Some fun for everyone - find the errors in this table of results from an ivermectin study Bonus points if you can identify the true numberspic.twitter.com/IbxiFU4Sbb
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
The winner here - most of the percentages in the follow-up columns cannot possibly be true, given the sample size. Either the authors had dropouts that they didn't mention, or something else went wrong!https://twitter.com/undafiend/status/1435466075550924802?s=20 …
well, for example 144 * .05 = 7.2 and you can't have 7.2 people with diarrhea
You are the winner! Here is your prize 
Ivermectin made 0% difference in the treatment of COVID-19
why are there 2 sets of p-values? Are they comparing treatment and control symptom frequency, or pre vs post enrollment within each group? Also, how the hell do they get p-values that tiny when they're comparing %s, in what is presumably a presence/absence non-parametric test?
None of the p-values are correct, but it's hard to figure that out just from the table. There's a more basic error there!
I’m proud of myself for spotting this!! But I was too afraid to say it. I’ve never been great at maths but have been paying attention to your threads and from others and have been able to understand far more than I did pre-Twitter and pre-ivermectin 
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.