"The truth wears off: Is there something wrong with the scientific method?"

Gur Huberman asks what I conceive of this entrepot article by Johah Lehrer (see also here).

My state is that it reminds me a bit of what I wrote here. Or wager here for the hurried powerpoint version: The brief news is that if you concealment for statistical meaning when estimating diminutive effects, you module necessarily appraisal the magnitudes of effects, sometimes by a huge amount. I undergo that Dave Krantz has intellection about this issue for awhile; it came up when Francis Tuerlinckx and I wrote our essay on Type S errors, decade eld ago.

My current intellection is that most (almost all?) research studies of the sort described by Lehrer should be attended by retro noesis analyses, or consultative theorem inferences. Either of these approaches--whether Hellenic or Bayesian, the key is that they incorporate actual preceding information, just as is finished in a Hellenic likely noesis analysis--would, I think, moderate the tendency to appraisal the ratio of effects.

In respond to the question posed by the title of Lehrer's article, my respond is Yes, there is something criminal with the scientific method, if this method is defined as running experiments and doing accumulation analysis in a patternless artefact and then reporting, as true, results that transfer a statistical meaning threshold.

And corrections for binary comparisons module not solve the problem: much adjustments merely agitate the boundary without partitioning the difficulty of estimate of diminutive effects.