"[Decisions should] be made based on the costs, benefits, and likelihoods of all possible outcomes, not via arbitrary cutoffs applied to statistical summaries such as p-values which capture little of this picture."
https://t.co/p5kePFkZ46
.@StatModeling has a few things to say about the proposal to get rid of stat sig in published reports: https://t.co/yLDsoA2N71
TL;DR: he’s a fan. https://t.co/PfEgiLStF8
Abandoning statistical significance is both sensible and practical [PeerJ Preprints]. Great reply by @vamrhein , @Lester_Domes , @StatModeling and McShane to comment by Ioannidis. https://t.co/EC4mdzfPMR
"The replication crisis has arisen because unreliable findings are presented as reliable."
Amrhein, Gelman, Greenland, McShane ...
https://t.co/L4wYDTA2TZ
reply to Ioannidis ...
https://t.co/v9DHXy7Ihg
who replies to "Retire statistical significance"
https://t.co/CTesqcX8XV