Taking the average of every place in the country with a population under 10,000 does not strike me as being even close to "as biased... as you can get" toward my point. You're still discussing the mean, while the original claim was about the median. If crime is indeed highly clumpy, as I've proposed, then it's probably highly clumpy in communities of under 10,000 people too. The mean could therefore easily be high while the median remains low.
And yes, a lot of it is handwaving and speculation. The data I'd need just doesn't appear to exist, since you'd have to take it down to the neighborhood level. That said, I think it's superior to discuss without data, while admitting that the data isn't there, than tho discuss with data that doesn't actually say what it's claimed to say.
Regarding healthcare, your first link does not mention inferior outcomes at all. It does not state, but heavily implies, that our outcomes are not significantly better or worse. I briefly searched around that site for other articles but couldn't find anything that discusses the health care system in isolation from the confounding factors. And there is no way I'm going to believe that obesity is not a significant factor here without something to back it up, given that obesity in the US is a factor in about 1/5th of all deaths.
For more on my reasoning, see my reply elsewhere:
https://news.ycombinator.com/item?id=7024053
And yes, a lot of it is handwaving and speculation. The data I'd need just doesn't appear to exist, since you'd have to take it down to the neighborhood level. That said, I think it's superior to discuss without data, while admitting that the data isn't there, than tho discuss with data that doesn't actually say what it's claimed to say.
Regarding healthcare, your first link does not mention inferior outcomes at all. It does not state, but heavily implies, that our outcomes are not significantly better or worse. I briefly searched around that site for other articles but couldn't find anything that discusses the health care system in isolation from the confounding factors. And there is no way I'm going to believe that obesity is not a significant factor here without something to back it up, given that obesity in the US is a factor in about 1/5th of all deaths.