Components of variance

Abstract

So far, the examples of analysis-of-variance problems that we have considered all fall within the scope of what we call the ‘fixed effects’ analysis, or the ‘model I’ situation. In these problems the questions being asked are questions about contrasts of sets of means. For example, in Example 10.1 we wanted to know what differences there were between the mean yields of treatments A, B, C and D regarded as specific individuals. We now consider a different type of analysis-of-variance problem, suitable for application of what we call the ‘random effects’ or the ‘model II’ analysis. Consider the following example.