear that there would be a bias in one direction or the other.