What is a mixed or split plot design?
A split-Plot design is a "blocked experiment" in which the blocks serve as experimental units for a subset of factors" (Jones & Nachtshem 2009). In other words, according to Jones and Nachtshem, the blocks are considered as whole plots, and contained within the plots are split plots, split units or space plots. They provide an explanation in the following example. Consider Factor A and two fertilizers methods (Factor B) on the yield of a crop, using four available fields as experimental units. In the irrigation, it is not possible to apply different irrigation methods (factor A) in areas smaller than a field. However, different fertilizer types (Factor B) could be applied in smaller areas. Hence, one whole plot (filed) will be divided into split plots, and each of the fertilizer types can be applied once within each whole plot (p. 340). Based on the split plot design, the two irrigation types are randomly assigned to the four field (whole plots). Jones and Nachtshem explain why one would use a split-plot design, and provide three potential reasons: (a) cost, (b) efficiency, and (c) validity. Cost suggests that a ...
This solution describes the nature of split-plot or mixed designs.