Open Science Collaboration (2015). Estimating the reproducibility of psychological science. Science 349(6251), aac4716.

A large team repeated 100 published psychology studies. Thirty-six per cent of the replications were statistically significant, and the mean effect size was about half that of the originals. Open science answers this with more exposure, through registered plans and shared data. The wiki cites it in its brief on protective space as part of the case for early exposure, set beside the case for protection.

Cited by