Creating a Basic Box Plot

Let’s start with a basic example. Suppose we want to visualize the distribution of ozone levels in the airquality dataset. Here’s how you can create a plain box plot:

# Basic box plot for ozone levels
basic_box_plot <- ggplot(airquality, aes(x = factor(1), y = Ozone)) +
  geom_boxplot() +
  labs(title = "Basic Box Plot of Ozone Levels",
       x = "", y = "Ozone Levels") +
  theme_minimal()

basic_box_plot

Warning: Removed 37 rows containing non-finite values (`stat_boxplot()`).

In this example, we use ggplot() to initiate the plot and specify the x aesthetic as a factor to create a single box plot. The y aesthetic is set to the Ozone variable, and we add the geom_boxplot() layer to create the box plot itself. The labs() function helps us set the title and axis labels.

Adding Fill to Box Plots

If you want to add more visual depth to your box plots, you can use color to differentiate categories. Let’s create a box plot of ozone levels, grouped by the months:

# Box plot with fill for different months
filled_box_plot <- ggplot(
  airquality, 
  aes(
    x = factor(Month), 
    y = Ozone, 
    fill = factor(Month)
    )
  ) +
  geom_boxplot() +
  labs(title = "Box Plot of Ozone Levels by Month",
       x = "Month", y = "Ozone Levels") +
  scale_fill_discrete(name = "Month") +
  theme_minimal()

filled_box_plot

Warning: Removed 37 rows containing non-finite values (`stat_boxplot()`).

In this code, we add the fill aesthetic to the aes() function, which creates separate box plots for each month and fills them with different colors based on the Month variable.

Notching for Comparing Medians

A notched box plot can help you compare the medians of different groups. Let’s create a notched box plot to visualize the distribution of ozone levels for different temperatures:

# Notched box plot for ozone levels by temperature
notched_box_plot <- ggplot(
  airquality, 
  aes(
    x = factor(Temp), 
    y = Ozone, 
    fill = factor(Temp)
    )
  ) +
  geom_boxplot(notch = TRUE) +
  labs(title = "Notched Box Plot of Ozone Levels by Temperature",
       x = "Temperature", y = "Ozone Levels") +
  scale_fill_discrete(name = "Temperature") +
  theme_minimal() +
  theme(legend.position = "none")

notched_box_plot

By setting notch = TRUE within geom_boxplot(), you add notches to the boxes that provide a rough comparison of medians.

Base boxplot()

# Create a filled box plot of ozone by month
boxplot(
  airquality$Ozone ~ airquality$Month, 
  main = "Distribution of Ozone by Month", 
  xlab = "Month", 
  ylab = "Ozone", 
  col = "lightblue"
  )

Explanation:

• In this example, we use the formula notation (~) to create a filled box plot of the ozone variable (airquality$Ozone) grouped by the month variable (airquality$Month).
• We provide the same title, x-axis label, and y-axis label as in the previous example.
• Additionally, we specify the col argument to set the color of the boxes to “lightblue”.

Conclusion

Box plots are a fantastic tool for quickly understanding the distribution of your data. With the ggplot2 package in R, creating informative and visually appealing box plots is both accessible and customizable. I encourage you to experiment with different aesthetics, variations, and datasets to explore the insights these plots can reveal. So why not grab your R console and embark on your data visualization journey today? Happy plotting!

Remember, the best way to truly master box plots is by trying them yourself. Copy and paste the code snippets provided here into your R environment, modify them, and observe how the plots change. As you become more comfortable, you can start applying box plots to your own datasets and discover new patterns and trends. Happy coding!