Paper [https://journals.sagepub.com/doi/pdf/10.1177/2515245920919667]by Hoogeven
et. al., on how people were able to predict which social-science studies would
replicate, in other words, which social-science studies are likely to have true
results.
Sample size is quite small, 233