3 ms·
My limited understanding is that the simplistic formula for computing p-value assumes certain things about the lack of hidden correlations and effects. If those
by analog31 12y ago
My limited understanding is that the simplistic formula for computing p-value assumes certain things about the lack of hidden correlations and effects. If those things are present and correctly taken into account, then the correct p-value would be much higher. Arbitrarily multiplying reported p-values by 10 would not seem to be too far from the mark.