bee-movie-in-baltic-states.R

library(tidyverse)
library(rvest)

box_office <-
  read_html("https://www.boxofficemojo.com/title/tt0389790/?ref_=bo_se_r_1") %>%
  html_nodes("h3 + table") %>%
  # Select "Europe, Middle East, and Africa" table only
  .[[2]] %>%
  html_table() %>%
  janitor::clean_names() %>%
  mutate(
    gross = parse_number(gross)
  )

population <-
  read_html("https://www.worldometers.info/world-population/population-by-country/") %>%
  html_node("#example2") %>%
  html_table() %>%
  janitor::clean_names() %>%
  mutate(
    population_2020 = parse_number(population_2020)
  )

box_office_per_capita <-
  box_office %>%
  # This join isn't perfect, but the countries we care about *are* matches, so meh
  left_join(population, by = c("area" = "country_or_dependency")) %>%
  mutate(gross_per_capita = gross / population_2020)

box_office_per_capita %>%
  filter(area %in% c("Latvia", "Lithuania", "Estonia")) %>%
  ggplot(aes(x = area, y = gross_per_capita)) +
  geom_col() +
  labs(
    title = "Was Bee Movie more popular in Lithuania?",
    subtitle = "Gross Box Office per capita ($)",
    x = NULL,
    y = NULL,
    caption = str_c(c(
      "Box office data: https://www.boxofficemojo.com",
      "Population data: https://www.worldometers.info"
    ), collapse = "\n")
  ) +
  theme_minimal() +
  theme(plot.title = element_text(face = "bold"))