This menu is for generating orthogonal arrays for experiments for which not all factors have 2 levels. The menu calls function function oa.design from package DoE.base and has been opened internally using the function shown in the usage section, which normal users will never have to deal with.
Menu.oa()The most important technical point about this menu and the R-Commander in general:
The R GUI should be installed with the Single Document Interface (SDI) mode
instead of the default MDI mode. John Fox, the author of R-Commander, has described
how to change MDI to SDI, if R has been installed in the wrong mode.
It is strongly discouraged to run the R-Commander under the MDI mode,
as it happens very frequently that windows suddenly disappear. They usually
have iconized only and can be retrieved from the taskbar; nevertheless, this behavior
can be very annoying and can be avoided by using R in SDI mode.
It is possible within this menu to handle situations with 2-level factors only. However, most of the time, treating experiments in 2-level factors with special functions for this purpose is both more convenient and yields more efficient designs. Thus, it is recommended to use the menu for 2-level designs, if all factors are at 2 levels only. At the screening stage, where you are not yet sure which factors are important, it is very often reasonable to conduct the experiment with each factor at only 2 levels!
Exceptions: special resolution V designs in 128, 256 or 2048 runs
This user form is structured as a notebook with several tabs, and each tab has its own specific help button. It is recommended to work through the tabs from left to right, although it is possible to switch between tabs back and forth.
The buttons to the right of the tabs store or load form settings (cf. next section) for the complete form with all tabs.
On OK, the menu will create an experimental design as an R data frame with
some attributes (desnum, run.order and design.info).
The list design.info contains, among other things, the element
creator, which contains the stored form.
If requested on the Export tab, the R workspace with ONLY the experimental
design will be saved at the specified location with the ending rda,
and an Excel-readable html-file or a csv-file will be generated, depending
on the users choice. Alternatively, it is possible to permanently store an
R workspace with ONLY the design later using the menu entry
“Export design” from the “Design” menu,
or the full R workspace can be stored using the appropriate entry in the
“File” menu of R commander.
The menu entry for loading the stored R workspace in a future session
can be found under “Data Management”.
The settings of the form (all tabs) can be saved at any time with the
button “Store form”, which generates an object (of class menu.design2pb).
The button “Load form” can be used for loading these settings
into the form again in order to continue working on the entries.
The purpose of this functionality: work can be safely interrupted,
or a finished design can be modified after a team discussion or ...
The stored inputs will be an object within the R session.
If they are to be kept for future R sessions, the R workspace must be stored
on disk or on another storage medium (file type RData or rda).
This can be done from the file menu of the R commander (usually the leftmost menu).
The menu entry for loading the stored R workspace in a future session
can be found under “Data Management”.
Within an R session, the latest stored form inputs are normally loaded automatically on next usage of the menu. If you want to restart fresh, the button “Reset form” sets everything to default again.
As pointed out above, if a design is actually generated by pressing the OK button, the menu settings are saved within the design object. The button “Load form” knows how to load the form settings from designs, so that it is not necessary to separately store the form settings in this case.
Ulrike Groemping
See Also oa.design for the function behind this menu,
Menu.General for choice between full factorial designs and orthogonal
main effects designs.