Correlation Theory and Method Applied to Agricultural Research

Excerpt from the report Introduction: In no one volume are the theory and methods of correlation as applied to agricultural research in the Bureau of Agricultural Economics brought together in a form readily adapted to reference purposes. On the other hand not a few new statistical methods have been developed by members of the staff. This publication is an attempt to bring together and coordinate such methods. In order to be complete it presents usual correlation theory, but from a point of view easily adaptable to include the more recent theory and method. The approach is essentially that which has been given in the Graduate School of the Department of Agriculture for the past two years and is very similar in its treatment of simple correlation to that found in Prof, Frederick C. Mill's excellent text, "Statistical Methods Applied to Economics and Business". The new subject matter on multiple linear and curvilinear correlation, joint relationships, application to time series, and apportionment of importance to contributing variables is founded partly on articles published by members of the staff and partly on material as yet unpublished. Statisticians are especially indebted to H. R. Tolley and Mordecai Ezekiel of this Bureau for the notable contributions they have made to correlation methods, designated later in this work.