Wish I could edit my comment to add a statement that very sadly has been shown to have been needed. age is but one example out of many different conditions that would cause more frail people to self select. Stop replying to say one can zoom in and see age segments.
And please note that the main point of the article is an ELI5 of why absolute death numbers is misleading, but that eli5 doesn’t mention the sample bias which would be a good “next layer” to mention at least in the conclusion.
So yeah it’s all very fine that as so many people like to reply that a reader can dig around and find age groups assuming they already know about sample bias and thus are trying to control for those things. But oh wait one little problem. This article is explaining why one needs to look at rates not absolute numbers. If someone is aware of sample bias why would they need this article? Are you telling me people who need the first eli5 are going to know sample bias?!?!?
And please note that the main point of the article is an ELI5 of why absolute death numbers is misleading, but that eli5 doesn’t mention the sample bias which would be a good “next layer” to mention at least in the conclusion.
So yeah it’s all very fine that as so many people like to reply that a reader can dig around and find age groups assuming they already know about sample bias and thus are trying to control for those things. But oh wait one little problem. This article is explaining why one needs to look at rates not absolute numbers. If someone is aware of sample bias why would they need this article? Are you telling me people who need the first eli5 are going to know sample bias?!?!?