To show an example of adding variables to Graph Builder, make a scatterplot of weight versus height for males and females.
1.
|
2.
|
Select Graph > Graph Builder.
|
3.
|
4.
|
‒
|
5.
|
Side-by-side scatterplots (one for each level of sex) replace the initial scatterplot. You now see weight versus height for males and females.
Start from the graph in Example of a height Plot. Add the weight variable to the left of the height variable. Click weight and drag and drop it into the X axis, to the left of height.
To demonstrate combining two continuous variables, start from the graph in Example of a height Plot. Merge the weight variable with the height variable. Drag and drop weight to the center of the X zone, slightly above the axis. Before you drop the variable, a blue quadrilateral appears.
1.
|
2.
|
Select Graph > Graph Builder.
|
3.
|
This variable represents the size of the vehicle. Eight levels are listed alphabetically on the X axis: compact (comp), heavy (hev), lightweight (lt), medium (med), mini, multi-purpose (mpv), pick-up truck (pu), and van. Since the levels are listed alphabetically, they are not ordered by size. Heavy comes before mini and lightweight. Another variable, Wt (weight) can act as a good substitute for size.
4.
|
Click Wt and drag and drop it just above the X axis. Before you drop the variable, a blue quadrilateral appears.
|
The levels of Size are now ordered according to the average Wt of all vehicles in that level, in ascending order. Notice that mini and lightweight are now ordered before heavy. The axis label is updated, signifying that an ordering variable is in use.
To verify that Size is actually ordered by Wt, click on Wt under Variables and drag and drop it into the Y zone. Example of Size Ordered by Wt shows that the average Wt increases from the left to right.
5.
|
Right-click in the X zone.
|
6.
|
Deselect Ascending.
|
The default ordering statistic is the mean. To use another statistic, right-click in the X zone, select Order Statistic and change it to the statistic you want. To remove the ordering, select Remove Order.
1.
|
2.
|
Select Graph > Graph Builder.
|
3.
|
4.
|
5.
|
To place the batch variable above the popcorn variable, drag and drop batch into the left side of the Group X zone. Before you drop the variable, a left-justified blue polygon appears.
|
6.
|
To place the batch variable below the popcorn variable, drag and drop batch into the right side of the Group X zone. Before you drop the variable, a right-justified blue polygon appears.
|
7.
|
You can replace an existing variable with an incoming variable. To demonstrate replacing variables, start from the graph in Example of yield by popcorn. Replace popcorn with batch in the Group X zone. Drag and drop batch into the center of the Group X zone. Before you drop the variable, a blue quadrilateral appears.
1.
|
2.
|
Select Graph > Graph Builder.
|
3.
|
4.
|
5.
|
Click the Histogram element icon.
|
1.
|
2.
|
Select Graph > Graph Builder.
|
3.
|
Drag and drop Country into the Y zone. You can resize the Graph Builder window and the graph if necessary.
|
Note: Notice that the default ordering for Country is ascending alphabetical (starting point is at the bottom). You can change the sorting order within a data table by using either the Value Ordering or Row Order Levels commands. For details, see the Using JMP book.
4.
|
5.
|
6.
|
Change the default side-by-side bars to stacked bars by right-clicking on the graph and selecting Bar > Bar Style > Stacked.
|
7.
|
Go back to the Oil Use.jmp sample data table.
|
10.
|
With the new column highlighted, select Cols > Formula.
|
11.
|
Click Consumption.
|
12.
|
Click +/-.
|
13.
|
Click OK.
|
14.
|
Select Graph > Graph Builder.
|
15.
|
Drag and drop Country into the Y zone. You can resize the Graph Builder window and the graph if necessary.
|
16.
|
17.
|
18.
|
Right-click on the graph and select Bar > Bar Style > Stacked.
|
You can clearly see consumption on the left in red, and production on the right in blue. Sort the countries by their oil consumption. You can use the Consumption column as an ordering variable.
19.
|
1.
|
2.
|
In the Tables panel at top left, click the red triangle next to Full Factorial Model and select Run Script.
|
3.
|
Click on the disclosure icon next to Effect Tests to open the report.
|
The popcorn*batch interaction has a small p-value (0.0026). From this, you conclude that there is a significant interaction between popcorn and batch.
4.
|
Save the model predictions to the data table. From the red triangle menu next to Response yield, select Save Columns > Prediction Formula.
|
Notice that a new column is added to the data table, Pred Formula yield.
5.
|
Save the prediction intervals to the data table. From the red triangle menu next to Response yield, select Save Columns > Mean Confidence Interval.
|
Notice that two new columns are added to the data table: Lower 95% Mean yield and Upper 95% Mean yield.
6.
|
Select Graph > Graph Builder.
|
7.
|
8.
|
9.
|
10.
|
Format the graph to see interval bars for Lower 95% Mean yield and Upper 95% Mean yield, and to see points for Pred Formula yield.
12.
|
13.
|
Change the bar style to interval by right-clicking on the graph and selecting Bar > Bar Style > Interval.
|
The interval bar style currently spans from Lower 95% Mean yield to Pred Formula yield, but you want it to span up to Upper 95% Mean yield.
14.
|
Remove the bar element for Pred Formula yield by right-clicking on the graph and deselecting Bar > Y > Pred Formula yield.
|
15.
|
Right-click on the graph and select Points > Y, and deselect the Lower 95% Mean yield and Upper 95% Mean yield individually.
|
16.
|
Right-click on the graph, and select the XXL option under Graph > Marker Size. Do this for each quadrant of the graph.
|
From Example of Predicted Values and Confidence Intervals, you can see the following relationships:
1.
|
2.
|
Select Graph > Graph Builder.
|
3.
|
4.
|
5.
|
Right-click the plot and select Points > Change to > Contour.
|
The darker the density, the more data. In Example of Contour Plot of Diamond Characteristics, notice the following: