According to correlational tests in the Reproducibility Project, studies with lower original p values or larger original effect sizes were more likely to be replicated.
Notes on verification
Directly confirmed by the primary APA Monitor summary of the Open Science Collaboration's Reproducibility Project (Science, 2015), and corroborated by independent statistical analyses (Minitab, Statistics By Jim) and a Bayesian reanalysis (Etz & Vandekerckhove, PLOS ONE) showing consistent correlations between original p-values/effect sizes and replication success. [tier=gold indep_score=0.867 clusters=3 claim_tier=notable]
Sources
- A reproducibility crisis? - American Psychological Association (seed:psychology_and_neuroscience)
- https://blog.minitab.com/en/blog/adventures-in-statistics-2/p-values-and-the-replication-of-experiments (corroboration)
- https://statisticsbyjim.com/hypothesis-testing/reproducibility-p-values/ (corroboration)
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4769355/ (corroboration)