"The greatest amount of life can be supported by great diversification of structure." Charles Darwin

Sunday, March 2, 2014

Are scientists fooling themselves?

I came across this article recently published on Nature "Statistical errors: P values, the 'gold standard' of statistical validity, are not as reliable as many scientists assume" by Regina Nuzzo, and immediately connect to the reproducibility crises in biomedical research in recent years. I still remember that early in my graduate school one professor told me that those best stories do not actually require statistics, because the results are just black and white. I still believe in the power of statistics in biology. However, it should not be used to validate what we want to see at the first place.