It really is fun to objectively choose a level of alpha for NHT using Mudge's optimal alpha. It's so simple with a little simulation and elbow grease. And then you feel kinda boss-level when someone asks you why you did not use 0.05. https://t.co/yO923RRGP
@TimandraHarknes @stephensenn I can recommend reading Mudge et al. https://t.co/B8ob7NjkTj
@J_A_Quent The main thing I am interested in is minimizing error rates (https://t.co/jbChcr27EB) and the flexibility to set my error rates however I need for a specific research line. In general, I personally think power is more important than Type 1 error
Interesting paper on optimising alpha
@MarcusMunafo @tdienlin I know of earlier and better papers about this: https://t.co/jbChcr27EB. Scanning the abstract, this paper seems to say a similar but more limited thing? Obviously this whole approach was already mentioned by Neyman and Pearson in
@f2harrell @john_slp @lonnibesancon https://t.co/Xzwn01hlRm Alpha set using n and costs of making errors.
@StephenBHeard @biogeobiochem One of my buddies developed a pretty neat system for choosing an appropriate alpha level that minimizes Type I & Type II errors in NHSTs that he dubbed 'optimal alpha'. (@joemudge, one of Jeff Houlahan's students; @CB_Edge
@JPdeRuiter @RickCarlsson Sorry to interrupt, if the level of alpha can be discussed you agree the choice of alpha should NOT be one-size-fits-all! Sure, .05 likely won't do for most of psychology, but a formal method for optimising alpha should not exclud
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
@AHPG_UoA we didn't have a chance to talk about justifying (contra lowering) one's alpha levels, but this seems interesting. Ping @DanPowell83 @NJavornik, re: Wednesday's discussion on the topic https://t.co/wlEvtGMULF
RT @Research_Tim: Lovely tool! However, if (this particular form of) justifying your alpha requires a prior probability on hypotheses, why…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @Research_Tim: Lovely tool! However, if (this particular form of) justifying your alpha requires a prior probability on hypotheses, why…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
Lovely tool! However, if (this particular form of) justifying your alpha requires a prior probability on hypotheses, why not go all the way and calculate a posterior probability too? https://t.co/NLXrzjO2vV
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
RT @dsquintana: An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Ba…
Noice https://t.co/AHvMMdjazh
An online tool to calculate the optimal alpha that minimises Type 1 and Type 2 error for t-tests https://t.co/jnZkkvdmdI Based on this paper https://t.co/ktuFdsYlz8 https://t.co/kZzkhWSbZV
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: @hardsci @GordPennycook What is 'scientifically principled'? I think you might be having some unrealistic expectations about ho…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
@hardsci @GordPennycook What is 'scientifically principled'? I think you might be having some unrealistic expectations about how this is supposed to work. The perfect being enemy of the good, and all that. Here's one direct improvement to current approache
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge et al... who have an online Shiny app you can use to implement their approach. App: https://t.co/XoeMPaRAR4 Paper: htt
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
Interesting tool to get optimal alpha level that minimizes Type I and II errors, with assumptions laid out. Link to paper too. https://t.co/OSm2IGbCY7
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
A well-known and old result, that seems weirdly uncited in all these debates, is that minimising a linear combination of error rates for two point hypotheses is equivalent to choosing a likelihood ratio https://t.co/qz7RL6Uucv
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
RT @lakens: One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge e…
One (largely ignored) recommendation on how to justify your alpha level is to minimize the combined error rate. We cite Mudge et al in our commentary who have an online Shiny app you can use to implement their approach. App: https://t.co/33EdLknfir Paper:
@JPdeRuiter @StuartJRitchie I was wondering the same thing, what do you think “No one alpha to rule them all” means? I can come up with examples that would make our proposal support RSS’s plea for .005, also much lower than .005 or even higher than .05 See
RT @zerdeve: @lakens this 2012 paper seems to have addressed that as well: https://t.co/YeHynVXBDf instead of a fixed threshold, find the o…
@lakens this 2012 paper seems to have addressed that as well: https://t.co/YeHynVXBDf instead of a fixed threshold, find the optimum for ur research
#1 - https://t.co/RTWZl1UoIL #2 - https://t.co/P9wPGhDsU3 #3 - https://t.co/e2qNlXY3yg Is this a movement?
#STATISTICS #PLOSONE: Setting an Optimal α That Minimizes Errors in Null Hypothesis Significance Tests http://t.co/HTj4mb8Deh
@algaegirl328 completely arbitrary, unless you use my buddy @joemudge's novel approach to optimizing alpha :) http://t.co/w7YBG32fZ8
@katiesci it might be if you took this approach :) http://t.co/kD3XpS2E
Setting an Optimal α That Minimizes Errors in Null Hypothesis Significance Tests http://t.co/9XWa6o8w #bioinformatics #biostatistics
PLoS1: Setting an Optimal α That Minimizes Errors in Null Hypothesis Significance Tests http://t.co/QpVrWtzY. Aguardando [boas] criticas ;-)
A new way of approaching alpha choice for significance tests. Colleagues at UNBSJ introduce optimal alpha @FranLeanne http://t.co/kD41XsbO