Abstract: This paper discuss on errors disappearing in some empirical researches, such as defining error of index concept, mistiness of index calculating, etc. Some suggestions are put forward to governmental statistical departments as well as researchers.