data-raw/clean-household.R

In njtierney/conmat: Builds Contact Matrices using GAMs and Population Data

library(tidyverse)
library(here)
library(conmat)
abs_household_lga <- read_csv(
  file = here("data-raw/ABS_C16_T23_LGA_06092021160753604.csv")
) %>%
  rename(
    throw_hhcd_2016 = HHCD_2016,
    household_composition = `Household Composition`,
    throw_nprd_2016 = NPRD_2016,
    n_persons_usually_resident = `Number of Persons Usually Resident`,
    throw_state = STATE,
    state = State,
    throw_region_type = REGIONTYPE,
    throw2_region_type = `Geography Level`,
    throw_lga_code = LGA_2016,
    lga_name = Region,
    year = TIME,
    throw_year = `Census Year`,
    n_households = Value,
    throw_flag = `Flag Codes`,
    throw_flat2 = Flags
  ) %>%
  select(-starts_with("throw")) %>%
  relocate(
    year,
    state,
    lga_name,
    household_composition,
    n_persons_usually_resident,
    n_households
  ) %>%
  mutate(
    n_persons_usually_resident = case_when(
      n_persons_usually_resident == "Total" ~ "total",
      n_persons_usually_resident == "One person" ~ "1",
      n_persons_usually_resident == "Two persons" ~ "2",
      n_persons_usually_resident == "Three persons" ~ "3",
      n_persons_usually_resident == "Four persons" ~ "4",
      n_persons_usually_resident == "Five persons" ~ "5",
      n_persons_usually_resident == "Six persons" ~ "6",
      n_persons_usually_resident == "Seven persons" ~ "7",
      n_persons_usually_resident == "Eight or more persons" ~ "8+"
    )
  ) %>%
  filter(household_composition == "Total Households") %>%
  select(-household_composition) %>%
  mutate(state = abs_abbreviate_states(state)) %>%
  drop_na() %>%
  rename(lga = lga_name) %>%
  # renaming LGAs according to abs_lga_lookup & abs_pop_age_lga_2020
  mutate(
    lga = case_when(
      lga == "Botany Bay (C)" ~ "Bayside (A)",
      lga == "Rockdale (C)" ~ "Bayside (A)",
      lga == "Gundagai (A)" ~ "Cootamundra-Gundagai Regional (A)",
      lga == "Nambucca (A)" ~ "Nambucca Valley (A)",
      lga == "Western Plains Regional (A)" ~ "Dubbo Regional (A)",
      lga == "Mallala (DC)" ~ "Adelaide Plains (DC)",
      lga == "Orroroo/Carrieton (DC)" ~ "Orroroo-Carrieton (DC)",
      lga == "Break O'Day (M)" ~ "Break O`Day (M)",
      lga == "Glamorgan/Spring Bay (M)" ~ "Glamorgan-Spring Bay (M)",
      lga == "Waratah/Wynyard (M)" ~ "Waratah-Wynyard (M)",
      lga == "Kalamunda (S)" ~ "Kalamunda (C)",
      lga == "Kalgoorlie/Boulder (C)" ~ "Kalgoorlie-Boulder (C)",
      TRUE ~ lga
    )
  )

use_data(abs_household_lga, overwrite = TRUE)


abs_2016_lga_pop <- abs_pop_age_lga_2016 %>%
  group_by(year, state, lga) %>%
  summarise(total_popn = sum(population))

abs_2016_lga_pop

# multiply the number of households by the number people in a household
# e.g., 6000 households of size 1 - there are 6000 people who live alone
# do this for each group for one through to 7

household_raw %>%
  filter(n_persons_usually_resident %in% as.character(c(1:7))) %>%
  mutate(
    n_persons_usually_resident = parse_number(n_persons_usually_resident),
    n_persons_in_household = n_persons_usually_resident * n_households
  )

abs_pop_age_lga_2016
unique(household_raw$household_composition)

# question - with these categories in "household_composition"
# we are just interested in "total households", right?
c(
  "Total Households",
  "One family household with only family members present",
  "One family household with non-family members present",
  "Two family household with only family members present",
  "Two family household with non-family members present",
  "Three or more family household with only family members present",
  "Three or more family household with non-family members present",
  "Lone person household",
  "Group household"
)

household_raw %>%
  pull(household_composition) %>%
  unique()

household_raw %>%
  filter(n_persons_usually_resident %in% as.character(c(1:7))) %>%
  mutate(
    n_persons_usually_resident = parse_number(n_persons_usually_resident),
    n_persons_in_household = n_persons_usually_resident * n_households
  ) %>%
  group_by(lga_name) %>%
  summarise(total_in_residence = sum(n_persons_in_household))


household_raw %>%
  filter(
    n_persons_usually_resident %in%
      c(
        "total",
        "8+"
      )
  ) %>%
  left_join(
    abs_2016_lga_pop,
    by = c("year", "state", "lga_name")
  ) %>%
  arrange(lga_name)
# for 8 or more - take Total population, and subtract the sum of the other
# 1-7 groups (from LGA data)