Hacker News new | ask | show | jobs
by kspacewalk2 2207 days ago
That is exactly right - you can't do any cross-country comparisons, because the data simply is not comparable across countries. It can be used to a limited degree for tracking the epidemic within a country over time, and even that's got plenty of caveats. But cross-country comparison is a non-starter. Here's an incomplete list of giant confounders:

- for cases: how widespread was the testing? How quickly did it ramp up (e.g. how many tests per 1M in relation to 1000th case)?

- for deaths: what counts as a COVID death? Is death with COVID = death from COVID?

- how soon after the onset of the epidemic in the particular country were lockdown measures enacted? How long did they last? How stringent were they in reality vs on paper?

1 comments

Well I think you can compare, but with caveat you have to compare amount of testing, and various other situations. I only see people who say "you can't compare" to be those who want to minimize their own problems. It's endemic to the US where multiple states have been caught reducing the numbers, firing the people who organize them and prohibiting them from talking about it.
So wait, if they are reducing the numbers, we'll then turn around and use them for comparison?

There are many countries who figured out how to play this numbers game. Russia, for example, faithfully reports any and all cases of COVID and simultaneously employs every trick in the book to minimize the number of deaths. They have already and will continue to win propaganda points from people who then use these meaningless death numbers to compare to another country's far more accurate death numbers and proceed to draw garbage conclusions from garbage data.