Comment by hrhdkdlfnrne

In medical sciences is worse. In one attempt to replicate "lamdmark" cancer studies 89% failed to replicate.

https://www.nature.com/articles/483531a