Pre-registration is just year-old priors, statistician argues in Bayesian-frequentist spat

RexDouglass · x · 2026-09-29

Statistician Rex Douglass argues pre-registration is a uniquely frequentist idea: to a Bayesian, it's just "year-old priors," and a prior should reflect the state of the art now. Challenged by samth on how Bayesianism handles data-dredged spurious correlations, Douglass counters that pre-registration advocates should be reporting pre-reg confidence intervals — shrinking them or not — and "not a single one does," suggesting proponents don't practice what they preach. A pointed debate on research methodology and statistical inference.

Original post →

More from Research

Research channel →