dendrogram

R code of the ggplot2 example

library(ggplot2)
library(dplyr)
library(tidyr)
library(gapminder)
library(ggdendro)

## One row per country, one column per year of life expectancy.
europe <- gapminder |>
  filter(continent == "Europe") |>
  select(country, year, lifeExp) |>
  pivot_wider(names_from = year, values_from = lifeExp)
trajectories <- as.matrix(europe[, -1])
rownames(trajectories) <- europe$country

tree <- dendro_data(hclust(dist(trajectories), method = "ward.D2"))

p <- ggplot() +
  geom_segment(data = segment(tree), aes(x = y, y = x, xend = yend, yend = xend),
               colour = "#535353", linewidth = 0.4) +
  geom_text(data = label(tree), aes(x = -1, y = x, label = label), hjust = 1, size = 2.2) +
  scale_x_continuous(labels = lex_number, breaks = \(l) scales::breaks_extended()(c(0, l[2])),
                     expand = expansion(mult = c(0.3, 0.02))) +
  scale_y_continuous(NULL, breaks = NULL, expand = expansion(add = 0.8)) +
  labs(title = "Portugal and Turkey cluster with Eastern Europe\nin how their life expectancy changed",
       subtitle = "Countries clustered by their life expectancy from 1952 to 2007",
       x = "Dissimilarity (Ward's method)",
       caption = "Data: Gapminder, via the gapminder R package") +
  theme_lex_viz(base_size = 8, grid = "x", arrows = "")