RastaRocket (version 1.1.3)

desc_var: Generate Descriptive Tables for Variables

Description

This function creates descriptive tables for variables in a dataset. It can handle qualitative and quantitative variables, grouped or ungrouped, and supports multiple configurations for handling missing data (DM).

Usage

desc_var(
  data1,
  table_title = "",
  quali = NULL,
  quanti = NULL,
  add_total = TRUE,
  var_title = "Variable",
  by_group = FALSE,
  var_group = NULL,
  group_title = NULL,
  stat_var_quanti = c("{mean} ({sd})", "{median} ({p25}; {p75})", "{min}; {max}"),
  digits = list(r_quanti = 1, r_quali = 1),
  drop_levels = TRUE,
  freq_relevel = FALSE,
  tests = FALSE,
  show_n_per_group = FALSE,
  show_missing_data = NULL,
  var_tot = NULL,
  var_characteristic = NULL
)

Value

A gtsummary table object containing the descriptive statistics.

Arguments

data1

A data frame containing the dataset to be analyzed.

table_title

A character string specifying the title of the table.

quali

A vector of qualitative variables to be described. Defaults to NULL.

quanti

A vector of quantitative variables to be described. Defaults to NULL.

add_total

A boolean (default is TRUE) to add total column or not when var_group is specified.

var_title

A character string for the title of the variable column in the table. Defaults to "Variable".

by_group

A boolean (default is FALSE) to analyse by group.

var_group

A variable used for grouping (if applicable). Defaults to NULL.

group_title

A character string specifying the title for the grouping variable. Default is NULL and get the label or the variable.

stat_var_quanti

A character vector specifying the statistics to display for continuous variables. Default is c("{mean} ({sd})", "{median} ({p25}; {p75})", "{min}; {max}").

digits

A list, the number of decimal places to round categorical and continuous variable. r_quanti and r_quali can be a single integer or a vector of integer. Default is list(r_quanti = 1, r_quali = 1)

drop_levels

Boolean (default = TRUE). Drop unused levels.

freq_relevel

Boolean (default = FALSE). If TRUE, reorder factors by frequency (most to least frequent) using gtsummary.

tests

A value in order to add p value. Default to FALSE OPTION :

• FALSE: No p-value add

• TRUE: Add p-value made by default by gtsummary. See gtsummary add_p() options.

• list(): To force tests. See gtsummary add_p() options.

show_n_per_group

Default to FALSE. Should the 'N' appears in the column header of the groups. Can be either :

• FALSE: No 'N' is shown

• TRUE: 'N' is shown

show_missing_data

Default to NULL. Should the missing data be displayed. Can be either :

• FALSE: No missing data displayed

• TRUE: Missing data displayed

• NULL (default): will be switch to anyNA(data1) value.

var_tot

A string specifying the name of total column. Default is NULL and will guess from theme_gtsummary_language().

var_characteristic

A string specifying the name of characteristic column. Default is NULL and will guess from theme_gtsummary_language().

Details

The function processes the dataset according to the specified parameters and generates descriptive tables.

• It first uses the ajouter_label_ndm() function to append missing data statistics to variable labels.

• Depending on the group and DM arguments, it adjusts the dataset and creates tables using helper functions like desc_group, desc_degroup, and desc_degroup_group.

• Qualitative variables are reordered, and unused levels are dropped when necessary.

Examples

# Example usage:
library(dplyr)

# Sample dataset
data1 <- data.frame(
  group = c("A", "B", "B", "C"),
  var1 = c(1, 2, 3, NA),
  var2 = c("X", "Y", "X", NA)
)

# Generate descriptive table
table <- desc_var(
  data1 = data1,
  table_title = "Descriptive Table",
  quanti = "var1"
)

# Order categorical features by frequency
table1 <- desc_var(
  data1 = data1,
  table_title = "Descriptive Table",
  quanti = "var1",
  freq_relevel = TRUE)

  # Round quantitative and qualitative features using  a vector of integer
  table2 <- desc_var(
    data1 = iris,
    quanti = "Sepal.Length",
    stat_var_quanti = c("{sum}", "{mean} ({sd})"),
    digits = list(r_quanti = c(1, 3, 2), r_quali = c(0, 2))
    )