2022_02_22_world_freedom_index.Rmd

---
title: "TidyTemplate"
date: 2022-02-22
output: html_output
editor_options: 
  chunk_output_type: console
---

# TidyTuesday

```{r setup, include=FALSE}

knitr::opts_chunk$set(echo = TRUE)

library(tidyverse)
library(tidytuesdayR)
library(scales)
theme_set(theme_light())

```

# Load the weekly Data

Dowload the weekly data and make available in the `tt` object.

```{r Load}
freedom <- tt_load("2022-02-22")$freedom %>%
  janitor::clean_names() %>%
  rename(civil_liberties = cl,
         political_rights = pr) %>%
  mutate(country_code = countrycode::countrycode(country, "country.name", "iso2c"))

freedom %>%
  count(year, sort = TRUE) %>%
  arrange(desc(year))

freedom %>%
  distinct(country, region_name)
```

Look just at 2020

```{r}
summarize_freedom <- function(tbl) {
  tbl %>%
    summarize(n_countries = n(),
              avg_civil_liberties = mean(civil_liberties),
              avg_political_rights = mean(political_rights),
              pct_free = mean(status == "F"),
              .groups = "drop") %>%
    arrange(desc(n_countries))
}
```


```{r}
by_region <- freedom %>%
  filter(year == 2020) %>%
  group_by(region_name) %>%
  summarize_freedom()

by_region %>%
  ggplot(aes(avg_civil_liberties, avg_political_rights)) +
  geom_abline(color = "red") +
  geom_point(aes(size = n_countries)) +
  geom_text(aes(label = region_name), vjust = 1, hjust = 1) +
  expand_limits(x = 0, y = 0, size = 0)

freedom %>%
  ggplot(aes(avg_civil_liberties, avg_political_rights)) +
  geom_point()
```

```{r}
freedom %>%
  filter(year == 2020) %>%
  ggplot(aes(civil_liberties, political_rights)) +
  geom_abline(color = "red") +
  geom_jitter(height = .2, width = .2) +
  # geom_text(aes(label = region_name), vjust = 1, hjust = 1) +
  expand_limits(x = 0, y = 0, size = 0)

freedom %>%
  summarize(sd(civil_liberties),
            sd(political_rights))

freedom %>%
  filter(year == 2020) %>%
  gather(metric, value, civil_liberties, political_rights) %>%
  mutate(metric = str_to_title(str_replace_all(metric, "_", " ")),
         region_name = fct_reorder(region_name, value)) %>%
  count(region_name, metric, value) %>%
  ggplot(aes(value, n)) +
  geom_col() +
  facet_grid(region_name ~ metric) +
  labs(x = "World Freedom Index rating",
       y = "# of countries",
       title = "Distribution of World Freedom Index by region in 2020")
```

civil_liberties is a scale of 1-7

```{r}
freedom_gathered <- freedom %>%
  gather(metric, value, civil_liberties, political_rights) %>%
  mutate(metric = str_to_title(str_replace_all(metric, "_", " ")),
         region_name = fct_reorder(region_name, value))

overall <- freedom_gathered %>%
  group_by(year, metric) %>%
  summarize(avg_rating = mean(value)) 

freedom_gathered %>%
  group_by(year, region_name, metric) %>%
  summarize(avg_rating = mean(value)) %>%
  ggplot(aes(year, avg_rating)) +
  geom_line(aes(color = region_name)) +
  geom_line(data = overall, size = 3) +
  facet_wrap(~ metric) +
  expand_limits(y = 1) +
  scale_y_reverse(breaks = seq(1, 7)) +
  scale_color_discrete(guide = guide_legend(reverse = TRUE)) +
  labs(x = "Year",
       y = "World Freedom Index rating",
       title = "World Freedom Index rating over time by region",
       color = "Region",
       subtitle = "Black line shows overall trend")
```

```{r}
freedom %>%
  ggplot(aes(civil_liberties)) +
  geom_histogram()
```

### Worldbank data

```{r}
library(WDI)
library(countrycode)

gdp_percap <- WDI(indicator = "NY.GDP.PCAP.CD",
                  extra = TRUE,
                  start = 1995,
                  end = 2020) %>%
  as_tibble()

freedom_joined <- freedom_gathered %>%
  inner_join(gdp_percap, by = c(country_code = "iso2c", "year"),
             suffix = c("", "_wdi")) %>%
  mutate(income = fct_relevel(income, c("Low income", "Lower middle income", "Upper middle income")))

freedom_joined %>%
  filter(income != "Not classified") %>%
  group_by(metric, income, year) %>%
  summarize(avg_rating = mean(value)) %>%
  ggplot(aes(year, avg_rating)) +
  geom_line(aes(color = income)) +
  geom_line(data = overall, size = 3) +
  facet_wrap(~ metric) +
  expand_limits(y = 1) +
  scale_y_reverse(breaks = seq(1, 7)) +
  scale_color_discrete(guide = guide_legend(reverse = TRUE)) +
  labs(x = "Year",
       y = "World Freedom Index rating",
       title = "World Freedom Index rating over time by region",
       color = "Worldbank Income",
       subtitle = "Black line shows overall trend")

freedom_joined %>%
  filter(year == 2020) %>%
  ggplot(aes(NY.GDP.PCAP.CD, value)) +
  geom_point() +
  geom_jitter(height = .2, width = 0) +
  facet_wrap(~ metric) +
  scale_x_log10()
```

```{r}
library(broom)

civil_liberties_2020 <- freedom_joined %>%
  filter(metric == "Civil Liberties",
         year == 2020,
         !is.na(NY.GDP.PCAP.CD))

lin_mod <- civil_liberties_2020 %>%
  lm(value ~ region_name + log2(NY.GDP.PCAP.CD), data = .)

library(ggrepel)
lin_mod %>%
  augment(data = civil_liberties_2020) %>%
  select(country, NY.GDP.PCAP.CD, region_name, income, value, .fitted, .resid) %>%
  arrange(desc(abs(.resid))) %>%
  head(20) %>%
  ggplot(aes(.fitted, value)) +
  geom_point() +
  geom_text_repel(aes(label = country)) +
  geom_abline(color = "red") +
  labs(x = "Expected freedom index based on region + income",
       y = "Actual freedom index",
       title = "What are the largest outliers?") +
  expand_limits(x = 1, y = 1)
```

```{r}
library(fuzzyjoin)

freedom_2020 <- freedom_joined %>%
  filter(year == 2020)

world_map_freedom_2020 <- map_data("world") %>%
  as_tibble() %>%
  regex_left_join(maps::iso3166, c(region = "mapname")) %>%
  left_join(freedom_2020 %>% select(-region), by = c(a2 = "country_code")) %>%
  filter(region != "Antarctica")

world_map_freedom_2020 %>%
  filter(metric == "Civil Liberties") %>%
  ggplot(aes(long, lat, group = group)) +
  geom_polygon(aes(fill = value)) +
  coord_map(xlim = c(-180, 180)) +
  scale_fill_gradient2(low = "blue",
                       high = "red",
                       midpoint = 3.5,
                       guide = guide_legend(reverse = TRUE)) +
  ggthemes::theme_map() +
  labs(fill = "Civil Liberties Rating",
       title = "World Freedom Index: Civil Liberties",
       subtitle = "In 2020")
```

```{r}
library(gganimate)

world_map_freedom <- map_data("world") %>%
  as_tibble() %>%
  regex_left_join(maps::iso3166, c(region = "mapname")) %>%
  left_join(freedom_joined %>% select(-region), by = c(a2 = "country_code")) %>%
  filter(region != "Antarctica")

world_map_freedom %>%
  filter(metric == "Civil Liberties") %>%
  ggplot(aes(long, lat, group = group)) +
  geom_polygon(aes(fill = value)) +
  coord_map(xlim = c(-180, 180)) +
  scale_fill_gradient2(low = "blue",
                       high = "red",
                       midpoint = 3.5,
                       guide = guide_legend(reverse = TRUE)) +
  ggthemes::theme_map() +
  transition_manual(year) +
  labs(fill = "Civil Liberties Rating",
       title = "World Freedom Index: Civil Liberties ({ current_frame })")
```




```{r}
freedom %>%
  distinct(country, country_code) %>%
  View()

gdp_percap %>%
  filter(str_detect(country, "Iran"))
```