# Randomize a Factorial Design from an Orthogonal Block Structure

### Description

Randomize a factorial design according to a specified block structure formula.

### Usage

1 | ```
planor.randomize(blockformula, data, out.order, keep.initial=FALSE)
``` |

### Arguments

`blockformula` |
the block structure formula. |

`data` |
a data frame. |

`out.order` |
a list of |

`keep.initial` |
if TRUE, the initial row order of the design
is stored in column |

### Value

The input data frame after randomization.

### Note

Each name in `blockformula`

must correspond to a factor
of the dataframe `data`

. The only exception is
`UNITS`

. If `UNITS`

is used in `blockformula`

but absent from `data`

, a factor is added to `data`

,
with one level per row. See the examples below for the usage of
`UNITS`

in `blockformula`

.

### Author(s)

Monod, H. and Bouvier, A.

### References

Bailey, R.A. (1983) Generalized wreath products of
permutation groups. *Proc. London Math. Soc.*, **47**, 69–82.

Kobilinsky A. (1989) Randomization of a cartesian block structure. Technical Report. Laboratoire de Biométrie. INRA. Versailles.

### Examples

1 2 3 4 5 6 7 8 9 10 11 12 | ```
### Block design
Design <- data.frame(block=rep(1:4,rep(2,4)),
treatment=c("A1","B1","A2","B2","A3","B3","A4","B4"))
### No within-block randomization
planor.randomize(~block, data=Design)
### Blocks and units within blocks randomization
planor.randomize(~block/UNITS, data=Design)
### Row-Column design
RowColDes <- data.frame(row=rep(1:3,rep(3,3)),col=rep(1:3,3),
treatment=LETTERS[c(1:3,2,3,1,3,1,2)],
oldRow=rep(1:3,rep(3,3)),oldCol=rep(1:3,3))
planor.randomize(~row*col, data=RowColDes)
``` |