First analysis of ‘pre-registered’ studies shows sharp rise in null findings
399 by danso | 156 comments on Hacker News.