"Heat Map" of employers & employment level

Discussion created by lmisi on Aug 22, 2012
Latest reply on Aug 22, 2012 by mikeldub
I'm an economic analyst who has been tasked with becoming an ad hoc mapmaker in ArcGIS --- so I know just enough to cause trouble.

I've got a database of every healthcare industry employer in Michigan with their accompanying latitude and longitude.  I'm having an easy enough time mapping their X-Y coordinates, and changing the symbology to show employment size.

In addition, I'd like to create a heat map using the point data, in order to depict hot and cold spots.  This has given cause for my first foray into spatial analysis. 

The database consists of over 38,000 points, and as you might imagine, a great many of them are in the metro-Detroit area (southeast Michigan).  A few other clusters stand out throughout the state.  Given the rather dramatic clustering, I've had bad results with interpolation --- I believe this is because my data simply isn't evenly distributed.  (For example, when I run IDW, the clustering is so dramatic in southeast Michigan that I get a "hot spot" off in Canada.)

Do you folks have a suggestion for a method that might more successfully allow me to turn highly clustered point data into a thermal overlay?

Many thanks!