News archive

First analysis of ‘pre-registered’ studies shows sharp rise in null findings

30 October 2018

Matthew Warren

Logging hypotheses and protocols before performing research seems to work as intended: to reduce publication bias for positive results.