2 ms·
Just like calling every person who can whip up some html and JavaScript an "Engineer" is bad for the reputation of Engineering as a whole, so calling everything
by RcouF1uZ4gsC 10y ago
Just like calling every person who can whip up some html and JavaScript an "Engineer" is bad for the reputation of Engineering as a whole, so calling everything that has statistics and is peer reviewed "science" is bad for the reputation of science. I think a lot of damage to science has come from psychology and "social sciences" in corroboration with the popular press when one-off studies are popularized with supposedly profound implications for our understanding of how humans work, only to be contradicted a few months/years later by a different one-off study.
I think for the sake of our future, we should stop putting both physics which can predict, detect, and verify gravity waves from across the universe, and social sciences which are largely surveys with statistics into the same bucket.
- mankash666 10y agoEverything that's peer reviewed and backed by statistics is science. This article is stating that the (non) statistics used by the author, a.k.a cherry picking, is the root of fake results. If indeed one could replicate the results every single time, over a large population, then ESP would hold true. But that obviously fell flat on its face when the Berkeley coders backed it up with their online experiment
- RcouF1uZ4gsC 10y agoSo if Viking A measured thunder when it was cloudy vs clear, did some statistics and concluded with a sufficient P value that Thor was more likely to swing his hammer when it was cloudy, and then got a bunch of his buddies to agree his math was ok, you would call that science?
- mankash666 9y agoPeer review!= Collusion from buddies. What you describe here is collusion
- greglindahl 10y agoYou've never heard the old saw that everything with science in the name isn't? That comes from the same sentiment that you're expressing.