Find hot spots

Even random spatial patterns exhibit some degree of clustering. In addition, our eyes and brains naturally try to find patterns even when none exist. Consequently, it can be difficult to know if the patterns in your data are the result of real spatial processes at work or just the result of random chance. This is why researchers and analysts use statistical methods like Find Hot Spots (Getis-Ord Gi*) to quantify spatial patterns. When you do find statistically significant clustering in your data, you have valuable information. Knowing where and when clustering occurs can provide important clues about the processes promoting the patterns you're seeing. Knowing that residential burglaries, for example, are consistently higher in particular neighborhoods is vital information if you need to design effective prevention strategies, allocate scarce police resources, initiate neighborhood watch programs, authorize in-depth criminal investigations, or identify potential suspects.

Note:

To use the Find hot spots tool, you must have the publisher role assigned to your account. Contact your organization administrator to request role changes.

Press the Show Map Contents button to display the Contents pane.
On the Contents pane, choose the point layer you want to analyze.
The point layer must contain at least 60 points to perform a hot spot analysis.
From the map menu, choose Tools > Find Hot Spots.
The find hot spots pane opens.
Choose whether you would like to find clusters of point densities or attribute values.
The available options depend on the number of points on your layer. Choose from the following:
- By point densities: This option is available only for point layers. Choosing this option constructs a fishnet mesh and places it over the points in the analysis layer. The number of points that fall within each fishnet square are then counted and analyzed. Only fishnet squares with at least one point are analyzed. Any statistically significant hot spots (red) in the results layer reflect spatial clusters of fishnet squares with high count values. Similarly, statistically significant cold spots (blue) reflect spatial clusters of fishnet squares with very low count values.
  When you choose to find hot and cold spots by point densities, you can restrict the analysis to a specific area of the map, rather than on the entire map. Check the Restrict analysis check box, and choose one of the following options from the drop-down menu:
  - to current map extent: restricts the analysis to the map area displayed in the viewer
  - by drawing custom areas: use the drawing tools to define a study area on the map where you want to perform analysis. Press Clear to discard the area polygons.
  - to areas in a layer: use polygons on another map layer to define the boundaries for your analysis.
  Note:
  The results layer is not a density surface, but rather it indicates locations where high or low point counts are too clustered to be the result of random processes. You must have at least 60 features on your layer to use this option.
  You can choose to specify an area layer with a minimum of 30 aggregation polygons that overlay at least 30 of your points. When you provide aggregation areas, analysis is performed on the point counts within each area. Check the Use point counts within areas in a layer check box, and choose a polygon layer from the drop-down menu.
- By attribute values: When you choose this option, you specify the attributes to use as the analysis field; attributes defined in your data appear in the drop-down menu. Using the field you provide, the Find hot spots tool creates a layer that shows areas with statistically significant clusters of high values (hot spots: red), low values (cold spots: blue), and areas that aren't part of a statistically significant cluster (shown in beige). You must have at least 30 items in your layer to use this option.
- To restrict the analysis to the map area displayed in the viewer, check the Restrict analysis to the map's current extent check box. To apply the analysis to the entire map, uncheck the check box.
In the Result options, type a name to assign to the new layer.
Press Run analysis.
When the analysis is complete, a new layer is created and appears in the Contents pane. For the points or the areas in this result layer, the darker the red or blue colors appear, the more confident you can be that clustering is not the result of random chance. Points or areas displayed in beige are not part of any statistically significant cluster; the spatial pattern associated with these features could very likely be the result of random chance. Sometimes, the results of your analysis will indicate that there aren't any statistically significant clusters at all. This is important information to have. When a spatial pattern is random, there are no clues about underlying causes. In these cases, all of the features in the results layer will be beige. When you do find statistically significant clustering, however, the locations where clustering occurs are important clues about what might be creating the clustering. Finding statistically significant spatial clustering of cancer associated with certain environmental toxins, for example, can lead to policies and actions designed to protect people. Similarly, finding cold spots of childhood obesity associated with schools promoting after-school sports programs can provide strong justification for encouraging these types of programs more broadly.
For technical details on how the Hot Spot tool works, see How Hot Spot Analysis works.
For more information about spatial statistics, see Spatial Statistics Resources.

Note:

You cannot change the styling properties of a hot spot layer.

Feedback on this topic?