summarytabl (version 0.2.1)

select_group_tbl: Summarize multiple response variables by group or pattern

Description

select_group_tbl() displays frequency counts and percentages for multiple response variables (e.g., a series of questions where participants answer "Yes" or "No" to each item) as well as ordinal variables (such as Likert or Likert-type items with responses ranging from "Strongly Disagree" to "Strongly Agree", where respondents select one response per statement, question, or item), grouped either by another variable in your dataset or by a matched pattern in the variable names.

Usage

select_group_tbl(
  data,
  var_stem,
  group,
  var_input = "stem",
  regex_stem = FALSE,
  ignore_stem_case = FALSE,
  group_type = "variable",
  group_name = NULL,
  margins = "all",
  regex_group = FALSE,
  ignore_group_case = FALSE,
  remove_group_non_alnum = TRUE,
  na_removal = "listwise",
  pivot = "longer",
  only = NULL,
  var_labels = NULL,
  ignore = NULL,
  force_pivot = FALSE
)

Value

A tibble displaying the count and percentage for each category in a multi-response variable, grouped either by a specified variable in the dataset or by matching patterns in variable names.

Arguments

data

A data frame.

var_stem

A character vector with one or more elements, where each represents either a variable stem or the complete name of a variable present in data. A variable 'stem' refers to a common naming pattern shared among related variables, typically reflecting repeated measures of the same idea or a group of items assessing a single concept.

group

A character string representing a variable name or a pattern used to search for variables in data.

var_input

A character string specifying whether the values supplied to var_stem should be treated as variable stems (stem) or as complete variable names (name). By default, this is set to stem, so the function searches for variables that begin with each stem provided. Setting this argument to name directs the function to look for variables that exactly match the provided names.

regex_stem

A logical value indicating whether to use Perl-compatible regular expressions when searching for variable stems. Default is FALSE.

ignore_stem_case

A logical value indicating whether the search for columns matching the supplied var_stem is case-insensitive. Default is FALSE.

group_type

A character string that defines how the group argument should be interpreted. Should be one of pattern or variable. Defaults to variable, which searches for a matching variable name in data.

group_name

An optional character string used to rename the group column in the final table When group_type is set to variable, the column name defaults to the matched variable name from data. When set to pattern, the default column name is group.

margins

A character string that determines how percentage values are calculated; whether they sum to one across rows, columns, or the entire variable (i.e., all). Defaults to all, but can also be set to rows or columns. Note: This argument only affects the final table when group_type is variable.

regex_group

A logical value indicating whether to use Perl-compatible regular expressions when searching for group variables or matching variable name patterns. Default is FALSE.

ignore_group_case

A logical value specifying whether the search for a grouping variable (if group_type is variable) or for variables matching a pattern (if group_type is pattern) should be case-insensitive. Default is FALSE. Set to TRUE to ignore case.

remove_group_non_alnum

A logical value indicating whether to remove all non-alphanumeric characters (i.e., anything that is not a letter or number) from group. Default is TRUE.

na_removal

A character string that specifies the method for handling missing values: pairwise or listwise. Defaults to listwise.

pivot

A character string that determines the format of the table. By default, longer returns the data in the long format. To return the data in the wide format, specify wider.

only

A character string or vector of character strings of the types of summary data to return. Default is NULL, which returns both counts and percentages. To return only counts or percentages, use count or percent, respectively.

var_labels

An optional named character vector or list used to assign custom labels to variable names. Each element must be named and correspond to a variable included in the returned table. If var_input is set to stem, and any element is either unnamed or refers to a variable not present in the table, all labels will be ignored and the table will be printed without them.

ignore

An optional named vector or list indicating values to exclude from variables matching specified stems (or names), and, if applicable, from a grouping variable in data. Defaults to NULL, indicating that all values are retained. To specify exclusions for variables identified by var_stem, use the corresponding stems or variable names as names in the vector or list. To exclude multiple values from these variables or a grouping variable, supply them as a named list.

force_pivot

A logical value that enables pivoting to the 'wider' format even when variables have inconsistent value sets. By default, this is set to FALSE to prevent reshaping errors when values differ across variables in the returned table. Set to TRUE to override this safeguard and pivot to the 'wider' format regardless of value inconsistencies.

Author

Ama Nyame-Mensah

Examples

select_group_tbl(data = stem_social_psych,
                 var_stem = "belong_belong",
                 group = "\\d",
                 group_type = "pattern",
                 group_name = "wave",
                 na_removal = "pairwise",
                 pivot = "wider",
                 only = "count")

tas_recoded <-
  tas |>
  dplyr::mutate(sex = dplyr::case_when(
    sex == 1 ~ "female",
    sex == 2 ~ "male",
    TRUE ~ NA)) |>
  dplyr::mutate(dplyr::across(
    .cols = dplyr::starts_with("involved_"),
    .fns = ~ dplyr::case_when(
      .x == 1 ~ "selected",
      .x == 0 ~ "unselected",
      TRUE ~ NA)
  ))

select_group_tbl(data = tas_recoded,
                 var_stem = "involved_",
                 group = "sex",
                 group_type = "variable",
                 na_removal = "pairwise",
                 pivot = "wider")

depressive_recoded <-
  depressive |>
  dplyr::mutate(sex = dplyr::case_when(
    sex == 1 ~ "male",
    sex == 2 ~ "female",
    TRUE ~ NA)) |>
  dplyr::mutate(dplyr::across(
    .cols = dplyr::starts_with("dep_"),
    .fns = ~ dplyr::case_when(
      .x == 1 ~ "often",
      .x == 2 ~ "sometimes",
      .x == 3 ~ "hardly",
      TRUE ~ NA
    )
  ))

select_group_tbl(data = depressive_recoded,
                 var_stem = "dep",
                 group = "sex",
                 group_type = "variable",
                 na_removal = "listwise",
                 pivot = "wider",
                 only = "percent",
                 var_labels =
                   c("dep_1" = "how often child feels sad and blue",
                     "dep_2" = "how often child feels nervous, tense, or on edge",
                     "dep_3" = "how often child feels happy",
                     "dep_4" = "how often child feels bored",
                     "dep_5" = "how often child feels lonely",
                     "dep_6" = "how often child feels tired or worn out",
                     "dep_7" = "how often child feels excited about something",
                     "dep_8" = "how often child feels too busy to get everything"))