Introduction to the topic
In the first week we introduced geospatial analysis as:
A set of techniques for analyzing geospatial data
A set of techniques which results are dependent on the locations of the objects being analyzed
A set of techniques requiring access to both the locations of objects and also to their attributes
(after Goodchild)
Conventional data analytical techniques do not apply to geospatial/spatial data because of spatial autocorrelation, as described in Tobler’s First Law of Geography “everything is related to everything else, but near things are more related than distant things.”
In the next two lectures, we will examine spatial autocorrelation and the other challenges in analyzing geospatial data. We will also introduce the concepts that are essential to spatial analytic methods – distance, adjacency, interaction and neighborhood, and use two spatial statistics, global Moran’s I and Getis-Ord Gi*, to illustrate how the concepts are implemented in spatial analysis methods to address the challenges .
For those who are interested in the topics in spatial statistics , here is a good book:
- Geographic Information Analysis, 2nd edition. By David O’Sullivan and David Unwin.
- The Esri Guide to GIS Analysis, Volume 2: Spatial Measurements and Statistics, 2nd edition. By Andy Mitchell and Lauren Scott Griffin.
- Spatial Data Science. 8 recorded lectures from Luc Anselin, 2017.
