11/n There are dozens more such examples. While in some countries masking was temporally related to declines in case numbers, this is by no means true of every place in the world
I'm very confused by Table 1 - what you'd expect is a breakdown of country-by-country (or perhaps region) with percentages and variables. Instead, we get the results of some very odd logistic regressions - making it impossible to assess any of these variablespic.twitter.com/0KCYwPefNT
-
-
All of the duration variables are also calculated in a very strange way. In many cases the first case/death was on recent arrival to the country, so this is likely to skew their analysis quite a bit. Rather that they just used the date of the first case being identifiedpic.twitter.com/sKHnAB92g0
-
In terms of their findings with regards to masks, this makes me very hesitant to accept any of the conclusions - they are all based on a 'duration' variable which is opaque and may not really reflect reality
- Show replies
New conversation -
-
-
A statistician asked us to add Table 1 column with odds ratios. You can ignore it. Paper conclusions not based on that column. Purpose of Table 1 is to show characteristics of low and high mortality countries (focus on the other columns). https://www.researchgate.net/publication/342198360_Association_of_country-wide_coronavirus_mortality_with_demographics_testing_lockdowns_and_public_wearing_of_masks_Update_June_15_2020?uploadChannel=invalid …pic.twitter.com/0FkBMuEUFu
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.