[Skip to Content]
[Skip to Content Landing]
August 29, 1959


Author Affiliations

Philadelphia. University of Pennsylvania

JAMA. 1959;170(18):2231-2232. doi:10.1001/jama.1959.03010180083022

This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.


To the Editor:—  I should like to offer the following comments on Dr. Vollman's letter. First, in simple linear correlation there is no implication of a cause and effect relationship merely because one variable is referred to as the independent variable. Actually, the coefficient of correlation is identical regardless of which is made dependent and which independent. Hence, the association observed in the Sevag-Colton data would be the same even if the length of the cycle were, as Dr. Vollman suggests, the dependent rather than the independent variable. A variable is assigned the independent position in the equation because one plans to use it to estimate the other. If one is seeking to achieve pregnancies it seems likely that he will try to estimate the most favorable day for insemination or coitus, using the length of the menstrual cycle as the independent variable.Second, the fact that averages were used

First Page Preview View Large
First page PDF preview
First page PDF preview