Description Usage Arguments Details Value References Examples
View source: R/design.mad_function_vFldTrial.R
design.mad
generates a field layout whose various attributed originate from Lin and Poushinsky's modified augmented desing (type 2) (Lin and Poushinsky, 1985), with modifications made by the Barley CAP (http://www.barleycap.org) to accomodate standard row crop field plots (Figure 2). See Details
for more information.
1 2 3 4 5 | design.mad(enviro = format(Sys.Date(), "%x"), exp.name = NULL,
entries = NULL, nEntries = NULL, chks = NULL, nChk2 = NULL,
nFieldRows = NULL, nFieldCols = NULL, nChk2.min = 3,
plot.start = 1001, plot_id.template = NULL, minPerChks = 0.1,
maxPerChks = 0.5, fillWithEntry = T, dup.list = NULL, fillWithChk = F)
|
enviro |
Optional |
exp.name |
Required |
entries |
A |
nEntries |
|
chks |
A |
nChk2 |
|
nFieldRows |
Optional, but must be provided if nFieldCols is not. |
nFieldCols |
Optional, but must be provided if nFieldRows is not |
nChk2.min |
Optional |
plot.start |
Optional |
plot_id.template |
Optional |
minPerChks |
Optional |
maxPerChks |
Optional |
fillWithEntry |
Optional |
dup.list |
Optional |
fillWithChk |
Optional |
The modified augmented design (type 2) was originally proposed by Lin and Poushinsky (1985). The design originally arranged experimental entries in a grid pattern consisting of "whole plots" (WPs), with the dimension of each whole plot being 1 row x 5 columns. At the center of each WP was a "control" or check plot denoted as the primary check. Furthermore, a subset of the WPs also contained secondary check lines.
See Figure 1
from Lin and Poushinsky (1985) for details on the original (type 2) modified augmented desging. This type of design allowed for multiple adjustment procedures to correct for spatial heterogeneity (i.e. spatially-related error variance) (Lin and Poushinsky, 1985).
Because of its heavy use of check plots (20 percent guaranteed from primary checks alone!) the traditional type 2 MAD was modified by the Barley CAP (http://www.barleycap.org). The dimesion of the WPs have been changed to 3 rows x 5 columns (Figure 2), and the center plot is still retained as the "control" or check plot. Secondary checks (typically 2-3) are still assigned to a subset of the blocks.
The advantage of this modification is that fewer check plots are required, however it places a stronger restriction on overall field size, specifically that the field dimensions must be reducible by the 3 row x 5 column WP sizes. The adjustment procedures for correcting spatially-related error from Lin and Poushinsky (1985) are retained and described in more depth in the documentation for the function adjust.fld.
Figure 1:
Figure 2:
MAD may not be well-suited for all circumstances; in these cases consider using other design functions included in FldTrial: design.rcbd or design.aibd
mad.dgn A dataframe
of the resulting design. A line code of 0 denotes an experimental entry, codes of 1, 2, ..., (nChk2
+ 1) denote the primary, first secondary, up to the (nChk2
+ 1)-th secondary check, respectively.
If fillWithEntry
is TRUE
then a "D" will be placed in the "Duplicates" column for all duplicated entries.
Note that the column name "replication" is used to denote whole block replication, whereas "blk" denotes plot assignments to incomplete blocks.
plot.layout A matrix
of plot numbers reflecting the field layout. The first column contains field row numberis (i.e. 1:nFieldRows
) and the last row contains field column numbers (i.e. 1:nFieldCols
).
check.layout A matrix
of line codes reflecting the field layout. See plot.layout
for further details.
blk.layout A matirx
of block assignments reflecting the field layout. See plot.layout
for further details.
field.dims The dimensions of the resulting field design; nFieldRows x nFieldCols, or a x c (Figure 1).
nDuplicates The number of duplicated experimental entries. Will only be non-zero if fillWithEntry
is TRUE
.
rlzPerChks The final percentage of the designed represented by check lines. This may be different than minPerChks
, especially when fillWithCheck
is TRUE
.
Lin, C.-S. and G. Poushinsky. 1985. A modified augmented design (type 2) for rectangular plots. Can. J. Plant Sci. 65:743-749.
1 2 3 4 5 6 7 8 9 10 11 12 13 | ## Not run:
## Example 1 - Provide only nFieldRows
mad.ex1 <- design.mad(exp.name = "ex1", nEntries = 250,
nChk2 = 2, nFieldRows = 9)
## Example 2 - Provide entries and checks
Provide both nFieldRows and nFieldCols
mad.ex2 <- design.mad(exp.name = "ex2",
entries = paste("Line", 1:300, sep = "."),
chks = c("Larry", "Curly", "Moe"),
nFieldRows = 12, nFieldCols = 30)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.