第五章_实验设计.pptVIP

2.4 Nested Designs All analyses of variance are concerned with one or more factors or variables of classification. These factors can be of two types: 1. Main effects: Every level can be identified independently of any other lever (factorial or crossed designs). 2. Nested effects: If each of the levels of a main effect can be subdivided into randomly selected subgroups, the classification of the group is said to be nested. Nested designs always have at least one main effect, which may be a fixed or random effect. Nested effects are always random effects. Nested