Exploring Data Visually - Data Points: Visualization That Means Something - page 485

VISUALIZING CATEGORICAL DATA

You might like to group people, places, and things. The classifications and
categorizations lend structure to what otherwise would be an amorphous blob
of stuff. Figure 4-5 shows some of your options to visualize such categories.

The bar graph, of course, is one of the most common ways to show categorical
data. Each rectangle represents a category, and the longer the rectangle is, the
greater the value that it represents. Whether a higher value means better or
worse can, of course, vary by dataset and point of view.

For example, in February 2012, the Pew Internet and American Life Project surveyed
approximately 2,200 people about how they use the Internet, social networking
sites such as Facebook and Twitter, and whether politics was a regular occurrence
on those sites. Figure 4-6 shows the results for four of the fifty questions.

Note: The Pew Internet & American Life
Project makes its survey data freely available at
http://www.pewinternet.org/.

As you might expect, Google was the most common cho-
sen search engine; Facebook was far ahead of Twitter and
career-based social network LinkedIn. The responses to the
other questions probably aren't that surprising to you either.

In this example, the bar graph is the visual equivalent of a list. Each bar rep-
resents a value, and you use separate bars and charts for separation. Length
is your visual cue, with rectangles placed on a linear scale. You could however
use a different scale and shapes to represent the same data. Figure 4-7 shows
the same poll results with squares sized by area.

Notice that the differences among categories doesn't look as dramatic in the
symbols plots as they do in the bar graphs. For example, the bar for Google
looks a lot longer than the rest in the search engine bar graph, but when you
compare the square for Google, it looks bigger, but not quite by the same
magnitude relative to the other squares.

Note: Because there aren't many catego-
ries for each question, the bar chart seems
like a better choice in this example, but you
don't need to rule out area as a visual cue
automatically.

This might be considered a drawback, but it can also be an
advantage when you have hundreds of values that vary by orders
of magnitude. With symbol plots, you can organize squares and
circles in any way you want in two-dimensional space, as shown
in Figure 4-8. On the other hand, bar graphs are restricted in that
each bar must start at the zero-axis and must extend straight
across or upward to the corresponding value.

FIGURE 4-4 (following page)
Patterns and visual cues