Health Nerd is correct that the percentages cannot be the result of the numbers listed. But he's there is a lot of context that should be understood, and I find this in fully half the papers I read, regardless of other qualities. It happens when...
-
-
Replying to @EduEngineer @GidMK
It happens most often when the staff is small and not well enough trained in number sense to catch where they've changed the underlying patient pool after computing the percentages. Once in a blue moon, such an error is entirely critical to the outcome. Mostly, it's messy.
1 reply 0 retweets 2 likes -
Replying to @EduEngineer @GidMK
Typically, in nations outside the Western pharmasphere, this happens because it's mostly doctors working without a professional statistician, and that does not mean the labor itself was not quality.
2 replies 0 retweets 3 likes -
Replying to @EduEngineer
Lol, that's absolute nonsense. Some of the best research during COVID-19 has come from outside the "Western pharmasphere", most of it includes statisticians and does not have impossible values in the studies
1 reply 0 retweets 1 like -
Replying to @GidMK
I'm sure we can find some of the best papers that have a larger staff and can find all the errors. That's not the point. The point is that saying, "somebody made a number mistake, so all this is invalid" is a poor critique. Just ask for their data, and suggest a correction.
1 reply 0 retweets 1 like -
Replying to @EduEngineer
Normally, impossible values throughout a paper is cause for some concern about whether a study has been done correctly. There are also quite a few other concerns with this particular piece of research
2 replies 0 retweets 1 like -
Replying to @GidMK
You're making "impossible" sound like something it isn't. All that happened here is that the denominators were changed by 1 or 2 because the authors probably realized that a patient was miscategorized before publication. I see that all the time.
2 replies 0 retweets 0 likes -
Replying to @EduEngineer @GidMK
Again, the appropriate course of action is to write them, explain the problem, and if you're truly that concerned, ask for their data to check. It might be that the study was fabricated or something horrible, but I doubt it was worse than what I said.
1 reply 0 retweets 1 like -
Replying to @EduEngineer
We did ask for their data to check. They refused, with a range of excuses that became increasingly more bizarre
1 reply 0 retweets 0 likes -
Health Nerd Retweeted Kyle Sheldrick
Health Nerd added,
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.