Chapter 4 Describing Your Data 143
Another important component of a distribution’s shape is the distribu-
tion’s tails—the values located to the extreme left or right edge. A distribution
with very extreme observations is said to be a heavy-tailed distribution.
The historic sample of home prices we’ve examined appears to be posi-
tively skewed with a heavy tail (because there are a number of houses lo-
cated at the high end of the price scale). This is not surprising, because there
is a practical lower limit for housing prices (around $50,000 in this sample)
and an exceedingly large upper limit.
Breaking a Histogram into Categories
You can gain a great deal of insight by breaking your histogram into catego-
ries. In the current example, we may be interested in knowing how the 1993
Albuquerque prices compared when broken down by location: Were certain
locations more expensive than others? One of the more desirable locations
in Albuquerque at the time was the northeast sector. Was this refl ected in a
histogram of the sample home prices? Let’s fi nd out.
To create a histogram broken down by categories:
1 Click the Housing Data worksheet tab to return to the price data.
2 Click Single Variable Charts from the StatPlus menu and then click
Histograms. Click the Data Values button, select Price from the list
of range names as the source for the histogram, and click OK.
3 Click the Break down the histogram by categories checkbox.
The various categories can be displayed in a histogram as stacked on
top of each other, side by side, or in three dimensions. You’ll see the
effect of these choices on the histogram’s appearance in a moment.
For now, accept the default, Stack.
4 Click the Categories button, click the Use Range Names option but-
ton, select NE Sector, and then click the OK button.
The NE Sector variable is a qualitative variable that is equal to Yes if
the home is located in the northeast sector and is equal to No other-
wise. Now, defi ne the options for the histogram’s bins.
6 Click the Chart Options dialog tab.
7 Click the Values button, click the Use Range References option but-
ton, and then select the range G1:G20 on the Price Table with Bins
worksheet. Click the OK button.
8 Click the Right option button to set how bin values will be counted
in the histogram.
