查看文章

In a very interesting paper which was published in 1950, scholar WS Robinson pointed out that ecological correlations cannot be used as substitutes for individual correlations. He stated the purpose of his paper will be accomplished" if it prevents the future computation of meaningless correlations and stimulates the study of similar problems with the use of meaningful correlations between the properties of individuals." Since 1950, many papers which base conclusions on ecological correlations have been published. The purpose of this note is to help clarify this problem. It emphasizes the point that in general the study of the regression between ecological variables cannot be used as substitutes for the study of the behavior of individuals. It will also point out that in very special circumstances the study of the regression between ecological variables may be used to make inferences concerning the behavior of …