update rendering

.github/workflows/ggl.yml

@@ -54,6 +54,9 @@ jobs:
             any::vroom
             any::prettydoc
             any::DT
+      - name: Apt-get update things
+        run: |
+          sudo apt-get update 
       - name: Apt-get needed things
         run: |
           sudo apt-get install --no-install-recommends -y \

.github/workflows/r30.yml

@@ -20,6 +20,9 @@ jobs:
           quarto --version
       - name: Setup - Checkout repo
         uses: actions/checkout@v3
+      - name: Apt-get update things
+        run: |
+          sudo apt-get update 
       - name: Apt-get needed things
         run: |
           sudo apt-get install --no-install-recommends -y \

.github/workflows/r7.yml

@@ -20,6 +20,9 @@ jobs:
           quarto --version
       - name: Setup - Checkout repo
         uses: actions/checkout@v3
+      - name: Apt-get update things
+        run: |
+          sudo apt-get update 
       - name: Apt-get needed things
         run: |
           sudo apt-get install --no-install-recommends -y \

.github/workflows/render.yml

@@ -22,6 +22,9 @@ jobs:
           quarto --version
       - name: Setup - Checkout repo
         uses: actions/checkout@v3
+      - name: Apt-get update things
+        run: |
+          sudo apt-get update 
       - name: Apt-get needed things
         run: |
           sudo apt-get install --no-install-recommends -y \

data/comparison.csv

@@ -109,6 +109,9 @@ Date,cumulative_spend,days_until,election
 2023-11-17,934570,5,2023
 2023-11-18,1000850,4,2023
 2023-11-19,1106904,3,2023
+2023-11-20,1233814,2,2023
+2023-11-21,1366797,1,2023
+2023-11-22,1511741,0,2023
 2020-11-23,2897,112,2021
 2020-11-24,5786,111,2021
 2020-11-25,9175,110,2021

elex.R

@@ -0,0 +1,247 @@
+
+
+tk2021 <- read_csv2("data/TK20210317.csv") %>% 
+  janitor::clean_names() %>% 
+  mutate(pvv = pvv_partij_voor_de_vrijheid/opkomst)
+
+# tk2021$
+tk2023 <- read_csv("data/data-2SSi6.csv") %>% 
+  janitor::clean_names()
+
+
+
+tk2023 %>% 
+  mutate(regio_code = str_remove_all(lau_id, "M")) %>% 
+  select(regio_code, pvv2023=pvv) %>% 
+  left_join(tk2021 %>% select(regio_code, pvv2021 =pvv )) %>% 
+  ggplot(aes(pvv2021, pvv2023)) +
+  geom_point() +
+  geom_smooth()
+  # geom_abline(intercept = 0, slope = 1) +
+  # ylim(0,0.6) +
+  # xlim(0, 0.6)
+
+
+tk2023 %>% 
+  mutate(regio_code = str_remove_all(lau_id, "M")) %>% 
+  select(regio_code, pvv2023=pvv) %>% 
+  left_join(tk2021 %>% select(regio_code, pvv2021 =pvv )) %>% 
+  mutate(diff = pvv2023-pvv2021) %>% 
+  # mutate(regio_code = fct_r(regio_code, diff)) %>% 
+  ggplot(aes(regio_code, diff)) +
+  geom_col() 
+  # coord_flip()
+  # geom_histogram()
+
+
+
+# openxlsx::read.xlsx("data/ignore/GWB2022_PC6_huisnr.xlsx")
+
+library(openxlsx)
+read_all_sheets = function(xlsxFile, ...) {
+  sheet_names = openxlsx::getSheetNames(xlsxFile)
+  sheet_list = as.list(rep(NA, length(sheet_names)))
+  names(sheet_list) = sheet_names
+  for (sn in sheet_names) {
+    sheet_list[[sn]] = openxlsx::read.xlsx(xlsxFile, sheet=sn, ...)
+  }
+  return(sheet_list)
+}
+
+gwb <- read_all_sheets("data/ignore/GWB2022_PC6_huisnr.xlsx")
+
+
+
+saveRDS(gwb %>% bind_rows(), "data/ignore/GWB2022_PC6_huisnr.rds")
+
+gwb <- readRDS("data/ignore/GWB2022_PC6_huisnr.rds")
+
+gwb %>% 
+  janitor::clean_names() %>% 
+  as_tibble() %>% 
+  mutate(pc4 = str_sub(pc6, 1,4)) %>% 
+  count(pc4, gem_code, sort = T) %>% 
+  count(pc4, sort = T) %>% 
+  filter(n == 2) %>% pull(pc4) %>% dput()
+  select(pc4, pc6)
+
+  doubles<-c("1261", "1724", "2114", "4062", "4197", "4715", "5091", "5383", 
+    "5504", "6367", "6574", "6924", "6961", "7011", "7351", "7693", 
+    "7694", "7933", "7963", "7964", "8096", "9207", "9417", "9423")
+
+
+tobeadded <- gwb %>% 
+    janitor::clean_names() %>% 
+    as_tibble() %>% 
+    mutate(pc4 = str_sub(pc6, 1,4)) %>% 
+    count(pc4, gem_code, sort = T) %>% 
+    filter(pc4 %in% doubles) %>% 
+    group_by(pc4) %>% 
+    # filter(pc4 == "4062")
+    arrange(desc(n)) %>% 
+    slice(1) %>% 
+    ungroup() 
+
+
+dict_pc4 <-  gwb %>% 
+    janitor::clean_names() %>% 
+    as_tibble() %>% 
+    mutate(pc4 = str_sub(pc6, 1,4)) %>% 
+    count(pc4, gem_code, sort = T) %>% 
+    filter(!(pc4 %in% doubles)) %>% 
+    bind_rows(tobeadded)  %>% 
+  mutate(regio_code = str_remove_all(gem_code, "M"))
+
+
+elex30 <- readRDS("data/election_dat30.rds")
+
+
+
+all_zip %>% 
+  group_by(pc4, party) %>%
+  summarize(total_spend = sum(total_spend)) %>% 
+  arrange(desc(total_spend)) %>% 
+  ggplot(aes(total_spend)) +
+  geom_histogram() +
+  scale_x_log10() +
+  facet_wrap(~party)
+
+
+all_zip <- election_dat30 %>% 
+  filter(type == "location") %>% 
+  filter(location_type == "zips") %>% 
+  mutate(total_spend_formatted = total_spend_pct*total_spend_formatted) %>% 
+  filter(!is_exclusion) %>% 
+  group_by(page_id, total_spend_pct, num_ads, party) %>%
+  mutate(n_together = n()) %>%
+  # select(page_id, value, num_ads, total_spend_formatted, n_together) 
+  mutate(total_spend_formatted_old = total_spend_formatted) %>% 
+  mutate(total_spend_formatted = total_spend_formatted/n_together) %>% 
+  select(party, page_id, value, n_together, total_spend_formatted, total_spend_formatted_old) %>% 
+  ungroup() %>% 
+  arrange(value) %>%
+  group_by(value, party) %>% 
+  summarize(total_spend = sum(total_spend_formatted),
+            total_spend_old = sum(total_spend_formatted_old)) %>% 
+  ungroup() %>% 
+  # filter(party == "VVD")  %>% 
+  dplyr::mutate(pc4 = str_remove_all(value, ", Netherlands") %>% as.numeric) %>% 
+  right_join(cbs %>% 
+              janitor::clean_names())
+
+parties_with_zips <- all_zip %>%
+  count(party) %>% distinct(party)
+
+
+all_zip %>% count(party)
+
+unique_parties <- parties_with_zips
+unique_pc4 <- cbs %>% janitor::clean_names() %>% select(pc4)
+all_combinations <- full_join(unique_parties, unique_pc4, by = character())
+merged_data <- left_join(all_combinations, all_zip, by = c("party", "pc4")) %>% 
+  mutate(total_spend = ifelse(is.na(total_spend), 0, total_spend),
+         total_spend_old = ifelse(is.na(total_spend_old), 0, total_spend_old)) %>% 
+  drop_na(party)
+
+spend_per_party <- merged_data %>% 
+  mutate(pc4 = as.factor(pc4)) %>% 
+  group_by(pc4, party, .drop = F) %>%
+  summarize(total_spend = sum(total_spend_old)) %>% 
+  ungroup() %>% 
+  # filter(party != "GroenLinks-PvdA") %>% 
+  mutate(pc4 = as.character(pc4)) 
+  # left_join(cbs %>% 
+  #             janitor::clean_names() %>% 
+  #             select(pc4))
+# spend_per_party %>% distinct(pc4)
+# tk2021 %>% 
+spend_per_party %>% count(party)
+
+tk2021 <- read_csv2("data/TK20210317.csv") %>% 
+  janitor::clean_names() %>% 
+  mutate(pvv = pvv_partij_voor_de_vrijheid/opkomst)
+
+spend_per_party %>% 
+  left_join(dict_pc4) %>% 
+  left_join(tk2021) %>% 
+  mutate(turnout = opkomst/kiesgerechtigden*100) %>% 
+  filter(!(party  %in% c("BBB", "Ja21", "Libertaire Partij", "D66"))) %>% 
+  ggplot(aes(total_spend, turnout)) +
+  geom_point() +
+  geom_smooth(method = "lm") +
+  scale_x_log10() +
+  facet_wrap(~party) +
+  ggpubr::stat_cor()
+
+
+
+spend_per_party %>% 
+  left_join(dict_pc4) %>% 
+  left_join(tk2021) %>% 
+  mutate(turnout = opkomst/kiesgerechtigden*100) %>% 
+  # filter(!(party  %in% c("BBB", "Ja21", "Libertaire Partij", "D66"))) %>% 
+  mutate(targeted = ifelse(total_spend>0, "Yes", "No")) %>% 
+  ggplot(aes(targeted, turnout)) +
+  geom_boxplot() +
+  # facet_wrap(~party) +
+  ggpubr::stat_compare_means() +
+  EnvStats::stat_mean_sd_text()
+
+
+
+spend_per_party  %>% 
+  # filter(party == "D66") %>% View()
+  right_join(dict_pc4)%>% 
+  group_by(regio_code, party) %>% 
+  summarize(total_spend=mean(total_spend, na.rm = T)) %>% 
+  ungroup() %>% 
+  left_join(tk2023 %>% 
+              mutate(regio_code = str_remove_all(lau_id, "M")) %>% 
+              select(regio_code, pvv2023=pvv) %>% 
+              left_join(tk2021 %>% select(regio_code, pvv2021 =pvv )) ) %>% 
+  mutate(diff = pvv2023-pvv2021) %>% 
+  drop_na(party) %>% 
+  ggplot(aes(total_spend,diff)) +
+  geom_point() +
+  geom_smooth(method = "lm") +
+  ggpubr::stat_cor() +
+  scale_y_continuous(labels = scales::percent) +
+  labs(y = "Increase Vote Share PVV") +
+  facet_wrap(~party, scales = "free")
+
+
+
+
+spend_per_party  %>% 
+  filter(party == "GroenLinks-PvdA") %>% 
+
+  # filter(party == "D66") %>% View()
+  right_join(dict_pc4)%>% 
+  group_by(regio_code, party) %>% 
+  summarize(total_spend=mean(total_spend, na.rm = T)) %>% 
+  ungroup() %>% 
+  left_join(tk2023 %>% 
+              mutate(regio_code = str_remove_all(lau_id, "M")) %>% 
+              select(regio_code, gl_pvd_a2023=gl_pvd_a, pop_2021, pop_dens_2) %>% 
+              left_join(tk2021 %>% 
+                          mutate(gl_pvd_a2021 = (partij_van_de_arbeid_p_v_d_a + groenlinks)/opkomst) %>% select(regio_code, gl_pvd_a2021)) ) %>% 
+  mutate(diff = gl_pvd_a2023-gl_pvd_a2021) %>%
+  mutate(total_spend = ifelse(is.nan(total_spend), 0, total_spend)) %>% 
+  drop_na(party) %>% 
+  # mutate(total_spend = total_spend/pop_2021) %>% 
+  ggplot(aes(pop_dens_2,diff)) +
+  geom_point() +
+  geom_smooth(method = "lm") +
+  ggpubr::stat_cor() +
+  # scale_x_log10() +
+  scale_y_continuous(labels = scales::percent) +
+  labs(y = "Increase Vote Share GL-PvdA (2021-2023)", x = "Population Density") +
+
+  # labs(y = "Increase Vote Share GL-PvdA (2021-2023)", x = "Loggeed Per Capita Spend on Postcode by GL-PvdA") +
+  theme_minimal()
+  # facet_wrap(~party, scales = "free")
+
+
+
+# when you look at it by party spending, the negative correlation seems to be somewhat driven by D66 though
+