2025-6 Leisurely Cruise
The following is the February stop of our leisurely cruise (meeting 6 from my 2020 Seminar at the LSE). There was a guest speaker, Professor David Hand. Slides and videos are below. Ship StatInfasSt may head back to port or continue for an additional stop or two, if there is interest. Although I often say on this blog that the classical notion of power, as defined by Neyman and Pearson, is one of the most misunderstood notions in stat foundations. I did not know, in writing SIST, just how ingrained those misconceptions would become. I’ll write more on this in my next post. (The following is from SIST pp. 354-356, the pages are provided below)
Shpower and Retrospective Power Analysis
It’s unusual to hear books condemn an approach in a hush-hush sort of way without explaining what’s so bad about it. This is the case with something called post hoc power analysis, practiced by some who live on the outskirts of Power Peninsula. Psst, don’t go there. We hear “there’s a sinister side to statistical power, … I’m referring to post hoc power” (Cumming 2012, pp. 340-1), also called observed power and retrospective (retro) power. I will be calling it shpower analysis. It distorts the logic of ordinary power analysis (from insignificant results). The “post hoc” part comes in because it’s based on the observed results. The trouble is that ordinary power analysis is also post-data. The criticisms are often wrongly taken to reject both. Continue reading
Error Statistics Philosophy 