interactVariables
takes two or more variables and creates a new one that
is the cartesian product expansion of their unique values. For example, if
we cross ethnicity (with 2 categories) and race (with 4 categories), the new
variable would have 8 valid categories (e.g. black:hispanic, white:hispanic,
black:non-hispanic, etc.) and 7 categories where at least one of the
variables is missing (e.g. white:No Data).
interactVariables(..., name, collapse_missings = FALSE)
a sequence of categorical variables to make an interaction from as well as other properties to pass about the case variable (i.e. alias, description)
a character to use as the name for the interaction variable
a logical indicating whether to combine all
new categories that are formed from existing missing categories into
a single one (defaults to FALSE
).
A VariableDefinition
that creates the new interaction variable.
if (FALSE) {
ds$ethn_race <- interactVariables(
ds$ethnicity, ds$race, name = "Interaction of ethnicity and race"
)
}