R/household_type.R

In ech: Downloading and Processing Microdata from ECH-INE (Uruguay)

#' This function allows you to calculate the household type for each household in the survey. A household is composed of one or more people who occupy a housing unit.
#' @family demographic
#' @param data data frame with ECH microdata
#' @param numero Variable name of household id
#' @param e26 Variable name of sex
#' @param e27 Variable name of age
#' @param e30 Variable name of householder
#' @importFrom dplyr mutate group_by select
#' @importFrom glue glue
#' @importFrom rlang .data
#' @keywords household_type
#' @export
#' @return data.frame
#' @details
#' Disclaimer: This script is not an official INE product.
#' Aviso: El script no es un producto oficial de INE.
#' @examples
#' \donttest{
#' toy_ech_2018 <- household_type(data = ech::toy_ech_2018)
#' }

household_type <- function(data = ech::toy_ech_2018,
                           numero = "numero",
                           e26 = "e26",
                           e27 = "e27",
                           e30 = "e30") {

  # checks ---
  assertthat::assert_that(is.data.frame(data))
  assertthat::assert_that(e26  %in% names(data), msg =  glue::glue("Sorry... :( \n {e26} is not in data"))
  assertthat::assert_that(e27  %in% names(data), msg =  glue::glue("Sorry... :( \n {e27} is not in data"))
  assertthat::assert_that(e30  %in% names(data), msg =  glue::glue("Sorry... :( \n {e30} is not in data"))

  # if ("household_type" %in% names(data)) {
  #   message(glue::glue("The data.frame already contains a variable with the name household_type, it will be overwritten"))
  # }

  data <- data %>%
    dplyr::mutate(sex_householder = ifelse(.data[[e26]] == 1 & .data[[e30]] == 1,1, # 1 is man and householder
                                           ifelse(.data[[e26]] == 2 & .data[[e30]] == 1, 2, 0)), #0 is woman householder
                  partner = ifelse(.data[[e30]] == 2, 1, 0),
                  child = ifelse(.data[[e30]] %in% 3:5, 1, 0),
                  child_law = ifelse(.data[[e30]] == 6, 1, 0),
                  under_18 = ifelse(.data[[e27]] < 18, 1, 0),
                  parents_brosis = ifelse(.data[[e30]] %in% 7:10, 1, 0),
                  grandchild = ifelse(.data[[e30]] == 11, 1, 0),
                  other_rel = ifelse(.data[[e30]] == 12, 1, 0),
                  no_rel = ifelse(.data[[e30]] == 13, 1, 0)) %>%
    dplyr::group_by(.data$numero, .add = T) %>%
    dplyr::mutate(sex_householder = sum(sex_householder),
                  under_18 = sum(.data$under_18),
                  partner = sum(.data$partner),
                  child = sum(.data$child),
                  child_law = sum(.data$child_law),
                  parents_brosis = sum(.data$parents_brosis),
                  grandchild = sum(.data$grandchild),
                  other_rel = sum(.data$other_rel),
                  no_rel = sum(.data$no_rel),
                  household_type = ifelse(.data$partner ==0 & .data$child == 0 & .data$parents_brosis == 0 & .data$grandchild == 0 & .data$child_law == 0 & .data$other_rel == 0 & .data$no_rel == 0, 1, #Single person
                                          ifelse(.data$partner > 0 & .data$child == 0 & .data$parents_brosis == 0 & .data$grandchild == 0 & .data$child_law == 0 & .data$other_rel == 0 & .data$no_rel== 0, 2,#Couple without children
                                                 ifelse(.data$partner == 0 & .data$child > 0  & sex_householder == 1 & .data$parents_brosis == 0 & .data$grandchild == 0 & .data$child_law == 0 & .data$other_rel == 0 & .data$no_rel == 0, 3, #Single parent or Single father
                                                        ifelse(.data$partner == 0 & .data$child > 0 & sex_householder == 2 & .data$parents_brosis == 0 & .data$grandchild == 0 & .data$child_law == 0 & .data$other_rel == 0 & .data$no_rel == 0, 4, #Single parent or Single mother
                                                               ifelse(.data$partner > 0 & .data$child > 0 & .data$parents_brosis ==0 &  .data$grandchild == 0 & .data$child_law == 0 &  .data$other_rel == 0 & .data$no_rel == 0, 5, #Couple with children
                                                                      ifelse((.data$parents_brosis > 0 | .data$grandchild > 0 | .data$child_law > 0 | .data$other_rel > 0) & .data$no_rel == 0, 6, #Extended
                                                                             ifelse(.data$no_rel > 0, 7, NA))))))), # composite) %>%
                  household_type = haven::labelled(household_type, labels = c("Unipersonal" = 1, "Pareja" = 2, "Monoparental" = 3, "Monomarental" = 4, "Biparental" = 5, "Extendido" = 6, "Compuesto" = 7),
                                                   label = "Tipo de hogar")
    ) %>%
    ungroup()

  data <- data %>% dplyr::select(everything(), -sex_householder:-no_rel)
  message("A variable has been created: \n \t household_type (Tipo de hogar)")
  return(data)

}