12 Nested and split-plot designs
The designs so far have crossed every factor with every other: each level of one appears with each level of the others. Two common structures break this pattern. In a nested design the levels of one factor are different for each level of another, so the two cannot be crossed. In a split-plot design the factors are crossed, but the randomisation is restricted, so that different factors have different experimental units and different errors. Both are analysed with the random-effects tools of Section 11.
12.1 The two-stage nested design
12.2 The split-plot design
12.3 Split plots with random factors
12.4 Practice problems
12.2 The split-plot design
12.3 Split plots with random factors
12.4 Practice problems
Questions on this section
Stuck on something here? Ask below and it stays attached to this topic.