Sure... but conversely, if you keep dividing up and analyzing in different ways, eventually you'll get a p<0.05 result. This is why you "call your shot" before

Sure... but conversely, if you keep dividing up and analyzing in different ways, eventually you'll get a p<0.05 result. This is why you "call your shot" before beginning the analysis. Otherwise, people will be suspicious you just jiggered things around until you found something.