47 ms·
The people that want to make criticisms about how this association is not causal can start reading halfway through the first column of page 1252. The authors ar
by munin 9y ago
The people that want to make criticisms about how this association is not causal can start reading halfway through the first column of page 1252. The authors are careful to phrase their finding as an association, not a causal relationship.
They discuss the confounds you might think of (sleep a lot because of undetected major illness? depression? etc) and their efforts at controlling for them.
They then acknowledge that there are still enough unknowns that they can't make a causal link, and that there should be more in depth and controlled experiments to investigate this phenomenon.
- lucozade 9y agoI agree. I think publication of studies/experiments that don't find anything particularly interesting should be encouraged not condemned. It's much more of a problem that there's a bias to publishing to a headline. I'm, slightly irrationally, more inclined to believe this than a lot of papers.
- exclusiv 9y agoAgreed. What can we really say is causal? Not that much in the grand scheme of things.
- andybak 9y agoI suppose one could argue that in an absolute philosophical sense there is no causation that we can have knowledge of, only differing amounts of correlation. It would be an extreme position to take but - like solipsism - a difficult one to argue against. (Anyone with an actual background in epistemology care to help me out here?)
- anon1253 9y agoHume argues this in “enquiry concerning human understanding” https://plato.stanford.edu/entries/hume/#Cau https://plato.stanford.edu/entries/hume/#Cau
- arca_vorago 9y agoI always considered causality less as a binary thing that was always 0 until $somethreshold was crossed, and more of a bayesian (or other) probability model. Since in many cases we don't have the evidence for a deductive conclusion, we must often rely on inductive logic, even while still seeking the evidence until we can switch to deductive.
- Rhapso 9y agohttps://en.wikipedia.org/wiki/Granger_causality https://en.wikipedia.org/wiki/Granger_causality