More potential fraud in the ivermectin literature - my new piece builds on work by @K_Sheldrick on Brazilian data that looks very strangehttps://gidmk.medium.com/is-ivermectin-for-covid-19-based-on-fraudulent-research-part-2-a4475523b4e4 …
-
Show this thread
-
Replying to @GidMK @K_Sheldrick
fyi weight (and age) heaping is not unusual in real data, especially in LMICs. Paradoxically, I'd be more suspicious of an LMIC dataset with weights that *did* neatly follow Benford's law. Similarly, I'd expect spikes at 5, 7, 10, 14, 20, 21 days for days since symptom onset.
3 replies 0 retweets 3 likes -
You haven't shown the histogram for height trailing digits, which gives me an opporunity for a blinded guess! 1. Lots more 0s and 5s then expected 2. Then a slight preference for even digits maybe?
1 reply 0 retweets 0 likes -
-
Nope, for heights "5" is equal 4th most common, "0" is the second least common.pic.twitter.com/75K9ZAiHlx
3 replies 0 retweets 2 likes
Yeh the very distinct lack of 0s is one of the weirder things about the data. You'd expect way more 0s and 5s if this was just rounding or guesses! Also, ofc, you don't round to 1dp for height and weight very often...
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.