Moving Beyond p < .05: The Latest

A couple of days ago, the three authors of the Nature paper accompanying the special issue of TAS on moving beyond p < .05 sent the update below. (See below for lots of links.)

We are writing with a brief update on the Nature comment “Retire statistical significance” that you signed. The comment has so far been subject to spirited discussion in both traditional and social media (see below).

We believe keeping this discussion going by approaching colleagues and sharing links on social media is the only way a reform in statistical practice will come about. Your continued support in this is most welcome and appreciated! Together, our chance to make this happen has perhaps never been better.

With kind regards and many thanks,Valentin, Sander, Blake

Links:
Retire statistical significancehttps://www.nature.com/articles/d41586-019-00857-9
The American Statistician – Statistical inference in the 21st century: a world beyond p < 0.05https://www.tandfonline.com/toc/utas20/73/sup1

Media mentions:
Retraction Watchhttps://retractionwatch.com/2019/03/21/time-to-say-goodbye-to-statistically-significant-and-embrace-uncertainty-say-statisticians

Andrew Gelman’s blog (discussion with 400 comments)https://statmodeling.stat.columbia.edu/2019/03/20/retire-statistical-significance-the-discussion

Bloomberghttps://www.bloomberg.com/opinion/articles/2019-03-29/dump-statistical-significance-then-teach-scientists-statistics

The Guardianhttps://www.theguardian.com/commentisfree/2019/mar/24/the-guardian-view-on-statistics-in-sciences-gaming-the-unknown

Voxhttps://www.vox.com/latest-news/2019/3/22/18275913/statistical-significance-p-values-explained

NPRhttps://www.npr.org/sections/health-shots/2019/03/20/705191851/statisticians-call-to-arms-reject-significance-and-embrace-uncertainty

……………………………….

Browse Gelman’s blog and its comments to get an idea of the range of views held by, especially, statisticians. There may be a danger that the imperative to move forward from p values gets lost in the noise. We really do need to keep focus! (Yay for the new statistics! Estimation, CIs, meta-analysis to the fore! Let p values wither and die a natural death!)

Geoff