Africa population map in R

Photo by Sergey Pesterev on Unsplash

This time for Africa… let’s see how are different countries doing in terms of population.

install.packages(c("rvest", "dplyr", "rio"))
library(rvest)
library(dplyr)
library(rio)

# extract data from wikidepdia and do the necessary cleaning

aa = read_html("https://en.wikipedia.org/wiki/List_of_African_countries_by_population") %>%
    html_nodes("table") %>%
    .[[1]] %>%
    html_table(fill = TRUE) %>%  
    setNames(make.names(names(.)))  %>%  select(2:5) %>%
    dplyr::rename_with(~ c("country", "percent", "pop", 'percent.growth'), c(1:4)) %>%
    mutate(across(2:4, ~ as.numeric(gsub("[^0-9]", "", .)))) 

Here is the cleaned data:

# A tibble: 56 × 4
   country      percent       pop percent.growth
   <chr>          <dbl>     <dbl>          <dbl>
 1 Nigeria          153 223804632             24
 2 Ethiopia          87 126527060             26
 3 Egypt             77 111942661             16
 4 DR Congo          70 102262809             33
 5 Tanzania          46  67438106             30
 6 South Africa      41  59414495              9
 7 Kenya             38  55100587             20
 8 Uganda            33  48582334             28
 9 Sudan             33  48109006             26
10 Algeria           31  45606481             16
# ℹ 46 more rows
# ℹ Use `print(n = ...)` to see more rows

And here is the population rank:

library(ggplot2)
library(dplyr)

ggplot(aa, aes(x = country, y = pop)) +
    geom_segment(aes(x = country, xend = country, y = 0, yend = pop), color = ifelse(str_detect(aa$country, "^A"), "#6C5B7B", "#A48374")) +
    geom_point(data = filter(aa, str_detect(country, "^A")), aes(color = "#6C5B7B")) +
    geom_point(data = filter(aa, !str_detect(country, "^A")), aes(color = "#A48374")) +
    scale_color_identity() + labs(x='', y='population')+
    coord_flip() +
    theme_minimal() +
    theme(
        panel.grid = element_blank(),  
        plot.background = element_rect(fill = "white", color = NA), 
        axis.text.x = element_text(size = 12))

Let’s make a map as well to have a better spatial understanding of the demographics…

library(sf)
library(rnaturalearth)
library(ggplot2)

africa_map <- world %>%
    left_join(aa, by = c("name" = "country")) %>%
    filter(continent == "Africa")

ggplot() +
    geom_sf(data = africa_map, aes(fill = pop)) +
    geom_label_repel(
        data = africa_map %>% filter(pop > 5e6), 
        aes(label = name, fill = pop,
            x = longitude, y = latitude)) + scale_fill_gradient(low = "#F1EDE6",   high = "#FF6F00") + 
       theme_void()