ARCSPLITPLOT procedure

Options

No parameters

Description

ARCSPLITPLOT can be used to superimpose additional treatments onto the replicates of a row-column design, for example formed by CDNROWCOLUMNDESIGN, so that the design becomes a split-plot with the original replicates as the whole plots. It can also be used to generate row and column factors giving the locations of the plots of the row-column design within the entire experiment. (The row and column factors usually generated for the row-column design merely index the rows and columns within each replicate.)

The factor for the replicates of the row-column design is specified by the RCREPLICATES option. The RCROWS and RCCOLUMNS options specify the factors for the rows and the columns within those replicates.

The locations of the replicates within the whole design must be specified by the REPLOCATIONS option. This can supply a variate containing the numbers of replicates in adjacent columns of the design, like the REPLATINGROUPS option of the CDNROWCOLUMNDESIGN procedure. For example, setting REPLATINGROUPS=!(2,2,2) defines three columns of replicates, the first containing replicates 1 and 2, the second containing replicates 3 and 4, and the third column containing replicates 5 and 6. Alternatively, you can supply a matrix with each cell containing the number of the replicate at that location.

This provides the information needed to generate the factors to identify the row and column locations of the plots within the whole design. They can be saved using the ROWS and COLUMNS options, respectively.

To superimpose the extra treatments, ARCSPLITPLOT also needs to know how to define the replicate factor for the extended design. This is specified by the METHOD option, with settings:

The identifiers of extra treatment factors are specified by the TREATMENTFACTORS option. The LEVELS option can be used to define the numbers of levels of those factors, as a scalar if there is only one factor, or as a variate if there are several. The levels specified in the variate are assumed to be in the same order as the order in which the factors occur in the TREATMENTFACTORS list. LEVELS can be omitted if the factors have already been declared with the right numbers of levels. The REPLICATES option specifies the identifier of the replicate factor, and the WHOLEPLOTS option specifies the identifier of the whole-plot factor.

Printed output is controlled by the PRINT option, with settings:

The SEED option allows you to supply a seed for the random numbers used to randomize the allocation of the extra treatments. The default value of zero continues an existing sequence of random numbers if any have already been used in the current Genstat job, or obtains a random seed using the system clock if none have been used already. You can also set SEED=-1 if you want to suppress any randomization.

You can set option SPREADSHEET=design to put the factors into a spreadsheet.

Options: PRINT, LEVELS, TREATMENTFACTORS, REPLICATES, WHOLEPLOTS, ROWS, 
COLUMNS, RCREPLICATES, RCROWS, RCCOLUMNS, REPLOCATIONS, METHOD, SEED, SPREADSHEET.

Parameters: none.

See also

Procedure: CDNROWCOLUMNDESIGN.
Commands for: Design of experiments.

Example

CAPTION      'ARCSPLITPLOT example'; STYLE=meta
VARIATE      [VALUES=6] replocations
" Design constructed by CDNROWCOLUMNDESIGN using CycDesigN 
  This command can used in place of the FACTOR/READ statements below
  if you have a CycDesigN license 
CDNROWCOLUMNDESIGN [PRINT=design; LEVELS=15; TREATMENT=sptreat;\
             REPLICATES=rcreps; ROWS=rcrows; COLUMNS=rccols;\
             NREPLICATES=6; NROWS=5; NCOLUMNS=3; RESOLVABLE=yes;\
             NRLATIN=1; REPLATINGROUPS=replocations;\
             SEED=74674; TIMELIMITS=0.25] "
FACTOR [NVALUES=90; LEVELS=6] rcreps
READ rcreps
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 
3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 6 
6 6 6 6 6 6 6 6 6 6 6 6 6 6 :
FACTOR [NVALUES=90; LEVELS=5] rcrows
READ rcrows
1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 1 1 1 2 2 2 3 3 
3 4 4 4 5 5 5 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 1 
1 1 2 2 2 3 3 3 4 4 4 5 5 5 :
FACTOR [NVALUES=90; LEVELS=3] rccols
READ rccols
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 
3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 
2 3 1 2 3 1 2 3 1 2 3 1 2 3 :
FACTOR [NVALUES=90; LEVELS=15] sptreat
READ sptreat
12 14 3 1 15 6 11 5 8 9 7 4 10 2 13 3 12 6 5 7 14 9 11 10 1 4 13 15 8 2 4 12 
10 5 13 9 6 7 8 11 15 14 3 1 2 4 11 2 15 9 12 3 7 1 14 13 8 10 6 5 1 11 14 6 
2 9 10 3 8 5 15 4 7 13 12 14 6 4 9 8 1 3 11 13 7 15 10 2 12 5 :
PDESIGN      [BLOCKS=rcreps/(rcrows*rccols); TREATMENTS=sptreat]
ARCSPLITPLOT [PRINT=design,layout; LEVELS=2;\
             TREATMENTFACTORS=wptreat; REPLICATES=replicates;\
             WHOLEPLOTS=wplots; ROWS=rows; COLUMNS=columns;\
             RCREPLICATES=rcreps; RCROWS=rcrows; RCCOLUMNS=rccols;\
             REPLOCATIONS=replocations; METHOD=rows; SEED=945900
" whole-plot and sub-plot treatments in field layout "
PDESIGN      [BLOCKS=rows*columns; TREATMENTS=wptreat,sptreat
" whole-plot and sub-plot treatments in replicates, rows and columns "
PDESIGN      [BLOCKS=rcreps/(rcrows*rccols); TREATMENTS=wptreat,sptreat]