Nobody.
There’s a great little old book called *How to lie with Statistics*.
It was required reading for one of my college courses, and while dated in its examples, i the point it makes is still valid.
You can *prove* anything you want using a *study*, especially when you set it up the way you want, define the terms as you wish, and then repeat it endlessly until you get the desired results you want.
Then you publish it, and voila‘, you can now claim *A study proves.........*
I would highly recommend reading it.