3 ms·
> No, but really, the flaw behind these master controlling group theories is that there are so many of them that would have to compete that its the same as none
by thaumasiotes 5y ago
> No, but really, the flaw behind these master controlling group theories is that there are so many of them that would have to compete that its the same as none existing.
Andrew Gelman has made this same observation, which he calls "the piranha problem", in the context of sociology and psychology papers. The typical paper identifies a random-looking stimulus that nevertheless has a large effect on some behavior like voting or accepting a romantic partner. And the problem is, there are so many of these papers, all showing large effects, that it's impossible for any of the effects to be reliable, because even if the effect a particular paper identifies is real, there are so many other effects influencing the same outcome that the effect described by the particular paper cannot have any meaningful influence.