Science Fair
Scientist here. About 50% of all published results cannot be reproduced in another lab. A lot of statistics are tweaked with to get results that are ‘statistically significant,’ which is skimming the edge of what’s legal and what is not.
Credit: Reddit / @Holdin_McGroin