Facets

geom_line

Suppose we have data on average voter turnout (%) in national elections over several years for the US and the UK. We want to see the trend in participation.

library(ggplot2)
set.seed(123)

# Toy dataset with US and UK
eg5 <- data.frame(
  year = rep(c(2000, 2004, 2008, 2012, 2016, 2020), times = 2),
  turnout = c(
    # US presidential elections
    54, 60, 62, 58, 56, 65,
    # UK general elections (closest years aligned to US election years for teaching)
    59, 61, 65, 66, 68, 67
  ),
  country = rep(c("United States", "United Kingdom"), each = 6)
)

Facets

geom_line

Suppose we have data on average voter turnout (%) in national elections over several years for the US and the UK. We want to see the trend in participation.

This is what it looks like:

Facets

geom_line

Suppose we have data on average voter turnout (%) in national elections over several years for the US and the UK. We want to see the trend in participation.

This is what we had previously:

ggplot(eg5, aes(x = year, y = turnout, color = country)) +
  geom_line()

Facets

geom_histogram

Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.

library(ggplot2)
set.seed(123)
# Toy survey dataset
eg6 <- data.frame(
  gender = rep(c("Men", "Women"), each = 20),
  trust = c(
    rnorm(20, mean = 5, sd = 2),
    rnorm(20, mean = 8, sd = 1)
  )
)

Facets

geom_histogram

Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.

This is what it looks like:

Facets

geom_histogram

Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.

This is what we had previously:

ggplot(eg6, aes(x = trust, fill = gender)) +
  geom_histogram(bins = 10, color = "white", alpha = 0.5)

Putting Plots Side by Side

ggarrange()

Sometimes you don’t want facets (splitting one dataset).

Instead, you might want to combine different plots — for example:

library(ggpubr)

# Plot 1: voter turnout
p1 <- ggplot(eg5, aes(x = year, y = turnout, color = country, group = country)) +
  geom_line(linewidth=1.2) +
  labs(title = "Voter Turnout")

# Plot 2: trust in government
p2 <- ggplot(eg6, aes(x = trust, fill = gender)) +
  geom_histogram(bins = 10, color = "white", alpha = 0.7) +
  labs(title = "Trust in Government")

# Arrange them side by side
ggarrange(p1, p2, ncol = 2)

Coordinates

Coordinates in ggplot2

• Control how data space is mapped to the plot space
• Can zoom in/out or flip axes
• Crucial for clarity and honesty in visualizations

Coordinates

coord_cartesian

Let us check out another toy example inspired by real data.

The data looks like this:

brexit_data <- data.frame(
  side = c("Leave", "Remain"),
  support = c(52, 48))

Coordinates

coord_cartesian

This is how we can plot our data:

ggplot(brexit_data, aes(x = side, y = support, fill = side)) +
  geom_col()

Coordinates

coord_cartesian

This is how we can truncate the axis (zoom the view) without dropping data:

ggplot(brexit_data, aes(x = side, y = support, fill = side)) +
  geom_col() +
  coord_cartesian(ylim = c(45, 55))  # Truncates axis and exaggerates gap

Coordinates

coord_cartesian

Moral: Truncated axes exaggerate differences.

Use coord_cartesian() for transparency.

Flipping

coord_flip

• coord_flip() swaps x and y axes, rotating the entire plot.

• Useful for bar charts with long category labels (improves readability).

• Often makes rankings and comparisons easier to interpret.

Flipping

coord_flip

Suppose we have survey data on average voter turnout rates (%) across different education groups. Here we don’t want to plot individual points — instead, we’re comparing aggregated values (categories on x, turnout on y).

# Pre-aggregated counts instead of raw rows
eg2 <- data.frame(
  education = c(
    "Primary", "Primary", "High School", "High School", "High School",
    "College", "College", "College", "College"
  ))

Flipping

coord_flip

Suppose we have survey data on average voter turnout rates (%) across different education groups. Here we don’t want to plot individual points — instead, we’re comparing aggregated values (categories on x, turnout on y).

ggplot(eg2, aes(x = education)) +
  geom_bar()

Flipping

coord_flip

Suppose we have survey data on average voter turnout rates (%) across different education groups. Here we don’t want to plot individual points — instead, we’re comparing aggregated values (categories on x, turnout on y).

This is how we can use coord_flip

ggplot(eg2, aes(x = education)) +
  geom_bar()+
  coord_flip()

Annotations

Why Annotations?

Raw plots ≠ finished plots.

Annotations direct the reader’s eye and add context.

Labels vs Annotations

• Labels: Titles, axis labels, captions (labs()).
• Annotations: Extra text, marks on the plot itself

Annotations

geom_text

We learned to use geom_text in this example:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

eg1 <- data.frame(
  democracy = c(-8, -7, -5, -3, 0, 2, 5, 8, 9),   # democracy score
  gdp = c(2, 9, 4, 7, 8, 20, 15, 25, 27),  # GDP per capita in $1,000s
  country = c(
    "North Korea",   # very autocratic, very poor
    "Saudi Arabia",  # autocratic, but richer due to oil
    "Zimbabwe",      # authoritarian, low GDP
    "Russia",        # hybrid regime, middle income
    "Nigeria",       # similar position
    "India",         # low–mid democracy, growing GDP
    "Brazil",        # democracy, mid GDP
    "Poland",        # consolidated democracy, higher GDP
    "South Korea"))   # rich democracy

Annotations

geom_text

We learned to use geom_text in this example:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

Annotations

geom_text

We learned to use geom_text in this example:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

# Scatterplot with geom_point
ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5)

Annotations

geom_label

We could also use geom_label in this example:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

# Scatterplot with geom_point
ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_label(aes(label = country), vjust = -1, size = 5) 

Annotations

geom_text_repel

We could also use geom_text_repel in this example to avoid clutter:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

library(ggrepel)

# Scatterplot with geom_point
ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text_repel(aes(label = country), size = 5)

Annotations

geom_label_repel

We could also use geom_label_repel in this example to avoid clutter:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

library(ggrepel)
# Scatterplot with geom_point
ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_label_repel(aes(label = country), size = 5)

Annotations

geom_text

We learned to use geom_text in this example:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

# Scatterplot with geom_point
ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) + 
  annotate("text", x = -7, y = 11, label = "Oil-rich outlier", 
           color = "red", size = 6, fontface = "bold") +
  annotate("segment", x = -7, xend = -7, y = 9, yend = 10.5, 
           arrow = arrow(length = unit(0.2, "cm")), color = "red")

Labs

Why labs()?

Every good plot needs context. labs() lets you set:

• Title / Subtitle — what the figure is about
• Axis labels — what each dimension represents
• Caption — notes, sources, disclaimers
• Legend title — clarify what colors/shapes mean

Labs

Example with Democracy & GDP

Previously, we plotted democracy score vs GDP with labels:

ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point() +
  geom_text(aes(label = country), vjust = -1, size = 5)

Labs

Adding Context with labs()

Now, we can improve readability:

ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) +
  labs(
    title = "Democracy and Wealth",
    subtitle = "Higher democracy scores often align with higher GDP",
    x = "Democracy Score (-10 = Autocracy, +10 = Democracy)",
    y = "GDP per Capita (in $1,000s)",
    caption = "Toy dataset, inspired by real-world patterns"
  )

Themes

What are Themes?

Control non-data elements of the plot.

Fonts, background, gridlines, legend placement, margins.

Do not affect the data → only the presentation.

Good themes = clarity + professionalism.

Themes

Built-in Themes: theme_grey()

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

# Scatterplot with geom_point
ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) + 
  theme_grey()

Themes

Built-in Themes: theme_bw()

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) + 
  theme_bw()

Themes

Built-in Themes: theme_minimal()

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) + 
  theme_minimal()

Themes

Built-in Themes: theme_classic()

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) + 
  theme_classic()

Themes

Built-in Themes: theme_void()

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) + 
  theme_void()

Themes

Tweaking a Theme

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

# Scatterplot with geom_point
ggplot(eg1, aes(x = democracy, y = gdp)) +
  geom_point(size = 5) +
  geom_text(aes(label = country), vjust = -1, size = 5) + 
  theme_minimal()+
  theme(panel.background = element_rect(fill = "white"), # Set background color
        axis.title.x = element_text(color = "#BF0404", face = "bold"), # Set x-axis label color
        axis.title.y = element_text(color = "#BF0404", face = "bold"),
        axis.line = element_line(color = "#BF0404", size = 1.5), 
        panel.grid = element_blank(), 
        panel.border = element_blank(), 
        panel.grid.major.y = element_line(color = "#40403E", size = 0.5, linetype = "dotted"))

Interactive Plots

Time Series

Suppose we have data on average voter turnout (%) in national elections over several years for the US and the UK. We want to see the trend in participation.

set.seed(123)
library(gganimate)

# Toy dataset with US and UK
eg5 <- data.frame(
  year = rep(c(2000, 2004, 2008, 2012, 2016, 2020), times = 2),
  turnout = c(
    # US presidential elections
    54, 60, 62, 58, 56, 65,
    # UK general elections (closest years aligned to US election years for teaching)
    59, 61, 65, 66, 68, 67
  ),
  country = rep(c("United States", "United Kingdom"), each = 6)
)

Interactive Plots

Time Series

Suppose we have data on average voter turnout (%) in national elections over several years for the US and the UK. We want to see the trend in participation.

Interactive Plots

Time Series

Suppose we have data on average voter turnout (%) in national elections over several years for the US and the UK. We want to see the trend in participation.

ggplot(eg5, aes(x = year, y = turnout, color = country, group = country)) +
  geom_line() +
  transition_reveal(year)+
  labs(
    subtitle = "Year: {round(frame_along)}"
  )

Interactive Plots

Time Scatterplots

Let us first create a panel data.

library(dplyr)
library(tidyr)

# Base cross-section
eg1 <- data.frame(
  democracy = c(-8, -7, -5, -3, 0, 2, 5, 8, 9),   # democracy score
  gdp = c(2, 9, 4, 7, 8, 20, 15, 25, 27),  # GDP per capita ($1,000s)
  country = c(
    "North Korea",
    "Saudi Arabia",
    "Zimbabwe",
    "Russia",
    "Nigeria",
    "India",
    "Brazil",
    "Poland",
    "South Korea"
  )
)

# Add a fake panel (2000–2020 every 5 years)
set.seed(123)  # reproducible "wiggles"

eg_panel <- eg1 %>%
  slice(rep(1:n(), each = 5)) %>%           # repeat each country 5 times
  mutate(year = rep(seq(2000, 2020, 5), times = nrow(eg1))) %>%
  group_by(country) %>%
  mutate(
    # let democracy scores drift a bit
    democracy = democracy + cumsum(runif(5, -0.6, 0.3)),
    # let GDP grow with some noise
    gdp = gdp + cumsum(runif(5, 0, 2))
  )

Interactive Plots

Time Scatterplots

Let us first examine the data

Interactive Plots

Time Scatterplots

We learned to use geom_text in this example:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

# Scatterplot with geom_point
ggplot(eg_panel, aes(x = democracy, y = gdp, label = country)) +
  geom_point(size = 5) +
  geom_text(vjust = -1, size = 5, show.legend = FALSE) +
  labs(
    subtitle = "Year: {round(frame_time)}"
  ) +
  transition_time(round(year)) +
  ease_aes('linear')

Interactive Plots

Interactive Maps

library(ggplot2)
library(sf)
library(rnaturalearth)
library(rnaturalearthdata)
library(ggiraph)
library(glue)
library(scales)


world <- ne_countries(scale = "medium", 
                      returnclass = "sf")
europe_bounds <- list(x = c(-10, 40),
                      y = c(35, 70))

# Mapping Them
# Interactive ggplot
p <- ggplot() +
  geom_sf_interactive(
    data = world,
    aes(
      fill = log(pop_est),
      tooltip = glue("{admin}: {comma(pop_est)}")  # hover text
    ),
    color = "white", linewidth = 0.1
  ) +
  coord_sf(
    xlim = europe_bounds$x,
    ylim = europe_bounds$y
  ) +
  theme_grey(base_size = 25) +
  scale_fill_viridis_c(option = "viridis")

# Render interactive plot with tooltips
girafe(ggobj = p)

Saving Figures

This was the first example we had:

Imagine we have data about 9 countries that record their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).

eg1 <- data.frame(
  democracy = c(-8, -7, -5, -3, 0, 2, 5, 8, 9),   # democracy score
  gdp = c(2, 9, 4, 7, 8, 20, 15, 25, 27)  # GDP per capita in $1,000s
)

Saving Figures

We save the plot to an object and then print it:

# Scatterplot with geom_point
pic<-ggplot(data=eg1, 
       aes(x = democracy, y = gdp)) +
  geom_point()
print(pic)

Saving Figures

The command to save a picture is ggsave

We can save it as:

• PDF → small file, best for documents and printing

# Square picture
ggsave(pic, filename = "pic.pdf")

• PNG → crisp, good for websites or presentations

ggsave(pic, filename = "pic.png")

• JPG → also common, but usually larger file size

ggsave(pic, filename = "pic.jpg")

Saving Figures

Controlling Shape

When you save, you can tell R how wide and tall the picture should be.

ggsave(pic, filename = "pic_square.png", width = 20, height = 20, units = "cm")

Saving Figures

Controlling Shape

When you save, you can tell R how wide and tall the picture should be.

ggsave(pic, filename = "pic_height.png", width = 10, height = 20, units = "cm")

Saving Figures

Controlling Shape

When you save, you can tell R how wide and tall the picture should be.

ggsave(pic, filename = "pic_width.png", width = 20, height = 10, units = "cm")

Saving Figures

Image Quality

Here’s a new word: dpi = dots per inch.

• Higher dpi = sharper image, but larger file.
• Lower dpi = fuzzier image, but smaller file.

ggsave(pic, filename = "pic_hi.png", width = 20, height = 20, units = "cm", dpi = 300)

Saving Figures

Image Quality

Here’s a new word: dpi = dots per inch.

• Higher dpi = sharper image, but larger file.
• Lower dpi = fuzzier image, but smaller file.

ggsave(pic, filename = "pic_lo.png", width = 20, height = 20, units = "cm", dpi = 30)