Multiple Response variables are Categorical Arrays in which one or more categories are set as "selected". These methods allow you to view and set that attribute.
is.dichotomized(x)
dichotomize(x, i)
undichotomize(x)
is.selected(x)
is.selected(x) <- value
# S4 method for class 'Categories'
is.dichotomized(x)
# S4 method for class 'Categories,numeric'
dichotomize(x, i)
# S4 method for class 'Categories,logical'
dichotomize(x, i)
# S4 method for class 'Categories,character'
dichotomize(x, i)
# S4 method for class 'Categories'
undichotomize(x)
# S4 method for class 'CategoricalVariable,ANY'
dichotomize(x, i)
# S4 method for class 'CategoricalArrayVariable,ANY'
dichotomize(x, i)
# S4 method for class 'CategoricalVariable'
undichotomize(x)
# S4 method for class 'CategoricalArrayVariable'
undichotomize(x)
# S4 method for class 'Categories'
is.selected(x)
# S4 method for class 'Categories'
is.selected(x) <- value
# S4 method for class 'Category'
is.selected(x)
# S4 method for class 'Category'
is.selected(x) <- valueCategories or a Variable subclass that has Categories
For the dichotomize methods, the numeric or logical indices
of the categories to mark as "selected", or if character, the Category
"names". Note that unlike some other categorical variable methods,
numeric indices are positional, not with reference to category ids.
For is.selected<-,
A logical vector indicating whether the category should be selected.
For a single category the value should be either TRUE or FALSE. To change the
selection status for a Categories object, supply a logical vector which is the
same length as the number of categories.
Categories or the Variable, (un)dichotomized accordingly
dichotomize lets you specify which categories are "selected", while
undichotomize strips that selection information. Dichotomize converts
a Categorical Array to a Multiple Response, and undichotomize does the reverse.
is.dichotomized reports whether categories have any selected values.
is.selected is lower level and maps more directly onto the "selected"
attributes of categories. The best illustration of this difference is that
is.selected(categories(var)) returns a logical vector, a value for each
category, while is.dichotomized(categories(var)) returns a single
TRUE/FALSE value.
if (FALSE) { # \dontrun{
ds <- newExampleDataset()
is.MR(ds$allpets)
is.dichotomized(categories(ds$allpets))
is.selected(categories(ds$allpets))
ds$allpets <- undichotomize(ds$allpets)
is.CA(ds$allpets)
ds$allpets <- dichotomize(ds$allpets, "selected")
is.MR(ds$allpets)
} # }