HT @sapinker: The beginning of the end for null-hypothesis significance testing http://pss.sagepub.com/cgi/content/long/0956797613504966v1 … => #BigData and the New #Statistics
-
-
- 1 more reply
New conversation -
-
-
I think "Statistics 2.0" has a better ring to it, tho MT
@sapinker …the end for null-hypothesis significance testing? http://pss.sagepub.com/cgi/content/long/0956797613504966v1 …Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker BTW: shout-out for@AndrewRAConway's BAYES acronym (Biased by n size, Arbitrary rule, Yokel local, Error prone, Shady logic)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
.
@sapinker@JayMan471 Highlighting CIs instead of point estimates and p-values should've gone without saying; too many caveats in the latterThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker It isn't so much NHST that's problem as fetishization of P=0.05, an absurdly 'generous' value - 1 more reply
New conversation -
-
-
@sapinker Seem to be a discipline divide. Did PhD in Chemistry w. lots of data analysis wout encountering null hypothesis. -
@grodaeu is there an equivalent to hyp testing in chemistry? -
@sanderwagner a bit ad hoc the whole thing: A would give tau=5, B tau=12, C tau=20. D,E,F don't make sense. We measured 11+/-2 so B -
@grodaeu its a bit hard to exactly understand what u did in the abstract and wo being a chemist, but i think i roughly get it. -
@sanderwagner Then you can start thinking about refining model with aerosols/soot/solar cycles/whatever
End of conversation
New conversation -
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker - Hopefully the beginning of the end for abuses and biases of NHST as often reported. Let's see all the data.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker Null hypothesis says NHST end unlikely (p=0.05) Past hope: http://bit.ly/1l78NrK , http://bit.ly/1coVGCM , http://bit.ly/1jxsJsbThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker Bayesian statistics all the way!Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@sapinker So is the jist that estimated ranges (CIs) are safer than dichotomous significance tests?Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
at last
@sapinker: The beginning of the end for null-hypothesis significance testing[…] http://pss.sagepub.com/cgi/content/long/0956797613504966v1 …”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.