ABSTRACT
There are two ways to approach the question of determining the sample size, either by using theory or by considering the practicalities of various sample size possibilities. If one is really clever he would include both of these methods in his calculations. The sample size is connected to the level of accuracy one desires in his results. The greater the level of accuracy, the bigger the sample size. The chapter on sample surveys in Anderson et al, gives a basic introduction to sample size calculations. A common way of determining sample size is by including as many subjects in the survey as cost/time will allow. Another way is purely by gut feeling. It may be obvious that only 10 subjects are not enough, but that 10,000 are totally unnecessary. One needs to find a number between these two figures that provides enough information to give at least some validity to the results, that also within his time/budget constraints.
