ABSTRACT
In any empirical investigation or experimentation, experimental error is a fact of life. It would be very nice if tests or experiments conducted at the same factor settings would always give the same result. In the real world this is not the case and variability in data is something that must be expected. Figure 2.1 illustrates the problem. Although experimenters may hold the controllable factors constant, differences in experimental units, unknown changes in background or lurking variables, and measurement errors will cause the response in repeat experiments at the same factor settings to vary in what seems to be a random pattern.
