One can then test if the curve here is left skewed, with fewer highly significant than barely significant results, indicative of P -hacking [ 10 , 22 ].
← all excerpts
Evidence that nonsignificant results are sometimes preferred: Reverse P-hacking or selective reporting?
2
—
—
The sentences
Researchers might prefer not to provide exact P -values when they lie close to 0.05 (i.e., below 0.10) because such findings are often described, albeit erroneously, as ‘marginally significant’.