Thanks! RT @dbkell: Useful article on false discovery rates by @david_colquhoun http://t.co/MRaC2JIwr8
Thanks! RT @dbkell: Useful article on false discovery rates by @david_colquhoun http://t.co/MRaC2JIwr8
Useful article on false discovery rates by @david_colquhoun http://t.co/x1jEimBBjd
Useful article on false discovery rates by @david_colquhoun http://t.co/x1jEimBBjd
An investigation of the false discovery rate and the misinterpretation of p-values | Open Science http://t.co/xwdVbrS4oq
"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/S3EiGI4EdM
@bucksci & "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/iSCNBD8Z5x
If you use p=0.05 to suggest something "works", you will be wrong >30% of the time [assuming only 10% actually work] http://t.co/xWdQZXJ3aS
"having ‘constructed such a sturdy soapbox it would be a shame not to let [@david_colquhoun] speak from it’" http://t.co/GIaH1uTk9U
"An investigation of the false discovery rate and the misinterpretation of p-values" by @david_colquhoun http://t.co/6U2t54iw65
This should be compulsory reading for anyone using significance tests. P≤0.05 does not mean what you think! http://t.co/lQffD5Er7V
"having ‘constructed such a sturdy soapbox it would be a shame not to let [@david_colquhoun] speak from it’" http://t.co/GIaH1uTk9U
If you use p=0.05 to suggest something "works", you will be wrong >30% of the time [assuming only 10% actually work] http://t.co/xWdQZXJ3aS
Essentially, @david_colquhoun's Bayesian-esque point is that extraordinary claims require extraordinary evidence http://t.co/1WgL6HPnqy
Essentially, @david_colquhoun's Bayesian-esque point is that extraordinary claims require extraordinary evidence http://t.co/1WgL6HPnqy
If you use p=0.05 to suggest something "works", you will be wrong >30% of the time [assuming only 10% actually work] http://t.co/xWdQZXJ3aS
Essentially, @david_colquhoun's Bayesian-esque point is that extraordinary claims require extraordinary evidence http://t.co/1WgL6HPnqy
If you use p=0.05 to suggest something "works", you will be wrong >30% of the time [assuming only 10% actually work] http://t.co/xWdQZXJ3aS
An investigation of the false discovery rate and the misinterpretation of p-values http://t.co/DrsHiLO5Ev
An investigation of the false discovery rate and the misinterpretation of p-values http://t.co/DrsHiLO5Ev
“@ProfAndyField: MT via @xtaldave: Great open access paper on sig tests http://t.co/tN4NFfnfXU” @HLEBnapier thought you'd like this.
An investigation of the false discovery rate and the misinterpretation of p-values http://t.co/ROnPF42hqN #stat
"The problem with asking a Bayesian what to do is that you may get as many different answers as there are Bayesians” http://t.co/eZAWZ4WGyF
A very thought provoking read on interpretation of p values http://t.co/aXNBxVBlYS
この論文にも最後に"有意"を使うな、とある。An investigation of the false discovery rate and the misinterpretation… http://t.co/aT6ijOWl6y https://t.co/RIPxlPESe8
この論文にも最後に"有意"を使うな、とある。An investigation of the false discovery rate and the misinterpretation… http://t.co/aT6ijOWl6y https://t.co/RIPxlPESe8
A very thought provoking read on interpretation of p values http://t.co/aXNBxVBlYS
この論文にも最後に"有意"を使うな、とある。An investigation of the false discovery rate and the misinterpretation… http://t.co/aT6ijOWl6y https://t.co/RIPxlPESe8
"after having ‘constructed such a sturdy soapbox it would be a shame not to let the author speak from it’.” http://t.co/eZAWZ4WGyF
Pinches p's sarras, por eso luego nada es reproducible....♬ http://t.co/koJYL5JnRY
"The problem with asking a Bayesian what to do is that you may get as many different answers as there are Bayesians” http://t.co/eZAWZ4WGyF
"@LorenzLab: A thoroughly enjoyable paper. Needs to be read by a good many scientists. http://t.co/vHnP2gJdWy #statistics @david_colquhoun"
"@LorenzLab: A thoroughly enjoyable paper. Needs to be read by a good many scientists. http://t.co/vHnP2gJdWy #statistics @david_colquhoun"
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
required reading for anybody using hypothesis testing, p-values, fdr etc. http://t.co/GrAlJoTdHm #statistics #downwithsignificance
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
"@LorenzLab: A thoroughly enjoyable paper. Needs to be read by a good many scientists. http://t.co/vHnP2gJdWy #statistics @david_colquhoun"
An investigation of the false discovery rate and the misinterpretation of p-values http://t.co/bdkzAp8e0f sobering stuff for p<.05 hunters.
Very readable article by @david_colquhoun on how to judge significance in scientific results. Required reading. http://t.co/XsmGGWSrbQ
An investigation of the false discovery rate and the misinterpretation of p-values http://t.co/bdkzAp8e0f sobering stuff for p<.05 hunters.
Very readable article by @david_colquhoun on how to judge significance in scientific results. Required reading. http://t.co/XsmGGWSrbQ
Very readable article by @david_colquhoun on how to judge significance in scientific results. Required reading. http://t.co/XsmGGWSrbQ
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
Never use the word ‘significant’ ¿qué opina @sumediconosabe ?http://t.co/Sbmsm7sDZQ
Should be mandatory reading for all biomed students and researchers http://t.co/yZlqUoyqIg #statistics @david_colquhoun
Should be mandatory reading for all biomed students and researchers http://t.co/yZlqUoyqIg #statistics @david_colquhoun
Should be mandatory reading for all biomed students and researchers http://t.co/yZlqUoyqIg #statistics @david_colquhoun
Should be mandatory reading for all biomed students and researchers http://t.co/yZlqUoyqIg #statistics @david_colquhoun
筆者の主張が正当ならば、我々が提供している医学的介入の多くが恐ろしいほど不確かで、さらに再現性に乏しいことも説明できるかもしれません。http://t.co/6AFqEbskyw
p値カットオフ0.05の臨床試験を再現シミュレーションしたところ、第一種過誤(偽陽性率)は30%を上回った。第一種過誤を5%以内に抑える為のp値カットオフは0.001。http://t.co/6AFqEbskyw
"after having ‘constructed such a sturdy soapbox it would be a shame not to let the author speak from it’.” http://t.co/eZAWZ4WGyF
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
How not to make a fool of yourself with your choice of p-values, from @RSocPublishing by David Colquhoun http://t.co/YeDJ330iUv
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
筆者の主張が正当ならば、我々が提供している医学的介入の多くが恐ろしいほど不確かで、さらに再現性に乏しいことも説明できるかもしれません。http://t.co/6AFqEbskyw
p値カットオフ0.05の臨床試験を再現シミュレーションしたところ、第一種過誤(偽陽性率)は30%を上回った。第一種過誤を5%以内に抑える為のp値カットオフは0.001。http://t.co/6AFqEbskyw
"And never use the word ‘significant’." - ouch - http://t.co/cbrqBvoI8C
p値カットオフ0.05の臨床試験を再現シミュレーションしたところ、第一種過誤(偽陽性率)は30%を上回った。第一種過誤を5%以内に抑える為のp値カットオフは0.001。http://t.co/6AFqEbskyw
p値カットオフ0.05の臨床試験を再現シミュレーションしたところ、第一種過誤(偽陽性率)は30%を上回った。第一種過誤を5%以内に抑える為のp値カットオフは0.001。http://t.co/6AFqEbskyw
筆者の主張が正当ならば、我々が提供している医学的介入の多くが恐ろしいほど不確かで、さらに再現性に乏しいことも説明できるかもしれません。http://t.co/6AFqEbskyw
p値カットオフ0.05の臨床試験を再現シミュレーションしたところ、第一種過誤(偽陽性率)は30%を上回った。第一種過誤を5%以内に抑える為のp値カットオフは0.001。http://t.co/6AFqEbskyw
"And never use the word ‘significant’." - ouch - http://t.co/cbrqBvoI8C
"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/cbrqBvoI8C
Using p=0.05 as 'proof' makes you wrong at least 1/3 of time. Big source of #reproducibility problems #itoldyouso http://t.co/sFWFWoddKT
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
Using p=0.05 as 'proof' makes you wrong at least 1/3 of time. Big source of #reproducibility problems #itoldyouso http://t.co/sFWFWoddKT
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
An investigation of the false discovery rate and the misinterpretation of p-values http://t.co/XqleopIfOZ
Very readable article by @david_colquhoun on how to judge significance in scientific results. Required reading. http://t.co/XsmGGWSrbQ
"after having ‘constructed such a sturdy soapbox it would be a shame not to let the author speak from it’.” http://t.co/eZAWZ4WGyF
Pesky p values. Interpret yours with care: http://t.co/2qdLyas743 #OpenScience #OA #significance #stats @david_colquhoun
Never use the word ‘significant’ ¿qué opina @sumediconosabe ?http://t.co/Sbmsm7sDZQ
"after having ‘constructed such a sturdy soapbox it would be a shame not to let the author speak from it’.” http://t.co/eZAWZ4WGyF
"after having ‘constructed such a sturdy soapbox it would be a shame not to let the author speak from it’.” http://t.co/eZAWZ4WGyF
“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/Flw9bX5ZTL
"The problem with asking a Bayesian what to do is that you may get as many different answers as there are Bayesians” http://t.co/eZAWZ4WGyF
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Fascinating read - with some suggestions on how to go about it. But is it enough to buck the trend?! http://t.co/dFJhmkMXTd #pvalues
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Excellent paper on the misinterpretation of p-values (with bonus points for citing an @xkcdComic) http://t.co/FKuGevQkf9
required reading for anybody using hypothesis testing, p-values, fdr etc. http://t.co/GrAlJoTdHm #statistics #downwithsignificance
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Excellent paper on the misinterpretation of p-values (with bonus points for citing an @xkcdComic) http://t.co/FKuGevQkf9
An investigation of the false discovery rate and the misinterpretation of P values. (arXiv:1407.5296v3 [stat.AP]... http://t.co/YStBeOkZKN
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Still glorifying your p=0.047? Go read @david_colquhoun's thoru dissection of those treacherous p-values http://t.co/hOSIvOOGtO …
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Excellent paper on the misinterpretation of p-values (with bonus points for citing an @xkcdComic) http://t.co/FKuGevQkf9
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Excellent paper on the misinterpretation of p-values (with bonus points for citing an @xkcdComic) http://t.co/FKuGevQkf9
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Nice paper! 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/rWkoL4HA6Y
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
Interpretation of P values is now in a real journal. 5 months after going public on arXiv http://t.co/Kkg0nipAid
essential reading for all scientists “@david_colquhoun: Interpretation of P values is now in a real journal. arXiv http://t.co/y6uxlHG3hi”