>>13876696The problem is bias, which is the part people work very hard to eliminate in clinical studies, and many of the bad scientists produce shitty studies anyways because they don't follow the method carefully.
When you collect random comments, you have >99% noise and very little signal, so your error bars are through the fucking roof.
And that's without considering the biais of the person who does the screenshot. They only include the comments that already agree with their worldview, so you get a really warped picture.
Like when journalists takes a zoomed in photo of a burning twig at a protest to make it look like the whole street is on fire.
People are shockingly bad at self-reporting anything medical.
They did a study on Long Covid, and even though all of these people listen many symptoms, most of them had no actual link to having had Covid in the past.
Whatever it is they had, they were feeling some symptoms, but their immune system never saw any rona.
And then you have people who just go on the Internet and tell lies, because politics has become part of their identity and anything is valid if it protects your worldview.
You know, top shelf high IQ stuff. People you can trust.
