Tables of Optimal Allocations of Observations for Comparing Treatments with a Control

摘要

The problem under consideration is that of estimating simultaneously the differences between the means of p > or = 2 test treatments and the mean of a control treatment when the population variances of all p + 1 treatments are known. Tables are given which permit the experimenter to find the minimal total number of experimental units, and the optimal allocation of these units among the P + 1 treatments, in order to make one-sided or two-sided confidence interval estimates of the differences of interest. These intervals achieve a specified joint confidence coefficient 1-alpha for a specified allowance or yardstick associated with the common width of the intervals. The computations for these tables are based on the results of Bechhofer (1969) for one-sided comparisons, and Bechhofer and and Nocturne (1972) for two-sided comparisons. (Author)