1. Strip Plot Design

2. Strip Plots
Experimental units for factors are row strips (row factor) and column strips (column factor)
Convenient design for two-factor agricultural experiments
Alternative to split-plot design when split-plots are not practical as experimental units for either factor

3. Strip Plot Alternatives
The design requires replication of both row and column strips
Just as with split plot design, we can replicate across blocks, or replicate within a single block (i.e., plot)

4. Strip Plot Model I
Yandell Model
S=set. Draw diagram

5. Strip Plot I Discussion
In Yandell’s model, sk indexes set , which actually functions as a block
It’s surprising that Yandell doesn’t present nested strip plot errors
Strip plot errors are really just block (or set) by treatment interactions

6. Strip Plot Model II It may be more proper to consider the model as:
Draw layout (assume s=1)

7. Strip Plot Model III
Another version of the strip plot model (consistent with Yandell’s split plot design):

8. Strip Plot Design I & II ANOVA

9. Strip Plot Design I & II ANOVA (cont)

10. Strip Plot Design III ANOVA

11. Strip Plot Design III ANOVA (cont)

12. Pairwise Contrasts
Main effects contrast estimates are straightforward to compute, though their standard errors vary:
E.g.
Write out ybar(i..), E(ybar(i..)), V(ybar(i..))

13. Strip Plot Design LSMEANS
Use, e.g,. /E=A*S in LSMEANS to assign correct error term in PROC GLM
Standard interaction contrast estimates are straightforward as well
We did not discuss this, but split plot interaction contrasts involving cell means could involve comparisons across subplots, or plots or both.
Think how messy some of these comparisons could be based on the previous slide’s calculations

14. Strip Plot Design LSMEANS (cont)
This same problem occurs for interaction contrasts in the strip plot design
Care must be taken when estimating these contrasts
Yandell somewhat overstates severity of this problem

15. Strip Plot Example
Strip plot design with Variety (DPL 454BR, DPL 555BR, ST 5599BR, PHY 375WRF) as row factor and Fungicide (Ridomil, Terraclor, Quadras) as column factor
Fields are block
Nested row and column strip plots
Response is yield (cotton lint in pounds/acre)
Run SAS code. The model is a hybrid of the designs we listed.