"We need to step it up a notch, from asking where we have lots of  X to exploring where we have higher than expected proportions of X," said ESRI spatial statistics expert Lauren Scott about the importance of using spatial statistics tools to analyze GIS data.

Scott's workshop at the Health GIS Conference and Homeland Security GIS Summit events in Denver, Colorado, drew a standing-room-only crowd of more than 200 GIS users eager to learn more about tools with names like the "Getis-OrdGi*" and "Anselin Local Moran's I." The two methods help identify "hot and cold spots" and "clusters and outliers" respectively, and are part of the updated spatial statistics toolbox available in ArcGIS 9.2.

 What's so hot about "cold" spots?

"They can tell us where people are living longer," Scott said. "They can show us where there are fewer than expected illnesses during an epidemic and identify the safest locations to place essential workers."

Scott's animated presentation drew comments such as "cool" and "I wish we had more math teachers like Lauren" from the audience. She provided a glimpse into the secret life of statisticians as they explore underlying spatial processes at work and display them as visual patterns on a map.

"There seems to be a big difference between how a traditional statistician views spatial autocorrelation and how a spatial statistician views spatial autocorrelation," said Scott. "The traditional statistician sees it as a bad thing that needs to be removed from the data (through resampling, for example); spatial autocorrelation violates underlying assumptions of many traditional (non-spatial) statistical methods. For the geographer or GIS analyst, however, spatial autocorrelation is evidence of underlying spatial processes at work and is an integral component of spatial data. These spatial processes and spatial relationships are a primary interest and one of the reasons we get so excited about spatial data analysis."

Learn More about Spatial Statistics

View a demo about analyzing the location suitability of 911 emergency response centers called
Using Spatial Statistics Tools  [03:47]

Listen to a podcast by David Maguire, Director of Products and International, ESRI called Spatial Analysis and Modeling with ArcGIS. (This podcast was recorded at the 2006 ESRI International User Conference. ) [1:17:55 | 26.7 MB]

Read a series of articles on Putting Spatial Statistics to Work in a special section of the Spring 2005 issue of ArcUser magazine.