Spotting Differences - Data Points: Visualization That Means Something - page 241

face looks different than the others. The hair is higher and mouth width is
narrow, which corresponds to high free throw percentage and field goal
attempts, respectively.

To make the graphic more readable, you can add some more spacing
between rows, and at the least, provide a description of what feature pro-
vides what, as shown in Figure 7-12. Normally, I'd use a graphical legend,
but we used every facial feature, so it's a challenge to provide that many
pointers on a single face.

White space in
your graphics can
make them more
readable, espe-
cially when there
is a lot to look at
and evaluate.

Again, the usefulness of Chernoff Faces can vary by dataset and audience,
so you can decide whether you want to use the method. One thing though,
is that people who aren't familiar with Chernoff Faces, tend to take the
faces somewhat literally, as if the face that represents Shaquille O'Neal is
actually supposed to look like the player. Most of you know that O'Neal is
one of the physically biggest players of all time.

FIGurE 7-11 Chernoff Faces with names for players