The Effect of Increasing the Size of Samples in Certain Distributions

The problem of comparing two or more distributions has been under discussion for approximately 150 years. The classical method of comparing moments or functions of moments is without doubt the one which holds the favor of most of the eminent statisticians. It is not the purpose of this paper to enumerate the merits or demerits of any test of significance of differences of samples. Rather, a certain method of procedure is chosen, and in accordance with the hypotheses of this method the effect of increasing the size of the sample is shown. Comparative relations are easily shown with a few common statistics such as the standard deviations of standard deviations and means