aftermaths of retiring significance
Beyond mentions in the general press of the retire significance paper, as in Retraction Watch, Bloomberg, The Guardian, Vox, and NPR, not to mention the large number of comments on Andrew’s blog, and Deborah Mayo’s tribune on a ban on free speech (!), Nature of “the week after” contained three letters from Ioannidis, calling for more stringent thresholds, Johnson, essentially if unclearly stating the same, and my friends from Amsterdam, Alexander Ly and E.J. Wagenmakers, along with Julia Haaf, getting back to the Great Old Ones, to defend the usefulness of testing versus estimation.
Xi'an's Og 
April 10, 2019 at 4:45 am
Isn’t this overall debate about hypothesis tests getting a little tiresome? Most of the arguments are not in any sense new, but nevertheless are reported as if they are (giving one that groundhog day feeling!)
Avoiding the thresholding of significance (as carefully advocated in the original paper McShane et al. 2019) is of course a good idea, but the underlying problem appears to be a general one, that of doing bad statistics.
Perhaps many people will start doing even worse analyses to avoid the stigma of having used thresholded significance levels (or any other overly highlighted bad practice), and by so doing, feel that they have gained a seal of quality.
Adequately qualified statisticians (be they specialists or hands-on practitioners) are in ridiculously short supply. Unfortunately, it is unclear how it will ever be possible to genuinely correct for this deficiency.