Foundations of
Data Science

Spring 2026 | Data 2 (399)
Jeffrey M. Girard | Lecture 06c

Roadmap

• Customize the axis and plot labels in a figure
• Customize the appearance of scales in a figure

Labels

Labels

• By default, scales are named after data variables

• Changing names may enhance communication
  • Provide accessible names/definitions
  • Include units and other methods info.
  • Appear prettier and more professional

• We can also add plot-level information
  • Titles and subtitles for take-aways
  • Captions for asides or data sources

• This is all done by adding on labs()

Step 1: The Base Plot

# Create a base plot with default labels and scales
p1 <- 
  ggplot(mpg, aes(x = displ, y = hwy)) +
  geom_point(size = 3, aes(color = drv, shape = drv)) + 
  geom_smooth(method = "lm", color = "black")
p1

Step 1: The Base Plot

Step 2: Axis Titles

# Adjusting the positional (x and y) axis titles
p2 <- 
  p1 +
  labs(
    x = "Engine displacement (L)",
    y = "Highway fuel economy (mpg)"
  )
p2

Step 2: Axis Titles

Step 3: Legend Titles

# Adjusting the color and shape legend titles
p3 <- 
  p2 +
  labs(
    color = "Drivetrain",
    shape = "Drivetrain"
  )
p3

Step 3: Legend Titles

Step 4: Title, Subtitle, and Caption

# Adding context for key takeaways and data sourcing
p4 <- 
  p3 +
  labs(
    title = "Fuel efficiency generally decreases with engine size",
    subtitle = "Lightweight sports cars with rear-wheel drive break this rule",
    caption = "Data from fueleconomy.gov"
  )
p4

Step 4: Title, Subtitle, and Caption

Scales

Scales

• Scales are one of the basic gg elements

• So far, we have let R guess on the scales
  • Type (e.g., continuous or discrete)
  • Limits (minimum and maximum values)
  • Breaks (values to emphasize)

• But we can completely customize the scales!
  • Changes may enhance communication
  • We will add on various scale_ functions

Step 5: Y-Axis Scale

# Configuring the limits and breaks of the continuous y axis
p5 <- 
  p4 +
  scale_y_continuous(
    limits = c(0, 50),
    breaks = seq(0, 50, by = 10)
  )
p5

Step 5: Y-Axis Scale

Step 6: X-Axis Scale

# Configuring the limits and breaks of the continuous x axis
p6 <- 
  p5 +
  scale_x_continuous(
    limits = c(1, 7),
    breaks = seq(1, 7)
  )
p6

Step 6: X-Axis Scale

Step 7: Color Scale

# Manually specifying the color values and labels
p7 <- 
  p6 +
  scale_color_manual(
    breaks = c("r", "f", "4"),
    labels = c(
      "4" = "Four-Wheel",
      "f" = "Front-Wheel",
      "r" = "Rear-Wheel"
    ),
    values = c(
      "4" = "salmon",
      "f" = "cornflowerblue",
      "r" = "seagreen3"
    )
  )
p7

Step 7: Color Scale

Step 8: Shape Scale

# Matching the shape scale to the color scale for accessibility
p8 <- 
  p7 +
  scale_shape_discrete(
    breaks = c("r", "f", "4"),
    labels = c(
      "4" = "Four-Wheel",
      "f" = "Front-Wheel",
      "r" = "Rear-Wheel"
    )
  )
p8

Step 8: Shape Scale

Putting It All Together

ggplot(mpg, aes(x = displ, y = hwy)) +
  geom_jitter(size = 3, aes(color = drv, shape = drv)) + 
  geom_smooth(method = "lm", color = "black") +
  labs(
    x = "Engine displacement (L)",
    y = "Highway fuel economy (mpg)",
    color = "Drivetrain",
    shape = "Drivetrain",
    title = "Fuel efficiency generally decreases with engine size",
    subtitle = "Rear-wheel drive vehicles are often an exception due to sports cars",
    caption = "Data from fueleconomy.gov"
  ) +
  scale_y_continuous(
    limits = c(0, 50),
    breaks = seq(0, 50, by = 10)
  ) +
  scale_x_continuous(
    limits = c(1, 7),
    breaks = seq(1, 7, by = 1)
  ) +
  scale_color_manual(
    breaks = c("r", "f", "4"),
    labels = c(
      "4" = "Four-Wheel",
      "f" = "Front-Wheel",
      "r" = "Rear-Wheel"
    ),
    values = c(
      "4" = "salmon",
      "f" = "cornflowerblue",
      "r" = "seagreen3"
    )
  ) +
  scale_shape_discrete(
    breaks = c("r", "f", "4"),
    labels = c(
      "4" = "Four-Wheel",
      "f" = "Front-Wheel",
      "r" = "Rear-Wheel"
    )
  )

Putting It All Together