ABSTRACT
The previous chapter dealt with the comparison of a sample and a theoretical parameter. When the theoretical parameter is a control or standard value, this value is often not known precisely under the particular conditions of the experiment, so the investigator also ob tains a series of control observations. If the experimental and control observations are paired on nuisance characteristics in order to elim inate their effects, then individual differences should be computed for each pair, and the problem remains a one-sample problem of comparing the mean difference with zero. When it is not necessary to pair the experimental and control series, the problem becomes a two-sample problem.