3 ms·
I appreciate a study like this that's very straightforward with the limitations. 1. Some people say X. 2. We tested X. 3. Actually, it looks like X might be
by kevinpet 5y ago
I appreciate a study like this that's very straightforward with the limitations.
1. Some people say X.
2. We tested X.
3. Actually, it looks like X might be true.
4. So it would be worth doing a more powerful study.
- AshamedCaptain 5y agoThis is not. They are testing a shitton of variables so it's highly likely that by pure chance something "statistically significant" would appear. But not even that happened (there was NO significant change at all), so the recommendation of a "more powerful study" is absolutely baseless. If anything, the conclusion should be "we wasted our time so you don't have to waste yours in the future". Nothing to be ashamed of. Otherwise this is how you start another bunch of "science reproducibility crisis!".