@harrietinthemain.bsky.social
Reposted
This is a really interesting phenomenon. Various things very similar to p-hacking are a common risk in hacking stuff (fundamentally an experimentally driven process to break into unknown systems).
p-hacking is primarily an academic publication phenomenon due to the incentives involved -- researchers find no evidence of p-hacking in e-commerce A/B testing because it's counterproductive to try and fool yourself
alexmiller.phd/research/p-h...
alexmiller.phd/research/p-h...
January 2, 2025 at 8:01 AM
This is a really interesting phenomenon. Various things very similar to p-hacking are a common risk in hacking stuff (fundamentally an experimentally driven process to break into unknown systems).