True, but in the case of a reasonably large, reasonably well distributed sample set, like performance of schoolchildren, that statement is tautologically absurd the vast majority of the time.