ppepe-partyfacts.R

library(tidyverse)


# download and read Party Facts mapping table
file_name <- "partyfacts-mapping.csv"
if( ! file_name %in% list.files()) {
  url <- "https://partyfacts.herokuapp.com/download/external-parties-csv/"
  download.file(url, file_name)
}

partyfacts_raw <- read_csv(file_name, guess_max = 50000)

partyfacts <- 
  partyfacts_raw %>% 
  filter(! is.na(partyfacts_id)) %>% 
  select(country:name_english, partyfacts_id)

# link datasets (select only linked parties)
dataset_1 <- partyfacts %>% filter(dataset_key == "ppepe")
dataset_2 <- partyfacts %>% filter(dataset_key == "ches")
dataset_3 <- partyfacts %>% filter(dataset_key == "parlgov")
dataset_4 <- partyfacts %>% filter(dataset_key == "manifesto")

# Note: ParlGov import into Party Facts includes only parties with < 1% share in national parliament election

link_table <-
  dataset_1 %>%
  left_join(dataset_2, by = c("country" = "country", "partyfacts_id" = "partyfacts_id")) %>% 
  left_join(dataset_3, by = c("country" = "country", "partyfacts_id" = "partyfacts_id")) %>% 
  left_join(dataset_4, by = c("country" = "country", "partyfacts_id" = "partyfacts_id")) %>% 
  rename_with(~ str_replace(.x, fixed(".y.y"), "_manifesto")) %>% 
  rename_with(~ str_replace(.x, fixed(".x.x"), "_parlgov")) %>% 
  rename_with(~ str_replace(.x, fixed(".y"), "_ches")) %>% 
  rename_with(~ str_replace(.x, fixed(".x"), "_ppepe"))

# write results into file with dataset names in file name
file_out <- "partyfacts-linked.csv"
write_csv(link_table, file_out)