Essential Graphing • Treemaps • Additional Examples of the Treemap Platform
Previous
 • 
Next
Additional Examples of the Treemap Platform
The following examples further illustrate using the Treemap platform.
Example Using a Sizes Variable
Using the Cities.jmp sample data table, specify a Sizes variable to see a treemap of city sizes based on population.
1.
Select Help > Sample Data Library and open Cities.jmp.
2.
Select Graph > Treemap.
3.
Select POP (population) and click Sizes.
4.
Select city and click Categories.
5.
Click OK.
The report window appears.
POP as Sizes Variable
•
Cities with large populations are represented by large rectangles, such as New York and Los Angeles.
•
Cities with smaller populations are represented by smaller rectangles, such as Cheyenne and Dubuque.
Some rectangles are too small to show their labels (for example, Cheyenne and Dubuque). Click on a small rectangle to select the corresponding row in the data table, where you can see the label.
Example Using an Ordering Variable
For example, using the Cities.jmp sample data table, specify an Ordering variable as follows:
1.
Select Help > Sample Data Library and open Cities.jmp.
2.
Select Graph > Treemap.
3.
Select city and click Categories.
4.
Select POP (population) and click Ordering.
5.
Click OK.
The report window appears. Since you specified an Ordering variable, the cities in the treemap are ordered from bottom left (small cities) to upper right (big cities).
POP as Ordering Variable
Example Using Two Ordering Variables
For example, in the Cities.jmp sample data table, the X and Y columns correspond to the geographic location of the cities. Specify the X and Y columns as Ordering variables:
1.
Select Help > Sample Data Library and open Cities.jmp.
2.
Select Graph > Treemap.
3.
Select city and click Categories.
4.
Select X and click Ordering.
The X variable corresponds to the western and eastern US states.
5.
Select Y and click Ordering.
The Y variable corresponds to the northern and southern US states.
6.
Click OK.
The report window appears.
Treemap with Two Ordering Variables
The western and eastern US states are arranged horizontally, and the northern and southern US states are arranged vertically.
Example of a Continuous Coloring Variable
For example, using the Cities.jmp sample data table, specify a continuous Coloring variable as follows:
1.
Select Help > Sample Data Library and open Cities.jmp.
2.
Select Graph > Treemap.
3.
Select city and click Categories.
4.
Select OZONE and click Coloring.
5.
Click OK.
The report window appears.
City Colored by OZONE
Note that the size of the rectangles is still based on the number of occurrences of the Categories variable, but the colors are mapped to ozone values. The high ozone value for Los Angeles clearly stands out. Missing values appear as black rectangles.
Example of a Categorical Coloring Variable
For example, using the Cities.jmp sample data table, specify a categorical Coloring variable as follows:
1.
Select Help > Sample Data Library and open Cities.jmp.
2.
Select Graph > Treemap.
3.
Select city and click Categories.
4.
Select Region and click Coloring.
5.
Click OK.
The report window appears.
City Colored by Region
All of the cities belonging to the same region are colored the same color. The colors are chosen from JMP’s color theme.
6.
From the red triangle menu, select Layout > Squarify.
Squarify Treemap
The squares are now ordered according to size from the top left corner to the smallest rectangle in the lower right corner. This makes the data easier to analyze.
To order the squares using a combination of Split and Squarify, select the Mixed layout option.
Examine Pollution Levels
Using the Cities.jmp sample data table, examine the distribution of different pollution measurements (ozone and lead) across selected cities in the United States.
First, examine ozone levels in the selected cities.
1.
Select Help > Sample Data Library and open Cities.jmp.
2.
Select Graph > Treemap.
3.
Select POP (population) and click Sizes.
4.
Select city and click Categories.
5.
Select X and Y and click Ordering.
6.
Select OZONE and click Coloring.
7.
Click OK.
The report window appears.
OZONE Levels for Selected Cities
From OZONE Levels for Selected Cities, you observe the following:
•
Los Angeles has a high level of ozone. It is also a western state, and that is reflected by its positioning in the treemap.
•
Chicago and Houston have slightly elevated ozone levels.
•
New York and Washington have slightly lower-than-average ozone levels.
Next, examine lead levels in the selected cities. Perform the same steps as before, but substitute OZONE with Lead for the Coloring variable.
Lead Levels for Selected Cities
From Lead Levels for Selected Cities, you observe the following:
•
Cleveland and Seattle have high levels of lead. Cleveland is near the middle of the country, and Seattle is in the northwest. These locations are reflected by their positioning in the treemap.
•
Interestingly, most other cities have rather low lead levels.
•
Raleigh and Phoenix have missing values for lead measurements.
Examine Causes of Failure
This example uses the Failure3.jmp sample data table, which contains the common causes of failure during the fabrication of integrated circuits. Examine the causes of failure and when it occurs.
1.
Select Help > Sample Data Library and open Quality Control/Failure3.jmp.
2.
Select Graph > Treemap.
3.
Select N and click Sizes.
4.
Select failure and clean and click Categories.
5.
Select clean and click Coloring.
6.
Click OK.
The report window appears.
Failure Modes
From Failure Modes, you observe the following:
•
Contamination is the biggest cause of failure.
•
Contamination occurs more often before the circuits were cleaned, rather than after they were cleaned.
Examine Patterns in Car Safety
This example uses the Cars.jmp sample data table, which contains impact measurements of crash-test dummies in automobile safety tests. Compare these measurements for different automobile makes and models during the years 1990 and 1991.
1.
Select Help > Sample Data Library and open Cars.jmp.
Filter the data to show only the years 1990 and 1991 and create a subset of the Cars.jmp data table.
2.
Select Rows > Data Filter.
3.
Select Year.
4.
Click Add.
5.
Select 90 and 91 (hold down the SHIFT key).
The rows corresponding to 1990 and 1991 are highlighted in the data table.
6.
In the data table, select Tables > Subset.
7.
Ensure that Selected Rows is selected and click OK.
A new data table (Subset of Cars) appears that contains only the data corresponding to the years 1990 and 1991.
Now, using the Subset of Cars data table, create the treemap.
1.
Select Graph > Treemap.
2.
Select Wt (weight) and click Sizes.
3.
Select Make and Model and click Categories.
4.
Select L Leg and click Coloring.
L Leg represents a measurement of injuries resulting from the deceleration speed of the left leg, where more deceleration causes more injury.
5.
Click OK.
The report window appears.
Left Leg Deceleration Injuries
From Left Leg Deceleration Injuries, you can see that the Club Wagon and S10 Pickup 4x4 have the largest number of left leg deceleration injuries.
You can examine other safety measurements without re-launching the Treemap platform, as follows:
1.
From the red triangle menu, select Change Color Column.
2.
Select Head IC.
3.
Click OK.
The treemap updates to reflect head injuries instead of left leg injuries.
Head Injuries
From Head Injuries, you notice the following:
•
Although the S10 Pickup 4x4 had a high number of left leg deceleration injuries, it has a lower number of head injuries.
•
The Club Wagon still has a high number of head injuries, in addition to the high number of left leg deceleration injuries.
•
The Trooper II 4x4 had a low number of left leg deceleration injuries, but it has a high number of head injuries.