Methods for increasing the accuracy of experiments

William G. Cochran & Gertrude M. Cox

The results of experiments are affected not only by the action of the treatments, by also by extraneous variations which tend to mask the effects of the treatments. The term experimental errors is often applied to these variations, where the word errors is not synonymous with ‘mistakes’, but includes all types of extraneous variations. Two main sources of experimental errors may be distinguished. The first is the inherent variability in the experimental material to which the treatments are applied. We shall use the term experimental unit to denote the group of material to which a treatment is applied in a single trial of the experiment. The unit may be a plot of land, a patient in a hospital, or a lump of dough, or it may be a group of pigs in a pen, or a batch of seed. It is characteristic of such units that they produce different results even when subjected to the same treatment: these differences, whether large or small, contribute to the experimental errors. The second source of variability is lack of uniformity in the physical conduct of the experiment, or in other words, failure of standardize the experimental technique.

Fonte: Cochran, WG & Cox, GM. 1957. Experimental designs. NY, Wiley.