Module 5 - Spatial Statistics

Whoever thought statistics was fun was wrong. Fortunately ESRI made a set a spatial analyst tools for ArcMap that makes the work a little easier. In Module 5 for Cartography we took a look at spatial statistical analysis. I remember taking statistics. I remember thinking how useful statistics could be and how excited I was to learn to apply some of that math to may day job as an analyst. I also remember how overwhelmed I got at the end of the class and how I wished statistics never went beyond the point of standard deviation and normal distribution. Fast forward to today and now I'm back to thinking how cool statistics can be because the tools we learned about this week made the work much easier.

For our lab we followed an ESRI course (Exploring Spatial Patterns in Your Data Using ArcGIS) and spatially analyzed some of the attributes of a vector point file. For the graphic below we looked at weather stations in Western Europe and the temperatures they reported. The graphic simply displays the locations of the weather stations, the mean location, the median location, and the directional distribution of the locations. The great part is that I did not have to do any of the math behind the scenes to create the graphic because the Spatial Statistics Tools and the Geostatistical Analyst Toolbar did all of the hard work for me.

The Spatial Statistics Tools made finding the mean, median, and directional distribution of the weather stations easy because it was all done with the click of a button. Simply pick the tool you want to use, input the file you want to analyze, and click okay. The tool will process all of the data and create a new file displaying the statistic you chose.

What's not pictured in the graphic above is all of the other statistical analysis that the software can do. Using the  Geostatistical Analyst Toolbar I created a histogram, a normal qq plot, a voronoi map, a semivariogram/covariance cloud, a crosscovariance cloud, and a 3D trend line. My two favorite tools from the week were the histogram and the voronoi map. The histogram was something I easily recognized and am comfortable using and the voronoi map was really interesting. The polygon and color distribution of the voronoi map was much more intuitive for me than looking at a bunch of numbers and I think it will be a useful tool when trying to create graphics to explain statistic that I have to analyze.