Better statistical reporting does not lead to statistical rigour: lessons from two decades of pseudoreplication in mouse-model studies of neurological disorders.

The study was limited to neuroscience articles using animal models that permitted repeat sampling; it did not assess the entire literature or non-animal studies. Scoring for pseudoreplication involved subjective judgment, although double-blind scoring and consensus processes were used to mitigate bias. Articles lacking adequate statistical details were assumed to be free of pseudoreplication unless evidence suggested otherwise, which may underestimate the true prevalence.

Funded by the Simons Initiative for the Developing Brain and the Patrick Wild Centre at the University of Edinburgh.

Do not assume that a paper's compliance with statistical reporting guidelines guarantees that the data was analysed correctly. The study shows that rigorous reporting and rigorous analysis are separate issues, and pseudoreplication persists even in well-reported articles.