International Journal of
Information and Education Technology

Abstract

The orthogonal array and sample size determine the experimental cost. Selection of an appropriate sample size is one the most important aspects of any experimental design problem. The choice of sample size and the probabilities of type I error and type II error are closely connected in traditional experimental design. We can specify the and , and then decide the standard deviation based on the prior experience to calculate the ratio, , of absolute value for difference in means dividing by standard deviation. The required sample size n is obtained from operating characteristic curve. The purpose of this paper is to present economical models for selecting an appropriate sample size based on the testing cost of observations and loss by adopting a wrong level and probability of factor. Some tables are provided that can be used to select sample size for the variances of two populations are known or unknown.