Plot proportion of female parliamentarians over time

ipu_pct_female.R

library(tidyverse)

### Write a function to download the data from the IPU
dl_data <- function(y, m) {
    url <- paste0("https://data.ipu.org/api/women-ranking.csv?load-entity-refs=taxonomy_term%2Cfield_collection_item&max-depth=2&langcode=en&month=",
                  m,
                  "&year=",
                  y)
    download.file(url, destfile = paste0("ipu_", y, m, ".csv"))
}

### Generate all year/month combinations
holder <- expand.grid(y = 2019:2023, m = 1:12)
holder <- holder |>
    filter(!(y == 2023 & m > 3))

### Download the data
apply(holder, 1, function(x) dl_data(x["y"], x["m"]))

### Read it in
filelist <- list.files(pattern = "ipu_")
holder <- list()
for (f in filelist) {
    holder[[f]] <- read.csv(f, skip = 5,
                            col.names = c("rank", "country",
                                          "elex", "seats",
                                          "women_count", "women_pct",
                                          "upper_elex", "upper_seats",
                                          "upper_women", "upper_women_pct"),
                            na.strings = c("-", "NA")) |>
        mutate(filename = f)
}

### Tidy up a little
holder <- bind_rows(holder)
holder$year <- sub("ipu_", "", holder$filename)
holder$year <- substr(holder$year, 0, 4)

holder$month <- sub("ipu_[0-9]{4}", "", holder$filename)
holder$month <- gsub(".csv", "", holder$month)

holder$date <- paste0(holder$year, "-", holder$month, "-01")
holder$date <- as.Date(holder$date, format = "%Y-%m-%d")

### Calculate the month/year average
plot_df <- holder |>
    group_by(date) |>
    summarize(mean_women_pct = mean(women_pct, na.rm = TRUE))

ggplot(plot_df, aes(x = date, y = mean_women_pct)) +
    geom_point() +
    scale_x_date("Date") +
    scale_y_continuous("Average percentage of women in parliament") + 
    geom_smooth(method = "lm", se = FALSE) +
    theme_bw()