Estimating the reproducibility of psychological science

A large-scale assessment suggests that experimental reproducibility in psychology leaves a lot to be desired, and correlational tests suggest that replication success was better predicted by the strength of original evidence than by characteristics of the original and replication teams.