Controlling Axis Limits in ggplot2

ggplot2 is a powerful and versatile data visualization package in R. A common task when creating plots is to control the range of values displayed on the axes. This allows you to focus on specific regions of your data, improve plot clarity, and avoid misleading visualizations caused by extreme outliers. This tutorial demonstrates how to set axis limits using several methods in ggplot2.

Setting Axis Limits: The Basics

There are three primary ways to control axis limits in ggplot2: scale_x_continuous(), coord_cartesian(), and the shorthand xlim()/ylim(). Each approach has slightly different behavior, which we will explore.

1. scale_x_continuous() (and scale_y_continuous())

The scale_x_continuous() function (and its y-axis counterpart) directly modifies the data being plotted. It removes data points that fall outside the specified limits. This means the plot only displays data within the desired range.

library(ggplot2)

# Sample data
carrots <- data.frame(length = rnorm(500000, 10000, 10000))
cukes <- data.frame(length = rnorm(50000, 10000, 20000))
carrots$veg <- 'carrot'
cukes$veg <- 'cuke'
vegLengths <- rbind(carrots, cukes)

# Create a density plot
ggplot(vegLengths, aes(length, fill = veg)) +
  geom_density(alpha = 0.2) +
  scale_x_continuous(limits = c(-5000, 5000))

In this example, scale_x_continuous(limits = c(-5000, 5000)) ensures that only data points with lengths between -5000 and 5000 are displayed.

2. coord_cartesian()

coord_cartesian() offers a different approach. It adjusts the visible area of the plot without removing any data points. All data remains in the underlying dataset, but values outside the specified limits are simply clipped from the view. This is useful when you want to explore the full dataset but focus on a particular region.

ggplot(vegLengths, aes(length, fill = veg)) +
  geom_density(alpha = 0.2) +
  coord_cartesian(xlim = c(-5000, 5000))

Here, coord_cartesian(xlim = c(-5000, 5000)) restricts the displayed range on the x-axis, but the entire dataset remains available for analysis and potential further plotting.

3. xlim() and ylim() Shorthand

ggplot2 provides convenient shorthand functions xlim() and ylim() that act similarly to scale_x_continuous() and scale_y_continuous(), respectively, by removing data points outside the specified limits.

ggplot(vegLengths, aes(length, fill = veg)) +
  geom_density(alpha = 0.2) +
  xlim(-5000, 5000)

This is a concise way to achieve the same result as scale_x_continuous(limits = c(-5000, 5000)).

Choosing the Right Method

• scale_x_continuous() / xlim() / ylim(): Use these when you want to filter the data being displayed, effectively removing outliers or focusing on a specific range of values. This can be useful for creating cleaner visualizations or when you’re only interested in a subset of the data.
• coord_cartesian(): Use this when you want to zoom in on a specific region of the data without discarding any information. This is helpful for exploring the full dataset while focusing on a particular area.

Important Considerations

• Data Transformations: If you apply data transformations (e.g., log scaling), ensure your axis limits are appropriate for the transformed data.
• Coordinate Systems: Be mindful of the coordinate system you’re using. coord_cartesian() is designed for Cartesian coordinates.
• Flipped Coordinates: When using coord_flip() to swap x and y axes, you need to specify the limits within the coord_flip() function itself, rather than using coord_cartesian(). For example: + coord_flip(ylim = c(3, 5), xlim = c(100, 400))