ABSTRACT
Two important questions on the designing of any sample inquiry are the total cost of the survey and the precision of the main estimates. Both these are related to the size of the sample, given the variability of the data, the type of sampling and the method of estimation. Obviously, the larger the sample, the smaller will be the sampling error, i.e. the greater the precision of the estimates, but the higher will also be the cost. The survey should be so designed as to provide estimates with minimum sampling errors (i.e. with maximum precision) when the total cost is fixed, and to result in the minimum total cost when the precision is preassigned: a sample size fulfilling these conditions is called the optimal sample size. We shall see later in Chapter 25 that other considerations such as the existence of nonsampling errors and biases in data have also to be taken into account: these generally increase with sample size beyond a certain point.
