Hot Spot Analysis

Juan_CarlosSoto_Palomo — Wed, 16 Mar 2011 00:27:18 GMT

Hi Lauren, Im new using the spatial stats tools, so am very excited to use these tools in one f my university projects. So what i did was to run the hot spot analysis for a dataset of districts from my country. I wanted to model the spatial distribution of a socioeconomic variable with the hot spot analysis at district scale.
First, i did some of the common tasks shown in the tutorials that you presented, for example use the "Calculate distance band from neighbor count" tool to ensure that all features had at least one neighbor.
Also, i performed the "Average nearest neighbor" tool to define the increments on the distance bands. Then i graphed the results and chose as the distance band the first peak in tha graph (By the way, it was the only peak, and from there the curve started to go down and down as distances increased). Also I chose as the conceptualization of relationships, the Fixed distance band because the polygon's size varies across the study area.
So im a little confused, because after obteined the results, i started to run the autocorrelation tool several times, and realized that even at smaller distances than the distance reported by "Calcule distance band from neighbor count''( wich i used as the initial distance) the data still autocorrelated,
im taking distances so small such as 100m and 500m considering that i'm working mainly with large
polygons (Districts polygons for the whole country).
I wonder if am doing something wrong or can i trust my results (I'll try to send you an image of the hotspot analysis map to give you an idea)

Thank you very much
Any suggestion will be appreciated

Re: Hot Spot Analysis

LaurenRosenshein — Tue, 22 Mar 2011 15:44:05 GMT

Hi Juan!

I'm glad to hear that you're finding the resources useful. It sounds like you are on the right track using the Incremental Spatial Autocorrelation tool to graph the intensity of spatial clustering at increasing distances, and its great that you found a pronounced peak, indicating that at that distance the spatial clustering of your phenomenon is most intense. All of that sounds great.

Your concern about running the Spatial Autocorrelation tool with a distance that does not ensure that every feature has at least one neighbor is very valid. If the distance that you choose does not ensure that every feature has at least one neighbor, then you cannot trust the results of your analysis. And this is also true for the Hot Spot Analysis tool, or any of the spatial statistics tools that use a conceptualization of spatial relationships. So, the first steps that you took by running the Calculate Distance Band from Neighbor Count and the Incremental Spatial Autocorrelation tool are important first steps to ensure that you do have enough neighbors being considered so that you can trust the results of your analysis.

Hope this helps!

Lauren Rosenshein
Geoprocessing Product Engineer

Check out the latest Spatial Statistics resources at http://esriurl.com/spatialstats

topic Re: Hot Spot Analysis in Spatial Statistics Questions

Hot Spot Analysis

Re: Hot Spot Analysis