RT @ken_rothman: @bj_cairns @EpidLife I'm not a stats educator, but I have been advocating estimation over testing for decades. Mostly it…
RT @ken_rothman: @bj_cairns @EpidLife I'm not a stats educator, but I have been advocating estimation over testing for decades. Mostly it…
@bj_cairns @EpidLife I'm not a stats educator, but I have been advocating estimation over testing for decades. Mostly it seems that nothing changes, but if you take the long view, there has been some movement. For example, there is the ASA statement: ht
@altmetric A3.3 Another good example would be The American Statistical Association’s @AmstatNews 2016 statement on p-Values https://t.co/38LIfgLPai - an important statistical model (often mis)used in many fields #altmetchat https://t.co/pC9QTFXFDl
Scientific reasoning The ASA's statement on P-value misuses and misconceptions https://t.co/dJ81uEOkjo
@ra_fun @DLBHATTMD @DrDerekConnolly P values help you judge the likelihood that the null hypothesis is true, but there are ways that studies can be wrong that a p value doesn't take into account. Here's more on that: https://t.co/2snURd2dK8 https://t.co/gW
@rjwthree @saralpulit "Scientific conclusions... should not be based only on whether a p-value passes a specific threshold." Population frequencies and effect sizes are necessary for determining relevance, as is a good understanding of population structur
RT @iamdavecampbell: ASA statement on p-values: https://t.co/llNdiXuQTi #birsmath relates to #StatisticalClarity but doesn't address langua…
RT @iamdavecampbell: ASA statement on p-values: https://t.co/llNdiXuQTi #birsmath relates to #StatisticalClarity but doesn't address langua…
RT @iamdavecampbell: ASA statement on p-values: https://t.co/llNdiXuQTi #birsmath relates to #StatisticalClarity but doesn't address langua…
ASA statement on p-values: https://t.co/llNdiXuQTi #birsmath relates to #StatisticalClarity but doesn't address language per se
Not new, but P values are not the be-all-end-all for evaluating scientific hypotheses: https://t.co/RL7pjhLR2z
RT @ken_rothman: @dscharf3 @irishinneburg Even the American Statistical Association supports the point that @dscharf3 is making. From thei…
RT @ken_rothman: @dscharf3 @irishinneburg Even the American Statistical Association supports the point that @dscharf3 is making. From thei…
@dscharf3 @irishinneburg Even the American Statistical Association supports the point that @dscharf3 is making. From their official statement: "A p-value, or statistical significance, does not measure the size of an effect or the importance of a result."
@JacobPlieth Nice article, but definition of p is not correct. p denotes the probability to obtain the data found under the assumption of a specific statistical model but not the probability that a finding is due to chance. See statement by the Am Stat Ass
@lonnibesancon "5. A p-value, or statistical significance, does not measure the size of an effect or the importance of a result." https://t.co/Gz2QmT1pYX Might take a bolder reviewer to dispute the ASA on statistical practice? 3/3
RT @MaartenvSmeden: @kaz_yos From the ASA statement on p-values: https://t.co/mnTRsWU1if https://t.co/QbMyj44WpU
@kaz_yos From the ASA statement on p-values: https://t.co/mnTRsWU1if https://t.co/QbMyj44WpU
@AMAManual You’ll notice, for example, that this statement on P from the American Statistical Association uses the convention I describe: https://t.co/PfPuZ9XfzG
@Eric_Fauman @SbotGwa The other issue is that these plots put priority on p-value, which is not appropriate. "A p-value, or statistical significance, does not measure the size of an effect or the importance of a result." https://t.co/IP9ShAP7yG We've dis
Regarding the P-value: it is just about a difference of some kind (degree) between two leaves, two apples, two coins, two colours. And that is all about the P-value. It is just an introductory measure. https://t.co/n4rD0c0VjC
The ASA's Statement on p-Values: Context, Process, and Purpose is still #FreeAccess. You can find it here: https://t.co/sTqKkej4ZD @AmstatNews @Ron_Wasserstein https://t.co/oS4uTrnuKw
@lastexposfan Amen! Moreover, see the ASA's official position statement on the use (and abuse) of p-values: https://t.co/bPkR4vzZEz
@wgervais @OstlundOllie @AndrewPGrieve @bmwiernik The odd interpretation IMO is the claim that a p-val is not evidence. Even the ASA statement on p-vals (https://t.co/k7puCxaUBf) says it is evidence against a null, but also cautions to not *only* report a
Look here for a more detailed explanation of the @AmstatNews statement: https://t.co/859a4fA0Gy Or here for the full statement: https://t.co/sC8QWjAcMp #pvalues #ScienceMatters #mathMatters #FactsMatter #FactsNotFear
@Duinhiron @MihoJnvr @Visions_Science I'm not sure I would. Defining *a* scientific method is a very difficult unsettled problem that many philosophers and statisticians are tackling. Many key features like p-values are harshly criticized by, eg, the Ame
@makennedy101 @JShendure Problem #4: Results should be ranked by a statistic that is related to the difference between groups. A p-value, or statistical significance, does not measure the size of an effect or the importance of a result. Use p-values togeth
@Eric_Fauman @SbotGwa "A p-value, or statistical significance, does not measure the size of an effect or the importance of a result" You can be *really* confident about an effect size, but if that effect size is functionally equivalent to zero, then the p
este link es el "me destruiste la infancia" para quienes les damos importancia a los asteriscos del p-value en una regresión https://t.co/HV1wVv5ePC
NO, this is not what a p-value means Clinical Trials, End Points, and Statistics—Measuring and Comparing Cancer Treatments in Practice https://t.co/Yio6Ql8hlC See (again): https://t.co/RWEUbULpKv
The American Statistical Association’s Statement on p-Values: Context, Process, and Purpose https://t.co/Y4LldQplL3
@johngraves @JonRMcClure The American Statistical Association has a useful statement concerning the use of p-values (and significance tests) that every journalist and scholar should know about. 1/ https://t.co/4ulz0LB4jK
In 2014, Professor Cobe asked a statistics discussion forum: "Q: Why do so many colleges and grad schools teach p = 0.05? A: Because that’s still what the scientific community and journal editors use. Q: Why do so many people sti…https://t.co/7kvwu93MVm h
@sheppard_lab @DanielFalush Although there's still a wrinkle about p-values: * genes / SNPs ranked by p-value * plots / statistics generally only report p-values * no indication that the most distinguishing SNP / gene was known, but at least allele frequen
RT @gonen_mithat: Required reading for everyone, especially for non-statisticians who use p-values routinely. https://t.co/t1elAuXqZM
Required reading for everyone, especially for non-statisticians who use p-values routinely.
Have you read "The ASA's Statement on p-Values: Context, Process, and Purpose" yet? It's free access and you can find it here: https://t.co/VrgxWy3dfK @AmstatNews https://t.co/OSRdwy3ebg
RT @RealRupertDBear: @RodneyTaveras @DTFinegood @nntaleb 1) p of 0.05 is arbitrary. 2) So-called "p hacking" Check this out: https://t.co/K…
@RodneyTaveras @DTFinegood @nntaleb 1) p of 0.05 is arbitrary. 2) So-called "p hacking" Check this out: https://t.co/KnSZdmR4Zh
@hoffsbeefs @NatureGenet Preferably by a statistic that has biological relevance. Absolute value of effect size is an example of that. P-values should only be used as a filter ("is this value reasonable?"), not for determining the degree of significance o
The ASA grapples with the "long-worrisome circularity in the sociology of science based on the use of bright lines such as p < 0.05: “We teach it because it’s what we do; we do it because it’s what we teach.” https://t.co/srgbBJg7no https://t.co/4B08FKM
"No single index should substitute for scientific reasoning". Bingo. #pvalue #Statistics #Science https://t.co/rFxM25RLXV
@chadinabhan You might be interested in the American Statistical Association's statement from 2016 on p-values along with the paper: https://t.co/0k7KKtfTCU https://t.co/IYiHSDUs8o
RT @Students4BE: The ASA’s Statement on p-Values: Context, Process, and Purpose https://t.co/MG3RPrHo03
The ASA’s Statement on p-Values: Context, Process, and Purpose https://t.co/MG3RPrHo03
@hammerheadbat This is why people should listen and take in the things that statisticians say ("zuhören" vs "hören" in German), rather than assuming p-value thresholds are meaningful: https://t.co/TUDvXx6TCp
@onechankeidansi @zhanpon 5. A p-value, or statistical significance, does not measure the size of an effect or the importance of a result. https://t.co/vR11wyrWSU
Also useful is the ASA's statement on p-values, followed by interesting discussions from 18 experts. https://t.co/onWlu89eJO
アメリカ統計学会 p値に関する声明 ASA Statement on Statistical Significance and P-Values https://t.co/xBd79mQUM8
RT @philecoPL: "By itself, a p-value does not provide a good measure of evidence regarding a model or hypothesis", in: American Statistical…
"By itself, a p-value does not provide a good measure of evidence regarding a model or hypothesis", in: American Statistical Association statement on p-values. Really worth reading https://t.co/FsKGX51scd
What’s the problem with p values? https://t.co/8ilQY5aLa1 #UKCCRF18 @NBT_ICU
@CmRoycostilla The exact p-value is being treated as important, and a value with a *really* low p-value (such as 9e-149) is seen as being *really* significant. This does not match the recommendations of ASA: https://t.co/NIKJ8PYAMa https://t.co/A4NZadikkp
@OlgaTroyanskaya "leading to significantly improved performance, p=6.7x10 -123" *sigh* Please at least include one other measure when you report p-values, as "A p-value, or statistical significance, does not measure the size of an effect or the importanc
RT @NickRHart: If you haven’t yet read @AmstatNews 2016 statement on statistical significance & #pvalues, set a few minutes aside for #Mond…
RT @NickRHart: If you haven’t yet read @AmstatNews 2016 statement on statistical significance & #pvalues, set a few minutes aside for #Mond…
If you haven’t yet read @AmstatNews 2016 statement on statistical significance & #pvalues, set a few minutes aside for #MondayMorning to learn, understand, and reflect on what it means for our scientific endeavors. #phacking #openscience #scicomm #eval
RT @Students4BE: The American Statistical Association's Statement on p-Values: Context, Process, and Purpose. A must read if you want to d…
RT @Students4BE: The American Statistical Association's Statement on p-Values: Context, Process, and Purpose. A must read if you want to d…
RT @Students4BE: The American Statistical Association's Statement on p-Values: Context, Process, and Purpose. A must read if you want to d…
RT @Students4BE: The American Statistical Association's Statement on p-Values: Context, Process, and Purpose. A must read if you want to d…
RT @Students4BE: The American Statistical Association's Statement on p-Values: Context, Process, and Purpose. A must read if you want to d…
RT @Students4BE: The American Statistical Association's Statement on p-Values: Context, Process, and Purpose. A must read if you want to d…
The American Statistical Association's Statement on p-Values: Context, Process, and Purpose. A must read if you want to delve into the world of p-Values... https://t.co/MG3RPrHo03
RT @davidlederer: Can't say enough bad things about p-values. Here's a reminder (i guess mostly to me) about what p-values are and what the…
Can't say enough bad things about p-values. Here's a reminder (i guess mostly to me) about what p-values are and what they aren't...all taken from: https://t.co/gHy896gTR8 @Ron_Wasserstein @AmstatNews 1/x https://t.co/ZVQBonIzG9
@BayesicTony @molly_przew Fiddling with p-values to change the importance of a result is the wrong way to go about GWAS. https://t.co/i39uOkgZ56 I prefer internal subsampling to knock out highly-variable alleles and at least make sure that the associatio
@medrants Not quite. A p value of 0.05 means there is a 5% of observing results at least as extreme as were observed, IF there was in truth no effect/difference (or whatever the null hypothesis is). The condition is CRUCIAL, and often ignored! See @AmstatN
RT @consult_stat: 원문: Ronald L. Wasserstein & Nicole A. Lazar 2016 "The ASA's Statement on p-Values: Context, Process, and Purpose” https:/…
RT @consult_stat: 원문: Ronald L. Wasserstein & Nicole A. Lazar 2016 "The ASA's Statement on p-Values: Context, Process, and Purpose” https:/…
원문: Ronald L. Wasserstein & Nicole A. Lazar 2016 "The ASA's Statement on p-Values: Context, Process, and Purpose” https://t.co/zGCXpwBK58
RT @unlearnai: "By itself, a p-value does not provide a good measure of evidence" - American Statistical Association https://t.co/OOjRcEbQ…
RT @unlearnai: "By itself, a p-value does not provide a good measure of evidence" - American Statistical Association https://t.co/OOjRcEbQ…
"By itself, a p-value does not provide a good measure of evidence" - American Statistical Association https://t.co/OOjRcEbQDi
RT @BioVinci: "- Why do so many colleges and grad schools teach p = 0.05? - Because that’s still what the scientific community and journal…
"- Why do so many colleges and grad schools teach p = 0.05? - Because that’s still what the scientific community and journal editors use. - Why do so many people still use p = 0.05? - Because that’s what they were taught in college or grad school.” https:/
RT @drjwdietrich: @no_identd @CMichaelGibson @lakens @stephenbenning The ASA's statement on p-values is worth reading, too: https://t.co/K5…
@no_identd @CMichaelGibson @lakens @stephenbenning The ASA's statement on p-values is worth reading, too: https://t.co/K5gzPCn8uV
RT @cofactoranna: This post links to an editorial explaining the problems with p-values (tweeting to bookmark for later reading) https://t.…
Bilimin kısır döngüsü: "S: Neden üniversitelerde p<0.05 olarak anlatılıyor? C: Çünkü bilim topluluğu ve bilimsel dergiler bunu kullanıyor. S: Neden bir çok bilimsel yayında hala p<0.05 olarak kullanılıyor? C: Çünkü üniversitelerde öğretilen bu. "
RT @ICU_Management: On p values @GMeyfroidt recommends the ASA statement https://t.co/CZzUvvnbX1 #ISICEM18
On p values @GMeyfroidt recommends the ASA statement https://t.co/CZzUvvnbX1 #ISICEM18
どこぞのブログで、すごくズレた解釈をしていたようだけど、統計を扱う人は原文を読むべき。そのp値に至るまで検証した仮説や個々の仮説のp値、データの選定は明確にすべき。 https://t.co/iUbtrkiwXk
統計的有意性と P 値に関する ASA 声明(日本計量生物学会訳) https://t.co/n9ZemKJWU4 元 The ASA's Statement on p-Values - American Statistical Association https://t.co/RXpNUkecv4
I was recently reminded of the importance of experimental design, hypothesis testing and data manipulation: https://t.co/P5pHm5tg9i
@matthewladner @nhannahjones @sarahaprecourt Some reading to catch us up on the p-values crisis. https://t.co/3FY1dttwvd
RT @nachalca: @marianazerpa @nberpubs the role of significance tests and p-values have being (more actively) discussed in stat community er…
@marianazerpa @nberpubs the role of significance tests and p-values have being (more actively) discussed in stat community ercently. 2015: initial reactions to a NHST ban https://t.co/upOiPNpx9O 2016: ASA statement: https://t.co/yCtbyFXhkt 2017: beyond
RT @kenmetrics: ちょうどアメリカ統計学会がp値について歴史的な声明を出したところですが、この要点を端的にいえば、「p値は証拠・効果の大きさを満足に表現できない」ということだと思います。https://t.co/XPlyDBC0X0 (先の再現性投稿の続きですが…
RT @kenmetrics: ちょうどアメリカ統計学会がp値について歴史的な声明を出したところですが、この要点を端的にいえば、「p値は証拠・効果の大きさを満足に表現できない」ということだと思います。https://t.co/XPlyDBC0X0 (先の再現性投稿の続きですが…
@MarinePolicy @AcademiaObscura A nice summary discussion on P-values by the American Statistical Association (ASA) can be found here: https://t.co/HPp0ORxvni
良いペーパーです。リンク→ https://t.co/jpuceTEAnY