In emraher/tuik: Download Data File And Database Urls from TUIK

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%",
  warning = FALSE,
  error = FALSE,
  message = FALSE,
  cache = TRUE
)

tuik

The goal of tuik is to extract data file and database URLs from TUIK webpage. Package can also download data from Geographical Statistics Portal.

Installation

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("emraher/tuik")

Example

library(tidyverse)
library(tuik)

(st <- statistical_themes())

# stab <- statistical_tables("aaa")
#> Error in check_theme_id(theme) :
#>  You should select a valid theme ID!

# stab <- statistical_tables(c(123, 143))
#> Error in check_theme_id(theme) : You can select only one theme!

(stab <- statistical_tables("110"))

# sdb <- statistical_databases("aaa")
#> Error in check_theme_id(theme) :
#>  You should select a valid theme ID!

# sdb <- statistical_databases(c(123, 143))
#> Error in check_theme_id(theme) : You can select only one theme!

(sdb <- statistical_databases(110))

If you are having problems at this stage, please see this issue on GitHub.

# -------------------------------------------------------------------------- ###
# Saving Data Files----
# -------------------------------------------------------------------------- ###
# Read xls files into R
# NOTE: TUIK xls files are messy!!!
download.file(stab$datafile_url[1], destfile = "/tmp/file.xls")
(dt <- tibble::as_tibble(readxl::read_xls("/tmp/file.xls")))



# Download file from URL
filename <- paste0(
  janitor::make_clean_names(stab$data_name[1]),
  janitor::make_clean_names(stab$data_date[1])
)

download.file(stab$datafile_url[1],
  destfile = paste0("/tmp/", filename, ".xls"),
  mode = "wb"
)


# -------------------------------------------------------------------------- ###
# All DB Links---- NOT WORKING AT THIS TIME. TUIK CHANGED THE PAGE
# -------------------------------------------------------------------------- ###
# all_dbs <- purrr::map_df(.x = st$theme_id, .f = ~statistical_databases(.x))
#
# all_dbs %>%
#   dplyr::count(theme_name, name = "database_count")

# -------------------------------------------------------------------------- ###
# Download Geo Data----
# -------------------------------------------------------------------------- ###
# Download Variable Names and Codes
(dt <- geo_data())



# Download data for a given level and variable
dt |>
  filter(var_num == "SNM-GK160951-O33303")

(dt_x <- geo_data(
  variable_level = 3,
  variable_no = "SNM-GK160951-O33303",
  variable_source = "medas",
  variable_period = "yillik",
  variable_recnum = 5
))

# (dt <- geo_data(4, "TFE-GK105747-O23001"))
#> Error in value[[3L]](cond) :
#>  This data (TFE-GK105747-O23001) is not available at this NUTS level (level = 4)!!!

# -------------------------------------------------------------------------- ###
# Download Geo Map----
# -------------------------------------------------------------------------- ###
(dt_sf <- geo_map(9))

(dt_sf <- geo_map(3))

Map Examples

NUTS-2

dt |>
  filter(var_num == "HYV-GK1696800-O32507")

chicken <- geo_data(
  variable_level = 2,
  variable_no = "HYV-GK1696800-O32507",
  variable_source = "medas",
  variable_period = "yillik",
  variable_recnum = 20
) %>%
  dplyr::filter(date %in% c("2023", "2015")) |>
  mutate(
    yumurta_tavugu_sayisi_adet = as.numeric(yumurta_tavugu_sayisi_adet),
    date = as.numeric(date)
  )

geo_map(2) %>%
  left_join(chicken) %>%
  ggplot() +
  geom_sf(aes(fill = yumurta_tavugu_sayisi_adet), color = "white") +
  coord_sf(datum = NA) +
  rcartocolor::scale_fill_carto_c(palette = "Safe") +
  hrbrthemes::theme_ipsum_rc() +
  theme(legend.position = "bottom", legend.key.width = unit(3, "cm")) +
  labs(
    fill = "",
    title = "Yumurta Tavuğu (Adet)",
    caption = "Kaynak: TÜİK"
  ) +
  facet_wrap(~date, ncol = 2)

NUTS-3

dt |>
  filter(var_num == "INS-GK055-O006")

house <- geo_data(
  variable_level = 3,
  variable_no = "INS-GK055-O006",
  variable_source = "ilGostergeleri",
  variable_period = "yillik",
  variable_recnum = 5
) %>%
  filter(date == 2019)

# Let's select different colors
pal <- wesanderson::wes_palette("BottleRocket2", 50, type = "continuous")

geo_map(3) %>%
  left_join(house) %>%
  mutate(konut_satis_sayilari_toplam = as.numeric(konut_satis_sayilari_toplam)) %>%
  ggplot() +
  geom_sf(aes(fill = konut_satis_sayilari_toplam)) +
  coord_sf(datum = NA) +
  scale_fill_gradientn(colours = pal) +
  hrbrthemes::theme_ipsum_rc() +
  theme(
    legend.position = "bottom",
    legend.key.width = unit(3, "cm"),
    plot.title = element_text(hjust = 0.5)
  ) +
  labs(
    fill = "",
    title = "2019 Yılında Konut Satış Sayıları",
    caption = "Kaynak: TÜİK"
  )

LAU-1

dt |>
  filter(var_num == "ULE-GK160887-O29502")

pal <- c("#f7f7f7", "#d9d9d9", "#bdbdbd", "#969696", "#737373", "#525252", "#252525")

geo_data(
  variable_level = 4,
  variable_no = "ULE-GK160887-O29502",
  variable_source = "medas",
  variable_period = "yillik",
  variable_recnum = 5
) %>%
  filter(date == 2022) %>%
  left_join(geo_map(level = 4), .) %>%
  mutate(okuma_yazma_bilmeyen_sayisi = as.numeric(okuma_yazma_bilmeyen_sayisi)) |>
  ggplot() +
  geom_sf(aes(fill = okuma_yazma_bilmeyen_sayisi), lwd = 0.1) +
  coord_sf(datum = NA) +
  scale_fill_gradientn(colours = pal) +
  theme_bw() +
  theme(
    legend.position = "bottom",
    legend.key.width = unit(3, "cm"),
    plot.title = element_text(hjust = 0.5)
  ) +
  labs(
    fill = "",
    title = "2019 Yılında Okuma Yazma Bilmeyen Sayısı",
    caption = "Kaynak: TÜİK"
  )

geo_data(
  variable_level = 4,
  variable_no = "ADNKS-GK137473-O29001",
  variable_source = "medas",
  variable_period = "yillik",
  variable_recnum = 5
) |>
  filter(date == 2019) %>%
  left_join(geo_map(level = 4), .) %>%
  mutate(toplam_nufus = as.numeric(toplam_nufus)) |>
  filter(bolgeKodu == "TR51") %>%
  ggplot() +
  geom_sf(aes(fill = toplam_nufus)) +
  scale_fill_viridis_c(option = "E") +
  labs(fill = "Population in 2019") +
  hrbrthemes::theme_ipsum_rc()

Hex and Dorling

library(tidyverse)
library(sf)
library(cartogram)
library(geogrid)
library(tuik)

# -------------------------------------------------------------------------- ###
# Hex----
# -------------------------------------------------------------------------- ###
# Read map data and transform CRS
tur_hex_map <- st_transform(geo_map(level = 3), crs = 3395)

dt |>
  filter(var_num == "ULS-GK093-O009")


# Read data and merge
tur_hex_dt <- geo_data(
  variable_level = 3,
  variable_no = "ULS-GK093-O009",
  variable_source = "ilGostergeleri",
  variable_period = "yillik",
  variable_recnum = 5
) %>%
  filter(date == 2019) %>%
  left_join(tur_hex_map, .)

# Choose one seed according to these plots
par(mfrow = c(4, 4), mar = c(0, 0, 2, 0))
for (i in 1:16) {
  new_cells <- calculate_grid(shape = tur_hex_dt, grid_type = "hexagonal", seed = i)
  plot(new_cells, main = paste("Seed", i, sep = " "))
}

# Create hex map
new_hex <- calculate_grid(shape = tur_hex_dt, grid_type = "hexagonal", seed = 9)
result_hex <- assign_polygons(tur_hex_dt, new_hex)

# Plot
result_hex %>%
  mutate(name = stringr::str_replace(name, "AFYONKARAHİSAR", "AFYON")) %>%
  mutate(name = stringr::str_replace(name, "KAHRAMANMARAŞ", "K.MARAŞ")) %>%
  mutate(bin_kisi_basina_otomobil_sayisi = as.numeric(bin_kisi_basina_otomobil_sayisi)) |>
  ggplot() +
  geom_sf(aes(fill = bin_kisi_basina_otomobil_sayisi), lwd = 0.1) +
  geom_sf_text(aes(label = name), color = "black", size = 2) +
  coord_sf(datum = NA) +
  rcartocolor::scale_fill_carto_c(palette = "OrYel") +
  hrbrthemes::theme_ipsum_rc() +
  theme(
    legend.position = "bottom",
    legend.key.width = unit(3, "cm"),
    plot.title = element_text(hjust = 0.5)
  ) +
  labs(
    fill = "",
    y = "",
    x = "",
    title = "2019 Yılında Bin Kişi Başına Otomobil Sayısı",
    caption = "Kaynak: TÜİK"
  )

# -------------------------------------------------------------------------- ###
# Dorling----
# -------------------------------------------------------------------------- ###
# Read map data and transform CRS
tur_map <- st_transform(geo_map(level = 4), crs = 3395)

# Read data and merge
tur_pop <- geo_data(
  variable_level = 4,
  variable_no = "ADNKS-GK137473-O29001",
  variable_source = "medas",
  variable_period = "yillik",
  variable_recnum = 5
)
tur_ill <- geo_data(
  variable_level = 4,
  variable_no = "ULE-GK160887-O29502",
  variable_source = "medas",
  variable_period = "yillik",
  variable_recnum = 5
)

tur_dt <- left_join(tur_pop, tur_ill) %>%
  filter(date == 2019) %>%
  left_join(tur_map, .) %>%
  mutate(
    okuma_yazma_bilmeyen_sayisi = as.numeric(okuma_yazma_bilmeyen_sayisi),
    toplam_nufus = as.numeric(toplam_nufus)
  ) |>
  mutate(value = 100 * (okuma_yazma_bilmeyen_sayisi / toplam_nufus))

# Create dorling maps
tur_dorling1 <- cartogram_dorling(tur_dt, "okuma_yazma_bilmeyen_sayisi", 0.4)

tur_dorling1 %>%
  ggplot() +
  geom_sf(aes(fill = okuma_yazma_bilmeyen_sayisi), lwd = 0.1) +
  coord_sf(datum = NA) +
  scale_fill_viridis_c(option = "B") +
  theme_bw() +
  theme(
    legend.position = "bottom",
    legend.key.width = unit(3, "cm"),
    plot.title = element_text(hjust = 0.5)
  ) +
  labs(
    fill = "",
    title = "2019 Yılında Okuma Yazma Bilmeyen Sayısı",
    caption = "Kaynak: TÜİK"
  )

tur_dorling2 <- cartogram_dorling(tur_dt, "toplam_nufus", 0.4)

tur_dorling2 %>%
  ggplot() +
  geom_sf(aes(fill = toplam_nufus), lwd = 0.1) +
  coord_sf(datum = NA) +
  scale_fill_viridis_c(option = "B") +
  theme_bw() +
  theme(
    legend.position = "bottom",
    legend.key.width = unit(3, "cm"),
    plot.title = element_text(hjust = 0.5)
  ) +
  labs(
    fill = "",
    title = "2019 Yılında Nüfus",
    caption = "Kaynak: TÜİK"
  )

tur_dorling3 <- cartogram_dorling(tur_dt, "value", 0.4)

tur_dorling3 %>%
  ggplot() +
  geom_sf(aes(fill = value), lwd = 0.1) +
  coord_sf(datum = NA) +
  scale_fill_viridis_c(option = "B") +
  theme_bw() +
  theme(
    legend.position = "bottom",
    legend.key.width = unit(3, "cm"),
    plot.title = element_text(hjust = 0.5)
  ) +
  labs(
    fill = "",
    title = "2019 Yılında Okuma Yazma Bilmeyen Sayısının Nüfusa Oranı (%)",
    caption = "Kaynak: TÜİK"
  )