Tidy Tuesday: Henley Passport Index

This week I made a plot to visualise which countries’ passport holders had the most travel freedom - relative to how many countries from which other people can enter without a visa.

The plot excludes countries like Australia where no one can enter visa-free (hence why New Zealand tops Oceania). I didn’t have time to problem-solve how to include these.

Things I learned this week:

• how to parse json data in R

• how to use {ggtext} to add italics to a subtitle (which I didn’t end up using)

• how to control the jitter of geom_point using ‘position = position_jitter(width = x)’

• there is a small island country called Niue that looks delightful

Code

tuesdata <- tidytuesdayR::tt_load(2025, week = 36)

country_lists <- tuesdata$country_lists
rank_by_year <- tuesdata$rank_by_year

library(jsonlite)
library(tidyverse)
library(ggtext)
library(paletteer)

free_access <- country_lists %>%
  select(code, country, visa_free_access) %>%
  rename(origin_code = code,
         origin_country = country) %>%
  
# Temporarily create a new column where the parsed json data will be stored (json_data)
  
  mutate(json_data = map(visa_free_access, ~ { 

# the map() function from purrr applies the following functions to every row (each json dataframe, represented by '.x') of the visa_free_access column
    
      jsdata <- fromJSON(.x) # converts/parses the json data into an R list
      as_tibble(jsdata[[1]]) # turns the list into a tibble
    })) %>%

# unnest the data so that it is no longer dataframes within a dataframe (lengthens the df)
  
  unnest(json_data) %>%
  rename(dest_code = code,
         dest_country = name) %>%
  select(-visa_free_access)

free_access <- free_access %>%
  group_by(dest_country) %>%
  count(name = "countries_with_free_access") %>%
  ungroup() %>%
  left_join(rank_by_year, join_by(dest_country == country)) %>%
  filter(year == 2025) %>%
  group_by(region) %>%
  mutate(relative_travel_freedom = visa_free_count - countries_with_free_access,
         median_rel_travel_freedom = median(relative_travel_freedom)) %>%
  ungroup() %>%
  mutate(region = fct_reorder(factor(region), median_rel_travel_freedom, .desc = TRUE))

# create dfs for the labelling (top and bottom 5)
top5 <- free_access %>%
  group_by(region) %>%
  slice_max(order_by = relative_travel_freedom, n = 1) %>%
  ungroup()

bottom5 <- free_access %>%
  group_by(region) %>%
  slice_min(order_by = relative_travel_freedom, n = 1) %>%
  ungroup()

p <- ggplot(free_access, 
            aes(region,
                relative_travel_freedom,
                color = region)) +
  geom_point(position = position_jitter(width = 0.1)) +
  geom_text(data = top5, 
            aes(label = dest_country),
            nudge_y = 10) +
  geom_text(data = bottom5, 
            aes(label = dest_country),
            nudge_y = -10) +
  labs(x = NULL, 
       y = "Relative Travel Freedom",
       title = "Passport Power by Region",
       subtitle = "Shows how many places passport holders can travel to without a visa – relative to how many types of passports\ncan enter their country without one.") +
  scale_color_paletteer_d("MetBrewer::Austria") +
  theme_minimal() +
  theme(legend.position = "none",
        plot.title = element_text(size = 20, face = "bold"),
        plot.subtitle = element_text(size = 14),
        axis.text = element_text(size = 12),
        axis.title = element_text(size = 12),
        plot.margin = margin(20,20,20,20))