List the analyzable layouts of an augmented row-column field

Tries every combination of block size and number of checks for a field and returns the layouts that can be analysed, so a layout can be chosen without writing a loop over augmented_row_column_capacity(). Each block gets one plot of every check. Suggested blocks have at least as many rows and at least as many columns as there are checks, leaving room for each check to occupy a distinct row and column.

Usage

augmented_row_column_layouts(
  rows_in_field,
  cols_in_field,
  rows_per_block = NULL,
  cols_per_block = NULL,
  n_controls = 2:6,
  min_error_df = 1,
  n_entries = NULL
)

Arguments

rows_in_field, cols_in_field

Field size.

rows_per_block, cols_per_block

Block sizes to try. By default every divisor of the field size from 2 up to the field size.

n_controls

Numbers of checks to try.

min_error_df

Smallest error df to keep.

n_entries

Optional number of entries to plant. Only layouts with at least that many entry plots are kept, and spare_plots shows how many entry plots would be left over.

Value

A tibble with one row per analyzable layout, sorted by entry plots (most first) and then error df: block sizes, number of checks, blocks, entry and check plots, check_fraction, error_df, method (the construction that augmented_row_column_design() uses for it: "latin" or "optimal") and, when n_entries is given, spare_plots.

Details

There is a trade-off: more check plots mean fewer entries but more error degrees of freedom, i.e. more reliable F-tests and spatial adjustment. Aim for at least about 10 error df when the field allows it (Clarke and Stefanova, 2011, recommend about 50 for large trials).

See also

augmented_row_column_capacity(), augmented_row_column_design()

Examples

# Suggested analyzable layouts of a 12 x 12 field with 2 to 5 checks:
augmented_row_column_layouts(12, 12, n_controls = 2:5)
#> # A tibble: 9 × 9
#>   rows_per_block cols_per_block n_controls entry_plots error_df check_plots
#>            <int>          <int>      <int>       <int>    <dbl>       <int>
#> 1              4              4          4         108        6          36
#> 2              4              3          3         108        5          36
#> 3              3              4          3         108        5          36
#> 4              4              2          2         108        2          36
#> 5              2              4          2         108        2          36
#> 6              3              3          3          96       14          48
#> 7              3              2          2          96        9          48
#> 8              2              3          2          96        9          48
#> 9              2              2          2          72       23          72
#> # ℹ 3 more variables: n_blocks <int>, check_fraction <dbl>, method <chr>

# Layouts that fit 100 entries with at least 10 error df:
augmented_row_column_layouts(12, 12, n_controls = 2:6, min_error_df = 10, n_entries = 100)
#> # A tibble: 0 × 10
#> # ℹ 10 variables: rows_per_block <int>, cols_per_block <int>, n_controls <int>,
#> #   entry_plots <int>, error_df <dbl>, check_plots <int>, n_blocks <int>,
#> #   check_fraction <dbl>, method <chr>, spare_plots <int>