Wombling assesses local rates of change. Choose Detect Boundary > Wombling from the Data menu OR after right clicking on the leukemia dataset in the project window.
General Tab
Data: leukemia
New boundary name: local leukemia boundaries.
Detect boundary using: Single variable. Choose Cases/1000.
Choose not to standardize data before detection (clear the checkbox). Data standardization is only necessary when you analyze multiple variables.
Accept the defaults, choose to set thresholds using information provided below
Define a crisp boundary using the top 20% of Boundary Likelihood Values (BLVs).
Leave the box checked next to display histogram of BLVs.
Other Tab: Accept the gradient angle defaults.
Maximum angle between adjacent vectors: 90°
Minimum angle between gradient and boundary: 30°.
Look at the histogram of Boundary Likelihood Values, the Map, and the Table.