3 ms·
Well said. It doesn't reveal problems with current methodology, it reveals problems with its application, and understanding. Larger sample sizes would fix quite
by ineedasername 9y ago
Well said. It doesn't reveal problems with current methodology, it reveals problems with its application, and understanding. Larger sample sizes would fix quite a bit. Studies, even "pilot" or "priming" studies, are not reliable at a .05 significance. Especially when most of the false positives will get published, but nearly none of the false negatives will see the light of day.