Points of Significance: Split plot design

Journal name:
Nature Methods
Volume:
12,
Pages:
165–166
Year published:
DOI:
doi:10.1038/nmeth.3293
Published online

When some factors are harder to vary than others, a split plot design can be efficient.

At a glance

Figures

  1. Split plot design examples from agriculture.
    Figure 1: Split plot design examples from agriculture.

    (a) In CRD, levels of irrigation and fertilizer are assigned to plots of land (experimental units) in a random and balanced fashion. (b) In RCBD, similar experimental units are grouped (for example, by field) into blocks and treatments are distributed in a CRD fashion within the block. (c) If irrigation is more difficult to vary on a small scale and fields are large enough to be split, a split plot design becomes appropriate. Irrigation levels are assigned to whole plots by CRD and fertilizer is assigned to subplots using RCBD (irrigation is the block). (d) If the fields are large enough, they can be used as blocks for two levels of irrigation. Each field is composed of two whole plots, each composed of two subplots. Irrigation is assigned to whole plots using RCBD (blocked by field) and fertilizer assigned to subplots using RCBD (blocked by irrigation).

  2. In biological experiments using split plot designs, whole plot experimental units can be individual animals or groups.
    Figure 2: In biological experiments using split plot designs, whole plot experimental units can be individual animals or groups.

    (a) A two-factor, split plot animal experiment design. The whole plot is represented by a mouse assigned to drug, and tissues represent subplots. (b) Biological variability coming from nuisance factors, such as weight, can be addressed by blocking the whole plot factor, whose levels are now sampled using RCBD. (c) With three factors, the design is split-split plot. The housing unit is the whole plot experimental unit, each subject to a different temperature. Temperature is assigned to housing using CRD. Within each whole plot, the design shown in b is performed. Drug and tissue are subplot and sub-subplot units. Replication is done by increasing the number of housing units.

  3. The split plot design with CRD is commonly applied to a repeated measures time course design.
    Figure 3: The split plot design with CRD is commonly applied to a repeated measures time course design.

    (a) Basic time course design, in which time is one of the factors. Each measurement uses a different mouse. (b) In a repeated measures design, mice are followed longitudinally. Drug is assigned to mice using CRD. Time is the subplot factor. (c) Drug is blocked by housing. (d) A three-factor, repeated measures split-split plot design, now including tissue. Tissue is subplot and time is sub-subplot.

References

  1. Krzywinski, M., Altman, N. & Blainey, P. Nat. Methods 11, 11871188 (2014).
  2. Krzywinski, M. & Altman, N. Nat. Methods 11, 699700 (2014).
  3. Ganju, J. & Lucas, J.M. J. Stat. Plan. Infer. 81, 129140 (1999).
  4. Krzywinski, M., Altman, N. & Blainey, P. Nat. Methods 11, 977978 (2014).

Download references

Author information

Affiliations

  1. Naomi Altman is a Professor of Statistics at The Pennsylvania State University.

  2. Martin Krzywinski is a staff scientist at Canada's Michael Smith Genome Sciences Centre.

Competing financial interests

The authors declare no competing financial interests.

Author details

Additional data