Assign to groups based on GroupConditions

Using GroupConditions object, assign observations to one of the groups. It can export either indices of the observations, or their unique ID: if column name is provided in id argument. Mostly used internally by more complex functions and R6 classes, but could also be useful on its own.

Usage

GroupAssignment(
  data,
  conditions,
  id,
  force_disjoint,
  force_exhaustive,
  skip_faulty = FALSE,
  .all = FALSE,
  ...
)

# S3 method for GroupAssignment
print(x, ...)

# S3 method for GroupAssignment
summary(object, ...)

Arguments

data

data.frame containing observations

conditions

GroupConditions object

id

character name of the column containing unique ID of the observations to assign to each group. If not provided, indices will be used instead.

force_disjoint

boolean indicating if groups disjointedness should be forced in case when one observation would pass conditions for more than one group. If TRUE, the first condition which will be met will indicate the group the observation will be assigned to. If not provided, the default from conditions will be used

force_exhaustive

boolean indicating if groups exhausiveness should be forced in case when there are observations that don't pass any of the provided conditions. If TRUE, then they will be assigned to .NA group. If not provided, the default from conditions will be used

skip_faulty

boolean should the faulty condition be skipped? If FALSE as in default, error will be produced. Faultiness of seemingly correct condition may be caused by variable names to not be present in the data.

.all

boolean. If TRUE, then additional group named .all will be created, which will contain all observations. Useful when object will be used for creation of GroupedFrequencyTable()

...

additional arguments to be passed to or from method

x

object

object

GroupAssignment object

Value

GroupAssignment object

list of summaries, invisibly

See also

Other observation grouping functions: extract_observations(), intersect_GroupAssignment()

Examples

age_grouping <- GroupConditions(
  conditions_category = "Age",
  "to 20" ~ age < 20,
  "20 to 40" ~ age >= 20 & age <= 40,
  "40 to 60" ~ age >= 40 & age < 60
)

# on basis of GroupConditions create GroupAssignment

age_assignment <- GroupAssignment(
  data = HEXACO_60,
  age_grouping)
#> Warning: ! Some observations were not assigned on provided condition. Set the
#>   `force_exhaustive = TRUE` to gather them in `.NA` group.

print(age_assignment)
#> <GroupAssignment>
#> Total assigned: 202
#> Mode: "index"
#> Groups:
#> "to 20", "20 to 40", and "40 to 60"

# overwrite the default settings imposed by `GroupConditions`

age_assignment_forced <- GroupAssignment(
  data = HEXACO_60,
  age_grouping,
  force_exhaustive = TRUE)

summary(age_assignment_forced)
#> <GroupAssignment>
#> Status
#> • Mode: index
#> • Total assigned: 204
#> • Disjointedness: TRUE; Forced: TRUE
#> • Exhaustiveness: TRUE; Forced: TRUE
#> Assignment [tested vars: `age`]
#> • Group: to 20 [obs: 20]
#> • Group: 20 to 40 [obs: 173]
#> • Group: 40 to 60 [obs: 9]
#> • Group: .NA [obs: 2]

# you can also use other unique identifier from your data

age_assignment_forced_w_id <- GroupAssignment(
  data = HEXACO_60,
  age_grouping,
  id = "user_id",
  force_exhaustive = TRUE)

summary(age_assignment_forced_w_id)
#> <GroupAssignment>
#> Status
#> • Mode: id [default ID: `user_id`]
#> • Total assigned: 204
#> • Disjointedness: TRUE; Forced: TRUE
#> • Exhaustiveness: TRUE; Forced: TRUE
#> Assignment [tested vars: `age`]
#> • Group: to 20 [obs: 20]
#> • Group: 20 to 40 [obs: 173]
#> • Group: 40 to 60 [obs: 9]
#> • Group: .NA [obs: 2]