Two and More Factors ANOVA - Agricultural Statistical Data Analysis Using Stata - page 109

Predictive Margins of Fert#Green with 95% Cls

30

25

20

15

Fallow
Ve tch

Barley
Barley-vetch

10

0 120

Fertilizer Rate (lbs/acre nitrogen)

Figure 6.3 Output from marginplots command showing interactions of fertilizer rates and green
manure effects on sugar beet yields (tons/acre).

The Margin column lists the means for each combination of fer-
tilizer and green manure. This will not always be the case as we
will see in covariance analysis. After this command is entered, the
marginsplot command can be entered (this command must fol-
low the margins command), which results in the graph shown in
Figure 6.3.
Both the fertilizer and green manure treatments had an effect on
sugar beet yields. There also was an interaction effect between the two
factors. An examination of the green manure effects with and with-
out fertilizer indicated treatments with vetch (vetch or vetch-barley)
appeared to have higher yields than green manures without vetch
regardless of fertilizer application. Fertilizer also had a significant
effect on yield for all green manures. The fertilizer effect, however,
was greater with barley and barley-vetch than with vetch alone or for
the fallow treatment.

Split-Block Design

The split-block design, which is also referred to as a strip-plot design,
is a derivation of the split-plot design. In this design, the first factor
is randomly assigned to plots in one direction and the second factor is
randomly assigned perpendicular to the first factor. This type of design
is often used where treatment application is applied by equipment (e.g.,
fertilizer or herbicide application equipment). Treatments are applied