Microsoft Office Tutorials and References
In Depth Information
Chapter 27: Data Mining Using Pivot Tables
You can try any of the other parameters: Product type, Gender, Marital Status,
Old/New Customer, or the type of Credit Card. It is easy. Drag and drop the dif-
ferent fields into the PivotTable and it will give you the information you need.
What I want to show now are the unconventional features that are of more
interest. Start with taking out the current parameters (Homeowners and Children),
leaving the table with the Average of Purchases only. See Figure 27.8.
Now drag Income to the left where the word Total is (under the Average of
Purchases header). The results are overwhelming. It is too much data to look at. See
Figure 27.9. The trend is clear if you look at the chart.
To be able to look at the table and make sense of the data, use the Grouping
feature of the PivotTable. Right-click a single cell in the Row data (column A) and
choose Group. See Figure 27.10. The Grouping menu attempts to group it for you
based on the data. It starts from 11,000, the lowest value in the data, and ends at the
highest value of 188,000. It will group it by increments of 10,000. I chose the lowest
number to be 10,000 and the highest 190,000.
The results of the grouping are shown in Figure 27.11. It is easier to follow. See
the results and the chart.
One more thing I want to do is to investigate the relationship between purchases
and the time of the year. Pull out (drag and drop) the Income, and replace it with the
FIGURE 27.8 Average of Purchases Only
FIGURE 27.9 The Upward Trend Is Clear on the Chart. It Is Dif
cult to View in the Table