2 ms·

I remembr a discussion here recently regarding "p-hacking". With so many thousands of genes, aren't a fair number going to be correlated or predictive with stat

I remembr a discussion here recently regarding "p-hacking". With so many thousands of genes, aren't a fair number going to be correlated or predictive with statistical significance by sheer probability?

pjscott 9y ago
The more genetic variants they test, the more they correct for multiple comparisons. https://en.wikipedia.org/wiki/Genome-wide_association_study#Methods https://en.wikipedia.org/wiki/Genome-wide_association_study#...