Choice of sample size in parasitological experiments

One of the most awkward practical problems confronting a research worker setting up an experiment is to decide how big the experiment should be. The larger the number of subjects then the smaller is the degree of uncertainty in the conclusions; but large trials are costly and may have to be of long duration, so there is every incentive to keep trials as small as possible. Statistics is the science of coping; with uncertainty, and in this article Michael Healy explains the statistical basis that helps to decide how big a sample size should be - how to balance minimum error rates with optimal cost efficiency.