Spatial statistics are distinct from general statistics in that they incorporate space in their calculations. Things that are near are considered more related than things that are far apart. For example, an analysis of the cost of rent may find a wide variation in the cost of rent across the country but a similar cost within cities and neighborhoods. This concept allows the tools in the Spatial Statistics toolbox to describe and model the effects of geography across a wide range of domains and problems.
Near is defined by associating locations based on their spatial relationships. In some cases, features are only considered near when they are contiguous while in others, features are associated based on a measure of distance. There are many methods to determine which features are near (neighbors) and the impact (weight) of those neighbors. Collectively, these methods are called conceptualizations of spatial relationships.
Neighborhood Explorer is a capability in ArcGIS AllSource that allows you to configure, visualize, and refine conceptualizations of spatial relationships, which are often used by the tools in the Spatial Statistics toolbox. Exploring neighborhoods is an important step when beginning and refining spatial statistics workflows.
Neighborhood Explorer allows you to do the following:
- Open a Neighborhood Explorer session for a layer in an ArcGIS AllSource project.
- Select and configure a conceptualization of spatial relationships.
- Explore conceptualizations of spatial relationships by selecting features on the map.
- Edit neighborhoods by adding and removing neighbors or altering the influence of existing neighbors.
- Create a custom conceptualization of spatial relationships by selecting the neighborhood and weighting method.
- Save and load a spatial weights matrix file.
Learn more about getting started with Neighborhood Explorer
Use cases
The following are examples of how you can use Neighborhood Explorer:
- Compare how different neighborhood and weighting methods configure neighborhoods in a feature class before running the Hot Spot Analysis (Getis-Ord Gi*) tool.
- Learn about conceptualizations of spatial relationships and compare their relative merits as they apply to your data.
- Edit and refine neighborhoods. The starting conceptualizations of spatial relationships may not capture all the nuanced relationships found in the real world. For example, if you apply contiguity to define neighborhoods, an island will not have neighbors; however, domain expertise may suggest that an island is well connected to a coastal community. You can use Neighborhood Explorer to edit the island's neighborhood so it includes the coastal community.