There is a footnote for that claim - https://journals.sagepub.com/doi/full/10.1177/2515245920919667 https://journals.sagepub.com/doi/full/10.1177/25152459209196

There is a footnote for that claim - https://journals.sagepub.com/doi/full/10.1177/2515245920919667 https://journals.sagepub.com/doi/full/10.1177/25152459209196... - but basically "things that are hard to believe" and "things that barely passed statistical analysis" (high p-values).