ABSTRACT
We have seen that, in a contingency table, if the two variables are independent the distributions of all parallel arrays are similar to each other and to the distribution in the total universe of discourse. In such a case the means of every parallel array would be the same. Thus, suppose the length of the head was entirely independent of the breadth in any particular race of people. If we selected different groups of persons whose head-lengths were between different assigned limits we should expect their mean head-breadths to be approximately the same; otherwise one would be in some way dependent on the other. https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315009810/61a4cc26-c726-4977-a558-75fbf99de4a5/content/fig17_B.tif" xmlns:xlink="https://www.w3.org/1999/xlink"/>
