One-variable visualisation

Topic 2 · Day 4 · 2 hours

Christian González Martel
christian.gonzalez@ulpgc.es

Department of Quantitative Methods in Economics and Management · ULPGC

Juan M. Hernández Guerra
juan.hernandez@ulpgc.es

Department of Quantitative Methods in Economics and Management · ULPGC

April 29, 2026

Outline

  • Why visualise before you model.
  • Distribution at a glance: histograms, density, rug plots.
  • Robust alternatives: boxplots and violin plots.
  • Categorical: bars, lollipop, waffle.
  • From base R to ggplot2: the grammar of graphics.

Base R · quick and local

hist(hotels$price,
     breaks = 20,
     col    = "#0067a2",
     main   = "Distribution of nightly price",
     xlab   = "Price (€)")

ggplot2 · the grammar

library(ggplot2)

ggplot(hotels, aes(x = price)) +
  geom_histogram(bins = 30, fill = "#0067a2", alpha = 0.85) +
  labs(title = "Distribution of nightly price",
       x = "Price (€)", y = "Hotels") +
  theme_minimal(base_size = 12)

Histogram vs density

ggplot(hotels, aes(x = price)) +
  geom_histogram(aes(y = after_stat(density)),
                 bins = 30, fill = "#0067a2", alpha = 0.55) +
  geom_density(linewidth = 1) +
  theme_minimal()

Boxplot · robust summary

ggplot(hotels, aes(y = price)) +
  geom_boxplot(fill = "#f4f3ee") +
  coord_flip() +
  theme_minimal()

Categorical · ordered bar chart

library(forcats)

tourists |>
  count(origin_country) |>
  mutate(origin_country = fct_reorder(origin_country, n)) |>
  ggplot(aes(n, origin_country)) +
  geom_col(fill = "#0067a2") +
  labs(x = "Tourists", y = NULL) +
  theme_minimal()

Rules of thumb

  • One variable → one plot type: continuous → histogram/density; categorical → bar chart.
  • Order matters for categories — reorder by frequency or by value.
  • Minimise ink; use theme_minimal() (or theme_void() for maps).
  • Titles, units, source line — always.

Hands-on

Produce publication-ready histogram and boxplot of nights spent in an ISTAC series of your choice, for 2024. Push to your fork.