Split-Plot Design

A split-plot design is an experimental design in which the levels of one or more experimental factors are held constant for a batch of several consecutive experimental runs, which is called a whole plot. The levels of the remaining factors are varied during these experimental runs, and each level combination is considered as a subplot within the whole plot. Split-plot designs therefore consist of two types of experimental units: whole plots and subplots, where the subplots are nested within the whole plots. A split-plot design results in correlated responses for the experimental runs in the same whole plot. A correct analysis of the data from split-plot designs should take this correlation into account. In practice, split-plot designs are often used inadvertently, thereby often ignoring the ...