Yes, we really live in a world where this needs to be said. Face. Palm.
@sapinkerhttps://twitter.com/sapinker/status/711259206695800832 …
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@BrendanEich That argument is highly misleading. p-values meant to do "x". He criticizes that it isn't equivalent to doing "y". -
@BrendanEich Stephen Senn has some great thoughts.https://twitter.com/NeuroStats/status/700710395254693888 …
End of conversation
New conversation -
-
@sapinker It simply states the level of confidence with which one claims the hypothesis is tentatively true. Or should I not use 'true'? -
@galibahsan@sapinker says nothing about prob. of theory, only prob. of experimental result: how rare result should be if theory is false.
End of conversation
New conversation -
-
-
@sapinker Use estimation statistics instead of significance testing.https://twitter.com/adamcchang/status/693130680373374976 …Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker i always new my stats teacher was full of shitThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker@IOPractitioners Reliance on P values may be predicated on the requirement of demonstrated stat. sig. 2 publish.#gamechangerThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker No - but it still indicates something. Perhaps we should focus on better (richer?) descriptions of what null hypotheses implyThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker 1 in 20 fluke results, out of interest who originally set p at > 0.05? would have thought with modern scientific method p > 0.005 -
@vinc3h4ll@sapinker 0.05 is not universally accepted - not acceptable standard at CERN by a very long stretch! -
@Rickenhacker@sapinker still interested in who set 0.05? in clinical medicine we seem to use 0.1 on both tails, both crazily high to me
End of conversation
New conversation -
-
-
@sapinker p-value depends on sample size. Can be made arbitrarily small via a large enough sample, in general. Bad test. What am I missing? -
New conversation -
-
-
@sapinker Drat. I was almost .5% sure. Just kidding. Stats = the music of math.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker Sadly this was the "correct" answer in the Psych GRE when I took it back in 20<cough>. Curious if they ever fixed it.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker please share it with banking industryThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker Oh thank goodness!Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.