Very nice response, and I agree with your arguments. What astonished me was the *appalling* wording of Q9 in the questionnaire. I would expect much better from a methodologist than to use double barrelled questions asking for a graded response to an absolute statement ("clearly")
-
-
-
Also, if frequentists want to start talking about priors when interpreting & reporting p-values, then they need to either stop being frequentists, or start using frequentist stats in a way that explicitly considers priors (cf
@david_colquhoun). Hand waving doesn't wash.
Kraj razgovora
Novi razgovor -
-
-
There certainly is a lot of appalling
#science and dichotomania - but that isn't inherent in either term or concept of "statistical significance". The heart of the problem: ignorance and naivete abt the aims and the methodology of science.https://twitter.com/d_spiegel/status/1221752176600436737?s=19 …Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Hats-off, David! Very well said!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
A really great post, thank you! Is the main problem here p-values or the use of thresholds? And do you think the probabilistic statements in your Bayesian analysis are also vulnerable to thresholds (e.g. 97 vs 94)?
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Great read, thank you for a good scientific start of a difficult week
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Excellent article but it's a fast moving debate. My name will be on an industry guidance doc to be published soon where I have persuaded them to move away from bright line p-values to confidence intervals & professional judgement but some still hanker for bright lines.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
@jcerdal Mire profe, lo que vimos en clase el semestre pasado
-
Cierto, nos adelantamos.
Kraj razgovora
Novi razgovor -
-
-
Good discussion
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.