1 comment

[ 1.9 ms ] story [ 17.0 ms ] thread
Quote: "The negative correlation between effect size and samples size, and the biased distribution of p values indicate pervasive publication bias in the entire field of psychology."