A p-value below a threshold says little about magnitude and nothing about practical relevance. We make the case for effect sizes with intervals as the default reporting unit, and work through examples where the two framings lead to opposite conclusions.
Authors choose licences quickly and live with them for a long time. This report traces how licence selection affects whether figures are reused in teaching materials, review articles and encyclopaedic works.
Repository records are only as findable as their metadata. We audit a sample of institutional deposits against a minimal quality rubric, and find that the fields most often incomplete are precisely those that discovery services rely on.
Registered reports move review before data collection. We summarise uptake across disciplines, compare outcome distributions against conventional submissions, and consider what the format costs authors in time.
Names are ambiguous, affiliations change, and both are recorded as free text far more often than as identifiers. This article reviews the identifier landscape for people and organisations, and what a journal gains by requiring them at submission.
Methods sections are where reproducibility is won or lost. Using automated extraction across a large corpus, we quantify how often key parameters - sample size justification, software versions, exclusion criteria - are stated at all.
Every submission is assessed by subject experts. Manuscripts receive methodological evaluation, ethical screening and constructive feedback before a decision is reached.