Construction of row-column designs by CycDesigN

This menu uses the CycDesigN package for the computer generation of experimental designs to form a row-column design with a single treatment factor. See the CycDesigN manual or the book "Cyclic and Computer Generated Designs" by John & Williams (1995) for more details about the underlying methods and theory.

Number of treatments

A scalar specifying the number of treatments.

Number of replicates

A scalar specifying the number of replicates of each treatment.

Number of rows

A scalar specifying the number of rows.

Number of columns

A scalar specifying the columns.

Seed

A scalar specifying a seed for the random numbers used by CycDesigN to randomize the design.

Store design in spreadsheet

Whether to put the design factors into a spreadsheet.

Printed output

designdisplays the design
reportdisplays a report from CycDesigN about the design

Two-stage construction

Whether or not to form the design in two stages. The first constructs the column component design, and the second forms the row-column design. Two-stage construction is faster and may be advisable with larger designs, but one-stage construction should find a more efficient design.

Resolvable

Whether or not to form a resolvable design i.e. one where whether the rows and columns can be grouped into replicates, each of which contains a single replicate of each treatment combination.

Latinize rows if resolvable

If this is set to 1, the rows of a resolvable design are latinized. The replicates are assumed to be set out in a grid, and the treatments are allocated to occur as equally as possible in each row. If this is set to an integer value t greater than 1 (and less than the number of rows) a t-latinized design is produced, in which the treatments occur as equally as possible in groups of t adjacent rows. The structure of the grid of replicates is defined by the variate specified by Numbers of replicates in each column (for latinizing).

Latinize columns if resolvable

In a similar way, this latinizes the columns of a resolvable design.

Numbers of replicates in each column (for latinizing)

This defines the layout of the replicates when the design is to be latinized. The idea is that the replicates are arranged in adjacent columns. So this should be set to a variate whose values specify the numbers of replicates in each of the columns in turn, running from left to right across the field. If this is unset, the replicates are assumed to be in a single column.

Treatments

A factor to save the levels generated for the treatment factor.

Replicates

A factor to save the levels generated for the replicate factor.

Rows

A factor to save the levels generated for the row factor.

Columns

A factor to save the levels generated for the column factor.