Data Visualization with ggplot2

Leykun (MSc)¹ & Yebelay (MSc)²

¹NDMC, EPHI    ²DMU & C4ED


April 28 – May 1, 2026

Module 3: Learning Objectives

By the end of this module you will be able to:

• Explain the Grammar of Graphics and how ggplot2 implements it
• Build plots layer by layer using ggplot() + geom_*()
• Choose the right chart type for categorical, numerical, and time-series data
• Customise titles, axes, colours, and themes
• Create faceted plots to compare subgroups
• Save publication-quality figures with ggsave()

Why ggplot2?

ggplot2 is built on a coherent Grammar of Graphics:

• A plot = data + aesthetic mappings + geometric layers
• Every component is explicit and modifiable
• Consistent syntax across all chart types
• Publication-quality output with minimal effort

library(ggplot2)   # or loaded automatically with library(tidyverse)

Note

In public health: the same ggplot2 skills apply whether you’re mapping disease burden, plotting age pyramids, or presenting outbreak curves.

Components of the layered grammar

• Data — raw dataset
• Aesthetics — map variables to x, y, colour … (aes())
• Geometries — shapes that draw the data (geom_*())
• Facets — sub-plots by group (facet_wrap())
• Statistics — computed summaries (stat_*())
• Coordinates — axis system (coord_*())
• Theme — fonts, grid, background (theme())

Common Aesthetic Mappings

Aesthetic	What it controls
x, y	Position on axes
colour	Outline / point colour
fill	Interior fill colour
size	Size of points / lines
shape	Point shape (0–24)
alpha	Transparency (0 = invisible, 1 = opaque)
linetype	Solid, dashed, dotted, etc.

Geometries (geom_*()) function

The general syntax is:

• ggplot(data = data, mapping = aes(mapings))+ geom_function()

• Geom Components

  Geom	Description	Input
  geom_histogram	Histograms	Continous x
  geom_bar	Bar plot with frequncies	Discrete x
  geom_point	Points/scattorplots	Discrete/continuous x and y
  geom_boxplot	Box plot	Disc. x and cont. y
  geom_smooth	Adds a smoothed conditional mean / regression line	Continuous x and y
  geom_line	Line plots	Discrete/continuous x and y
  geom_abline	Reference	line intercept and slope value
  geom_hline, geom_vline	Horizontal and vertical reference lines	yintercept or xintercept

The ggplot2 Template

Every ggplot follows this pattern:

ggplot(data = <DATA>, mapping = aes(<MAPPINGS>)) +
  <GEOM_FUNCTION>() +
  <OPTIONAL LAYERS>

The three required components:

Component	Function	Example
Data	The data frame	data = penguins
Aesthetic	Map variables to visual properties	aes(x = species, y = body_mass_g)
Geometry	The type of plot	geom_boxplot()

Building a Plot Layer by Layer

We use the penguins dataset (Palmer Archipelago, Antarctica) as a familiar reference before applying skills to surveillance data.

• data
• aesthetics
• geometries
• facets and add color

library(ggplot2)
library(palmerpenguins)
ggplot(data = penguins)

ggplot(data = penguins,
       aes(x = bill_length_mm, y = bill_depth_mm))

ggplot(data = penguins,
       aes(x = bill_length_mm, y = bill_depth_mm)) +
  geom_point()

ggplot(data = penguins,
       aes(x = bill_length_mm, y = bill_depth_mm,
           colour = species)) +
  geom_point(alpha = 0.7) +
  theme_minimal()

Part 1: Visualising Categorical Variables

Bar Chart with geom_bar()

ggplot(penguins, aes(x = fct_infreq(species), fill = species)) +
  geom_bar(show.legend = FALSE, width = 0.7) +
  labs(title = "Penguin Count by Species",
       x = "Species", y = "Count") +
  theme_minimal(base_size = 14)

Tip

fct_infreq() orders bars by frequency — always more informative than alphabetical ordering.

Stacked Bar Chart — Two Categorical Variables

ggplot(penguins, aes(x = island, fill = species)) +
  geom_bar(position = "fill") +   # position = "fill" shows proportions
  scale_y_continuous(labels = percent_format()) +
  labs(title = "Species Composition by Island",
       x = "Island", y = "Proportion", fill = "Species") +
  theme_minimal(base_size = 14)

Part 2: Visualising Numerical Variables

Histogram with geom_histogram()

ggplot(penguins, aes(x = body_mass_g)) +
  geom_histogram(bins = 30, fill = "#2E86AB", colour = "white") +
  labs(title = "Distribution of Penguin Body Mass",
       x = "Body Mass (g)", y = "Count") +
  theme_minimal(base_size = 14)

Density Plot — Comparing Groups

ggplot(penguins, aes(x = body_mass_g, fill = species)) +
  geom_density(alpha = 0.4) +
  labs(title = "Body Mass Distribution by Species",
       x = "Body Mass (g)", y = "Density",
       fill = "Species") +
  theme_minimal(base_size = 14)

Tip

Density plots are ideal for comparing distributions when group sizes differ — common in age-stratified epidemiology.

Boxplot — Distribution Across Categories

ggplot(penguins,
       aes(x = species,
           y = body_mass_g, fill = species)) +
  geom_boxplot(show.legend = FALSE) +
  labs(title = "Body Mass by Species (ordered by median)",
       x = "Species", y = "Body Mass (g)") +
  theme_minimal(base_size = 14)

Part 3: Relationships Between Variables

Scatter Plot with Regression Line

ggplot(penguins,
       aes(x = flipper_length_mm, y = body_mass_g)) +
  geom_point(alpha = 0.7, size = 2) +
  geom_smooth(method = "lm", se = FALSE) +
  labs(title = "Flipper Length vs. Body Mass",
       x = "Flipper Length (mm)", y = "Body Mass (g)",
       colour = "Species") +
  theme_minimal(base_size = 14)

Facets — Small Multiples

Facets create a separate panel for each level of a grouping variable.

ggplot(penguins,
       aes(x = flipper_length_mm, y = body_mass_g,
           colour = species)) +
  geom_point(alpha = 0.7) +
  facet_wrap(~species) +
  labs(title = "Flipper Length vs. Body Mass by Island",
       x = "Flipper Length (mm)", y = "Body Mass (g)") +
  theme_minimal(base_size = 13) +
  theme(legend.position = "bottom")

Part 4: Time Series

# Using measles data
measles_ts_data <- read_csv("data/measles_ts_data.csv", show_col_types = FALSE)
measles_ts_data |>
  ggplot(aes(x = date, y = measles_total)) +
  geom_line(color = "steelblue", linewidth = 0.9) +
  geom_point(color = "steelblue", size = 0.6) +
  labs(title = "Monthly Measles Cases (2012-2024)",
       x = "Date", y = "Total cases (monthly)") +
  scale_x_date(date_breaks = "2 year", date_labels = "%Y") +    
  theme_minimal(base_size = 14)

Customising Your Plot

Key theme() elements for polished slides and reports:

ggplot(penguins, aes(x = fct_rev(fct_infreq(species)), fill = species)) +
  geom_bar(show.legend = FALSE, width =0.7) +
  geom_text(stat = "count",
            aes(label = after_stat(count)),
            vjust = -0.5, fontface = "bold", size = 4) +
  labs(title = "Penguin Species Count",
       x = "Species", y = "Count") +
  scale_fill_manual(
    values = c("Adelie" = "#1f77b4", "Chinstrap" = "#ff7f0e",
               "Gentoo" = "#2ca02c")) +
  theme_minimal(base_size = 14) +
  theme(
    plot.title = element_text(face = "bold", hjust = 0.5, size = 16),
    axis.text  = element_text(size = 12)
  )

Saving Plots with ggsave()

# Assign your plot to an object
my_plot <- ggplot(penguins,
                  aes(x = flipper_length_mm, y = body_mass_g,
                      colour = species)) +
  geom_point() +
  theme_minimal()

# Save it
ggsave(
  filename = "figures/penguin_scatter.png",
  plot     = my_plot,
  width    = 8,
  height   = 5,
  dpi      = 300   # 300 dpi for publication quality
)

Exercise 1 — Customised Bar Chart

Task: Create a customised bar chart showing the count of students by race/ethnicity (race4).

Requirements:

1. Remove missing values from race4 before plotting.
2. Order categories by frequency, most frequent on the right.
3. Add the count as a bold, dark-red label above each bar.
4. Use custom fill colours as shown in the figure.
5. Center and bold the plot title.
6. Use theme_minimal() with increased base font size.

Exercise 2 — Zero-Dose by Region

Using measles_clean (from Module 2):

1. Calculate zero-dose prevalence (%) for each region
2. Create a horizontal bar chart ordered highest to lowest
3. Add percentage labels to the right of each bar
4. Use fill colour #D62828
5. Add a title, subtitle, and axis labels using labs()

Module 3 Summary

✅ ggplot2 builds plots layer by layer: data → aesthetics → geometry
✅ Chart selection depends on variable type (categorical, numerical, time)
✅ facet_wrap() creates small multiples for group comparison
✅ theme() gives full control over plot appearance
✅ ggsave() saves print-ready figures at 300 dpi
✅ ggplotly() converts any ggplot to an interactive visualisation

Resources

• ggplot2-book.org
• r4ds.hadley.nz/data-visualize