Skater
TerraView provides a tool for area regionalization to group small areas
(such as municipalities and neighborhoods) in larger contiguous areas
where associated attributes are similar to the small areas (such as
homicides and kidnappings). TerraView tool for this task is the SKATER
tool.
The
Skater (Spatial "k"luster Analysis by
Tree Edge Removal) method is a cluster analysis that considers the
spatial localization of objects. It is applied when a map is
partitioned in areas, each one with a given geographic localization,
and areas with similar characteristics must be part of the same spatial
grouping. The characteristics are obtained by values of measured
variables.
The variables are standardized due to the impact different scales may
have in the dissimilarity function and in the sum of squares in the
clusters. Therefore, variables to be used for regionalization have a
standard scale with mean zero and standard deviation one. The
standardization was obtained by subtracting the average from the
original value and dividing by the standard deviation.
Aggregation Method
Choose the criteria to define the number of larger regions (groups) to
be created. Clusters option
allows the definition of a fixed number and Population option allows the
definition of a minimum population for each group.
Attribute Selection
Select the information to be used to calculate similarity among areas. Layer Attributes list displays all
allowed (numeric) attributes and Selected
Attributes list displays the attributes selected for clustering.
It is
accessible through:
Plugins > Spatial Analysis >
Skater...
This
interface consists of the
following steps:
1.
Input Information:
- Layer Name: Select the
desired Layer.
- Load GPM: Loads a gpm
from file, if not, creates a new one.
- Attribute Link:
Defines the attribute that identifies the objects of this layer.
- GPM: Opens a dialog
to select a file with a desired proximity matrix.
2.
Operation Parameters:
- Number of Clusters:
Value that defines the total number of clusters generated in skater
operation.
- Minimum Population:
Value used to define the minimum population in each cluster.
- Population Attribute:
Attribute from input data that represents the population information.
- Skater By Groups: Uses
an attribute for
grouping the clusters.
- Layer Attributes: List
of all valid attributes that could be used to compute the skater.
- Selected Attributes:
List of selected attributes used to compute the skater.
3. Output
Information:
- Repository: Defines
where the output data will be saved.
- Layer Name: Defines the
name to create the output layer.
Click OK and then the skater
will be calculated.