Part 9 Creating a simple box plot

R provides many options for creating plots.²² Let’s start simple and create a box plot. Because our Teacher variable is a factor, we can use a box plot to see whether the range of responses on our composite variable, Energy, depends on whether or not the school professional is a teacher.²³

par(bty="l")
boxplot(dat$Energy ~ dat$Teacher,
        ylim=c(1,5),
        ylab="Energy level", 
        xlab="Is the respondent a teacher?")

Figure 9.1: Boxplot of Teachers on X axis and Energy level on Y axis.

We see that the teachers provided lower responses on the Likert-type questions than non-teachers. It seems (in our hypothetical data example) that teachers have lower energy levels than the other school staff on a Wednesday afternoon.

This example uses the base R graphics. We can get fancier plots using the ggplot2 package.↩︎
The tilde, ~, is used in R to indicate a formula. We will see this in regression and other statistical equations, where it serves the same role as the “=” operator (e.g., Y ~ b₀ + b₁X). Here, we’re asking R to make a box plot with a response (or outcome) variable, dat$Energy, based on the independent (or explanatory) variable,dat$Teacher. To match the scale of the Likert-type items, we set the scale of the Y axis to go from 1 to 5 using the ylim= argument. We have also specified the X and Y axis labels with the xlab= and ylab= arguments. The line par(bty="l") is for aesthetics. bty stands for box type and the l is a lowercase L to tell R to print only the left and bottom lines of the box (like the shape of the letter L).↩︎