Added Wales IMD scores

R/data.R

@@ -376,7 +376,7 @@ NULL
 #'   ...
 #' }
 #' @source \url{https://gov.wales/}
-"imd_wales_lad"
+"imd2019_wales_ltla22"
 
 #' Index of Multiple Deprivation for Local Government Districts in Northern
 #' Ireland

data-raw/imd2019_wales_ltla22.R

@@ -5,17 +5,17 @@ library(sf)
 # Load package
 devtools::load_all(".")
 
-# ---- Check LTLA codes ----
-# Confirming that LTLA 2021 and LTLA 2022 codes are the same for Wales
-dplyr::symdiff(
-  geographr::lookup_ltla_ltla |>
-    filter(str_detect(ltla21_code, "W")) |>
-    pull(ltla21_code),
-
-  geographr::lookup_ltla22_ltla23 |>
-    filter(str_detect(ltla22_code, "W")) |>
-    pull(ltla22_code)
-)
+# ---- LSOA to LAD lookup ----
+# Double-check that Welsh LTLAs didn't change codes between 2021 and 2022
+geographr::lookup_ltla_ltla |>
+  distinct(ltla21_code, ltla22_code) |>
+  filter(str_detect(ltla21_code, "^W")) |>
+  filter(ltla21_code != ltla22_code)
+#--> They didn't, so we can refer to the 2021 codes as 2022
+
+lsoa_lad <-
+  geographr::lookup_lsoa11_ltla21 |>
+  select(lsoa11_code, ltla22_code = ltla21_code)
 
 # ---- Population estimates in LSOAs ----
 query_url <-
@@ -50,76 +50,55 @@ lsoa_pop <-
 # Select and rename vars
 lsoa_pop <-
   lsoa_pop |>
-  select(
+  dplyr::select(
     lsoa11_code = `LSOA Code`,
     population = `All Ages`
   ) |>
-  distinct()
+  dplyr::distinct()
 
 lsoa_pop_wales <-
   lsoa_pop |>
-  filter(substr(lsoa11_code, 1, 1) == "W")
+  dplyr::filter(substr(lsoa11_code, 1, 1) == "W")
 
 # ---- Aggregate IMD into MSOAs ----
 wimd_lsoa <-
   imd_wales_lsoa |>
-
   rename(lsoa11_code = lsoa_code) |>
-
-  # We don't have IMD scores for Wales so just set them as zero
-  mutate(
-    IMD_score = 0,
-    Income_score = 0,
-    Employment_score = 0,
-    Education_score = 0,
-    Health_score = 0,
-    Crime_score = 0,
-    Housing_score = 0,
-    Access_score = 0,
-    Environment_score = 0
-  ) |>
-
-  left_join(lsoa_pop) |>
-  left_join(
-    geographr::lookup_lsoa11_ltla21 |>
-      select(lsoa11_code, ltla22_code = ltla21_code)
-  )
+  dplyr::left_join(lsoa_pop, by = "lsoa11_code") |>
+  dplyr::left_join(lsoa_lad, by = "lsoa11_code")
 
 # Aggregate into LADs
-wimd_ltla22 <-
+wimd_lad <-
   wimd_lsoa |> aggregate_scores(IMD_score, IMD_rank, IMD_decile, ltla22_code, population)
 
-wimd_ltla22_income   <- wimd_lsoa |> aggregate_scores(Income_score, Income_rank, Income_decile, ltla22_code, population)
-wimd_ltla22_employ   <- wimd_lsoa |> aggregate_scores(Employment_score, Employment_rank, Employment_decile, ltla22_code, population)
-wimd_ltla22_edu      <- wimd_lsoa |> aggregate_scores(Education_score, Education_rank, Education_decile, ltla22_code, population)
-wimd_ltla22_health   <- wimd_lsoa |> aggregate_scores(Health_score, Health_rank, Health_decile, ltla22_code, population)
-wimd_ltla22_crime    <- wimd_lsoa |> aggregate_scores(Crime_score, Crime_rank, Crime_decile, ltla22_code, population)
-wimd_ltla22_housing  <- wimd_lsoa |> aggregate_scores(Housing_score, Housing_rank, Housing_decile, ltla22_code, population)
-wimd_ltla22_barriers <- wimd_lsoa |> aggregate_scores(Access_score, Access_rank, Access_decile, ltla22_code, population)
-wimd_ltla22_env      <- wimd_lsoa |> aggregate_scores(Environment_score, Environment_rank, Environment_decile, ltla22_code, population)
-
-wimd_ltla22_income   <- wimd_ltla22_income   |> dplyr::rename(Income_Proportion = Proportion, Income_Extent = Extent, Income_Score = Score)
-wimd_ltla22_employ   <- wimd_ltla22_employ   |> dplyr::rename(Employment_Proportion = Proportion, Employment_Extent = Extent, Employment_Score = Score)
-wimd_ltla22_edu      <- wimd_ltla22_edu      |> dplyr::rename(Education_Proportion = Proportion, Education_Extent = Extent, Education_Score = Score)
-wimd_ltla22_health   <- wimd_ltla22_health   |> dplyr::rename(Health_Proportion = Proportion, Health_Extent = Extent, Health_Score = Score)
-wimd_ltla22_crime    <- wimd_ltla22_crime    |> dplyr::rename(Crime_Proportion = Proportion, Crime_Extent = Extent, Crime_Score = Score)
-wimd_ltla22_housing  <- wimd_ltla22_housing  |> dplyr::rename(Housing_Proportion = Proportion, Housing_Extent = Extent, Housing_Score = Score)
-wimd_ltla22_barriers <- wimd_ltla22_barriers |> dplyr::rename(Access_Proportion = Proportion, Access_Extent = Extent, Access_Score = Score)
-wimd_ltla22_env      <- wimd_ltla22_env      |> dplyr::rename(Environment_Proportion = Proportion, Environment_Extent = Extent, Environment_Score = Score)
+wimd_lad_income   <- wimd_lsoa |> aggregate_scores(Income_score, Income_rank, Income_decile, ltla22_code, population)
+wimd_lad_employ   <- wimd_lsoa |> aggregate_scores(Employment_score, Employment_rank, Employment_decile, ltla22_code, population)
+wimd_lad_edu      <- wimd_lsoa |> aggregate_scores(Education_score, Education_rank, Education_decile, ltla22_code, population)
+wimd_lad_health   <- wimd_lsoa |> aggregate_scores(Health_score, Health_rank, Health_decile, ltla22_code, population)
+wimd_lad_crime    <- wimd_lsoa |> aggregate_scores(Crime_score, Crime_rank, Crime_decile, ltla22_code, population)
+wimd_lad_housing  <- wimd_lsoa |> aggregate_scores(Housing_score, Housing_rank, Housing_decile, ltla22_code, population)
+wimd_lad_barriers <- wimd_lsoa |> aggregate_scores(Access_score, Access_rank, Access_decile, ltla22_code, population)
+wimd_lad_env      <- wimd_lsoa |> aggregate_scores(Environment_score, Environment_rank, Environment_decile, ltla22_code, population)
+
+wimd_lad_income   <- wimd_lad_income   |> dplyr::rename(Income_Proportion = Proportion, Income_Extent = Extent, Income_Score = Score)
+wimd_lad_employ   <- wimd_lad_employ   |> dplyr::rename(Employment_Proportion = Proportion, Employment_Extent = Extent, Employment_Score = Score)
+wimd_lad_edu      <- wimd_lad_edu      |> dplyr::rename(Education_Proportion = Proportion, Education_Extent = Extent, Education_Score = Score)
+wimd_lad_health   <- wimd_lad_health   |> dplyr::rename(Health_Proportion = Proportion, Health_Extent = Extent, Health_Score = Score)
+wimd_lad_crime    <- wimd_lad_crime    |> dplyr::rename(Crime_Proportion = Proportion, Crime_Extent = Extent, Crime_Score = Score)
+wimd_lad_housing  <- wimd_lad_housing  |> dplyr::rename(Housing_Proportion = Proportion, Housing_Extent = Extent, Housing_Score = Score)
+wimd_lad_barriers <- wimd_lad_barriers |> dplyr::rename(Access_Proportion = Proportion, Access_Extent = Extent, Access_Score = Score)
+wimd_lad_env      <- wimd_lad_env      |> dplyr::rename(Environment_Proportion = Proportion, Environment_Extent = Extent, Environment_Score = Score)
 
 imd2019_wales_ltla22 <-
-  wimd_ltla22 |>
-  left_join(wimd_ltla22_income,   by = "ltla22_code") |>
-  left_join(wimd_ltla22_employ,   by = "ltla22_code") |>
-  left_join(wimd_ltla22_edu,      by = "ltla22_code") |>
-  left_join(wimd_ltla22_health,   by = "ltla22_code") |>
-  left_join(wimd_ltla22_crime,    by = "ltla22_code") |>
-  left_join(wimd_ltla22_housing,  by = "ltla22_code") |>
-  left_join(wimd_ltla22_barriers, by = "ltla22_code") |>
-  left_join(wimd_ltla22_env,      by = "ltla22_code") |>
-
-  # Don't have scores for Wales so drop these columns
-  select(-ends_with("Score"))
+  wimd_lad |>
+  dplyr::left_join(wimd_lad_income,   by = "ltla22_code") |>
+  dplyr::left_join(wimd_lad_employ,   by = "ltla22_code") |>
+  dplyr::left_join(wimd_lad_edu,      by = "ltla22_code") |>
+  dplyr::left_join(wimd_lad_health,   by = "ltla22_code") |>
+  dplyr::left_join(wimd_lad_crime,    by = "ltla22_code") |>
+  dplyr::left_join(wimd_lad_housing,  by = "ltla22_code") |>
+  dplyr::left_join(wimd_lad_barriers, by = "ltla22_code") |>
+  dplyr::left_join(wimd_lad_env,      by = "ltla22_code")
 
 # Save output to data/ folder
 usethis::use_data(imd2019_wales_ltla22, overwrite = TRUE)