Page 336 - Statistics for Dummies

320
Part VI: The Part of Tens
Pie charts
Pie charts are exactly what they sound like: circular (pie-shaped) charts that
are divided into slices that represent the percentage (relative frequency) of
individuals that fall into different groups. Groups represent a categorical
variable, such as gender, political party, or employment status. Figure 20-1a
is a pie chart showing a breakdown of voter opinions on some issue (call it
Issue 1).
Here’s how to sink your teeth into a pie chart and test it for quality:
✓ Check to be sure the percentages add up to 100 percent, or close to it
(any round-off error should be small).
✓ Be careful when you see a slice of the pie called “other”; this is the
catch-all category. If the slice for “other” is too large (larger than other
slices), the pie chart is too vague. On the other extreme, pie charts with
many tiny slices give you information overload.
✓ Watch for distortions that come with the three-dimensional (“exploded”)
pie charts, in which the slice closest to you looks larger than it really is
because of the angle at which it’s presented.
✓ Look for a reported total number of individuals who make up the pie
chart so you can determine how big the sample was before it was divided
up into slices. If the size of the data set (the number of respondents) is
too small, the information isn’t reliable.
Bar graphs
A bar graph is similar to a pie chart, except that instead of being in the shape
of a circle that’s divided up into slices, a bar graph represents each group as
a bar, and the height of the bar represents the number (frequency) or per-
centage (relative frequency) of individuals in that group. Figure 20-1b is a rel-
ative frequency–style bar graph showing voter opinions on some issue (call it
Issue 1); its results correspond with the pie chart shown in Figure 20-1a.
When examining a bar graph:
✓ Check for the sample size. If the bars represent frequencies, you find the
sample size by summing them up; if the bars represent relative frequen-
cies, you need the sample size to know how much data went into making
the graph.

3/25/11 8:12 PM
29_9780470911082-ch20.indd 320
29_9780470911082-ch20.indd 320 3/25/11 8:12 PM