Graphs are a key part of data analysis and communication. A typical workflow includes:

Defining a research question.
Collecting, cleaning, and transforming data.
Exploring patterns and refining hypotheses.
Communicating results clearly to stakeholders.

Visualizations are often the fastest way to make complex ideas understandable.

The ggplot2 package

This lesson focuses on creating charts with ggplot2, which is based on Wilkinson’s Grammar of Graphics (2005). ggplot2 is flexible and supports many geoms, themes, and transformations, but it does not provide true 3D plots or interactive graphics out of the box.

A ggplot can be built from:

Data
Aesthetic mappings (aes())
Geometric objects (geoms)
Statistical transformations (stats)
Scales
Coordinate systems
Faceting

The basic syntax is:

ggplot(data, aes(...)) + geom_*(...)

Scatter plot

We will use mtcars and start with a simple scatter plot.

Basic scatter plot

library(ggplot2)

ggplot(mtcars, aes(x = drat, y = mpg)) +
  geom_point() +
  theme_classic()

Scatter plot with groups

To color points by a categorical variable, map color inside aes(). Here we treat gear as a factor.

ggplot(mtcars, aes(x = mpg, y = drat, color = factor(gear))) +
  geom_point() +
  theme_classic() +
  labs(color = "Gears")

Transform axes (log scale)

A common strategy to reduce the impact of extreme values is to transform variables. Here we apply a log transform:

ggplot(mtcars, aes(x = log(mpg), y = log(drat), color = factor(gear))) +
  geom_point() +
  theme_classic() +
  labs(color = "Gears")

Add a fitted line

You can add a fitted regression line with stat_smooth(method = "lm").

my_graph <- ggplot(mtcars, aes(x = log(mpg), y = log(drat), color = factor(gear))) +
  geom_point() +
  stat_smooth(
    method = "lm",
    se = FALSE,
    color = "#C42126",
    linewidth = 1
  ) +
  theme_classic() +
  labs(color = "Gears")

my_graph

Make the plot informative

Good charts should be readable without extra documentation. Use labs() to add titles and axis labels.

Add a title

my_graph +
  labs(title = "log(drat) vs log(mpg)")

Dynamic title (example)

mean_mpg <- mean(mtcars$mpg)

my_graph +
  labs(
    title = paste("log(drat) vs log(mpg) — mean mpg:", round(mean_mpg, 2))
  )

Add subtitle and caption

my_graph +
  labs(
    title = "Relationship between log(mpg) and log(drat)",
    subtitle = "Colored by gear class",
    caption = "Source: mtcars (datasets package)"
  )

Rename axes and legend title

my_graph +
  labs(
    x = "log(miles per gallon)",
    y = "log(rear axle ratio: drat)",
    color = "Gears",
    title = "Relationship between log(mpg) and log(drat)",
    subtitle = "Colored by gear class",
    caption = "Source: mtcars (datasets package)"
  )

Control scales

You can control axis breaks with scale_*_continuous(breaks = ...).

my_graph +
  scale_x_continuous(breaks = seq(1, 3.6, by = 0.2)) +
  scale_y_continuous(breaks = seq(0.6, 1.6, by = 0.1)) +
  labs(
    x = "log(miles per gallon)",
    y = "log(rear axle ratio: drat)",
    color = "Gears",
    title = "Relationship between log(mpg) and log(drat)"
  )

Themes

ggplot2 includes multiple themes; switching theme changes the overall style.

my_graph +
  theme_dark() +
  labs(
    x = "log(miles per gallon)",
    y = "log(rear axle ratio: drat)",
    color = "Gears",
    title = "Relationship between log(mpg) and log(drat)",
    subtitle = "Colored by gear class",
    caption = "Source: mtcars"
  )

Save plots

ggsave() saves a plot to disk; by default it saves the last displayed plot, but it is safer to pass the plot explicitly.

directory <- getwd()
directory

## [1] "/home/runner/work/An-R-Tutorial-for-Beginners/An-R-Tutorial-for-Beginners"

plot_to_save <- my_graph +
  theme_dark() +
  labs(
    x = "log(miles per gallon)",
    y = "log(rear axle ratio: drat)",
    color = "Gears",
    title = "Relationship between log(mpg) and log(drat)",
    subtitle = "Colored by gear class",
    caption = "Source: mtcars"
  )

ggsave(
  filename = "my_fantastic_plot.png",
  plot = plot_to_save,
  dpi = 300
)

Note: opening a folder from R is OS-dependent; in many deployed environments (GitHub Actions, servers) it will not work, so it is usually better to rely on the printed path.

Summary

Objective	Example
Basic scatter plot	ggplot(df, aes(x, y)) + geom_point()
Color by group	ggplot(df, aes(x, y, color = factor(g))) + geom_point()
Transform variables (log)	ggplot(df, aes(log(x), log(y))) + geom_point()
Add fitted line	… + stat_smooth(method = ‘lm’, se = FALSE)
Add labels (title/subtitle/caption)	… + labs(title = …, subtitle = …, caption = …)
Control axis breaks	… + scale_x_continuous(breaks = …)
Change theme	… + theme_classic()/theme_dark()/…
Save plot	ggsave(‘plot.png’, plot = p)

A work by Gianluca Sottile

gianluca.sottile@unipa.it

Scatter Plot in R using ggplot2 (with Example)