My point is that you can't explain a statistic with one cause. Most of the time there are many. Some of them we know, some of them not.