Side-By-Side Boxplots

Side-By-Side Boxplot

Figure 1: Side-By-Side Boxplot

Also known as a parallel boxplot or comparative boxplot, a  side-by-side boxplot is a visual display comparing the levels (the possible values) of one categorical variable by means of a quantitative variable. As its name implies, the side-by-side boxplot is constructed by placing single boxplots adjacent to one another on a single scale.

A side-by-side boxplot has all the advantages of a single boxplot (which can be seen here) with the added benefit of providing clear comparisons between levels in:

  • Range
  • Variance
  • Center
  • Maxima and Minima
  • Skewness

The simplicity and comparative power of the side-by-side boxplot is the reason it is used much more often than the single boxplot. (In fact, a google image search of the word “boxplot” actually shows more side-by-side boxplots than single boxplots.)

Skewed Side-By-Side Boxplot

Figure2: Skewed Side-By-Side Boxplot

Side-by-side boxplots are fairly straightforward and contain a large amount of information considering how simple they are. However, some data are better represented by a side-by-side boxplot than others; skewed data are particularly bad (as seen in Figure 2). Additionally, the formatting of the graph can be misused to make good side-by-side boxplots misleading;  here is a site that illustrates this.

Advertisements

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s