A VIEW OF STATISTICAL METHOD.
Abstract The article presents information on the use and development of statistical method. All things vary. There is no exact science. The best that one can do is lay down rules of action that may be used for doing things in spite of variations. The more "exact" science becomes, the more evident it is that the ultimate development of science leads to the statistical method. The statistical method provides a way of setting probability tolerances on chance variations and of pointing to the existence of variations that arise from assignable causes. The statistician possesses the tools for doing these things and for collecting data to make his tools effective in formulating rational courses of action. There are two kinds of variations and it is the task of the statistician to distinguish between them. Some variation arise from causes than can not be traced, or should not be, because one almost always only wastes time and money in trying to trace them. These are chance causes. The other kind of variation arises from all assignable cause, a cause that one is justified in looking for. The statistician must distinguish between these two different kinds of variations.
- DOI
- 10.2308/tar-7036890
- Volume
- 19 (3)
- Pages
- 254-260
- Language
- en
- Export
- BibTeX
- Sources
- openalex crossref