Forms factors to represent carry-over effects in cross-over trials (R.W. Payne).
Option
NONELEVEL = scalar or text |
Level or label to use for the units with no carry-over |
|---|
Parameters
TREATMENTS = factors |
Factors identifying the (direct) effects of the treatments |
|---|---|
SUBJECTS = factors |
Factors identifying the subjects |
PERIODS = factors |
Factors identifying the periods |
CARRYOVERFACTOR = factors |
Factors to represent the carry-over effect of the treatments in the period immediately after the period in which they were applied |
NOCARRYOVER = factors |
Factors to represent the comparison between none and any carry-over effect of the treatments |
Description
Cross-over trials are designed to study the effects of various treatments on a set of plots (in a field experiment) or subjects (in a medical trial). The special feature of these experiments is that the same plots or subjects are treated during several successive time periods, and there is interest both in the direct effect of a treatment during the period in which it is applied and its carry-over (or “residual”) effect during later periods.
AFCARRYOVER can be used to construct the factors required to represent the carry-over effects. To do this it requires factors to identify the treatment, subject (or plot) and period corresponding to each unit of the data. These must be specified by the TREATMENTS, SUBJECTS and PERIODS parameters respectively. The NONELEVEL option can be used to supply a scalar to specify the level, or a single-line text to specify the label, to be used to represent the “no carry-over” treatment which occurs during the first period. If the level is not specified, an appropriate value is found automatically: zero if that is not an existing treatment level, or the minimum treatment level minus one otherwise.
Two factors can be generated. The CARRYOVERFACTOR parameter provised a factor that simply identifies the carry-over treatment on each unit, while the NOCARRYOVER parameter provides a factor representing the comparison between the units with no carry-over and those with any type of carry-over (essentially this is a comparison between the periods 2 onwards where there were carry-over effects from earlier times, and period 1 where there was none). The NOCARRYOVER factor may be required to be able to analyse the design using ANOVA (see the description of procedure AGCROSSOVERLATIN).
Option: NONELEVEL.
Parameters: TREATMENTS, SUBJECTS, PERIODS, CARRYOVERFACTOR, NOCARRYOVER.
Method
AFCARRYOVER generates the factors using the standard Genstat calculation and manipulation commands.
See also
Directive: ANOVA.
Procedure: AGCROSSOVERLATIN.
Commands for: Design of experiments.
Example
CAPTION 'AFCARRYOVER example',\
!t('This generates the carry-over factor for a cyclic',\
'cross-over design.'); STYLE=meta,plain
AGCYCLIC [PRINT=design; METHOD=changeover] 7; NBLOCKS=14; NUNITS=4;\
TREATMENT=Direct; BLOCKS=Subjects; UNITS=Periods; SEED=23156
AFCARRYOVER Direct; SUBJECTS=Subjects; PERIODS=Periods;\
CARRYOVERFACTOR=Carryover; NOCARRYOVER=Nocarryover
PDESIGN [BLOCKSTRUCTURE=Subjects*Periods;\
TREATMENTSTRUCTURE=Direct,Nocarryover,Carryover]