"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
“@david_colquhoun: Here's the link again http://t.co/LbYcPbBIiK”Good bank holiday mind food. Have I been kidding myself? I'll read it again.
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time." http://t.co/xA7kgJ0xuE
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
http://t.co/SvsDH2WIpf "Berger's calibration suggests that p=0.0027 corresponds to an FDR of 0.042" Two times p<.05 = p<.0025 => FDR< 5%.
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
Things I Didn't Have Room For In My Previous Tweet: This Link For That Quote - http://t.co/Z7mnUJRlTF #Science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
@R__INDEX of course -see Fig 3 in http://t.co/X1NEc1p2Lj
@R__INDEX The only snag is lack of unique relation between P and FDR. On basis of simulations. my suggestions are at http://t.co/Hb7so5LL6d
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
"If you use p=0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time": http://t.co/2nxcvGXmLK #science
.@bengoldacre I hope someone will investigate how much of this dismal result can be explained by false discovery rate http://t.co/X1NEc1p2Lj
.@bengoldacre I hope someone will investigate how much of this dismal result can be explained by false discovery rate http://t.co/X1NEc1p2Lj
.@bengoldacre I hope someone will investigate how much of this dismal result can be explained by false discovery rate http://t.co/X1NEc1p2Lj
.@bengoldacre I hope someone will investigate how much of this dismal result can be explained by false discovery rate http://t.co/X1NEc1p2Lj
Here's the link again http://t.co/X1NEc1p2Lj
@TimHarford Is the false discovery rate too complicated for More or Less? http://t.co/X1NEc1p2Lj and, simpler, http://t.co/jbEqYe5N1q
@MikeTonge @brembs @drchriscole it's where my paper sat, gathering feedback, for 4 months before it was publsihed http://t.co/relFUWcK7f
Any comments welcome at http://t.co/Hb7so5LL6d on my suggestions for how P values shjould be described in papers (NOT as "significant")
@drchriscole @MikeTonge @brembs Doubt it web space is very cheap now,. And it;s easy to format like a paper yourself http://t.co/relFUVV8IF
@R__INDEX one thing one could do is to change wording used to describe P values My suggestions http://t.co/Hb7so5LL6d
.@R__INDEX And I'm surprised that you don't mention false discovery rate which muat be big part of the problem 2/2 http://t.co/X1NEc1p2Lj
.@pennutrition Consult @david_colquhoun's P values article & video too:http://t.co/TQAtvcDl6Q http://t.co/pCn47pCF1c https://t.co/jmEpOJCpnd
If you use p=0.05 to suggest you've made a discovery you'll be wrong at least 30% of the time @david_colquhoun paper: http://t.co/VzEgoDeRyo
If you use p=0.05 to suggest you've made a discovery you'll be wrong at least 30% of the time @david_colquhoun paper: http://t.co/VzEgoDeRyo
If you use p=0.05 to suggest you've made a discovery you'll be wrong at least 30% of the time @david_colquhoun paper: http://t.co/VzEgoDeRyo
Going to pin this up. "An investigation of the false discovery rate & the misinterpretation of p-values" Open Science http://t.co/NMVcxjgF4e
Fantastic article on false discovery rates and misinterpretation of p-values, from @david_colquhoun here http://t.co/CaQk0gPuDp
@JeffRouder after 4 months on arXiv it was pubishedl http://t.co/X1NEc1GDCR Then a simplified version on Youtube http://t.co/jbEqYdObCQ
Fantastic article on false discovery rates and misinterpretation of p-values, from @david_colquhoun here http://t.co/CaQk0gPuDp
Fantastic article on false discovery rates and misinterpretation of p-values, from @david_colquhoun here http://t.co/CaQk0gPuDp
Fantastic article on false discovery rates and misinterpretation of p-values, from @david_colquhoun here http://t.co/CaQk0gPuDp
Fantastic article on false discovery rates and misinterpretation of p-values, from @david_colquhoun here http://t.co/CaQk0gPuDp
@WTeducation @Larryferlazzo @imascientist On that site, http://t.co/F8Tn367wNg should read http://t.co/R1pWe7gX2R and http://t.co/X1NEc1p2Lj
Wonderful paper by David Colquhoun on the "false discovery rate" http://t.co/kSSuho1I4M
@mc_hankins @MaryDixonWoods Agreed: too poor to consider bringing @david_colquhoun False Discovery into discussion http://t.co/vr6YXjUexX
An investigation of the false discovery rate and the misinterpretation of p-values http://t.co/6iYYVtsIOj