«
»

Introduction to Classifying Map Data


Step 1) Determine if there is a standard for data classification that you want to use. For example, analyses on impervious surface in the Northwest using 30 meter resolution data are often split into class-breaks of 5%, 10%, and 20% due to these being important breakpoints for environmental degradation in the Northwest*. Likewise, if there are a set of intuitive classes that make sense for the visualization, use those. Otherwise proceed to Step 2.

Step 2) Graph the data values. Determine if the data are skewed or normally distributed.

Step 3) Consult this chart as a starting point.

Step 4) Read more about classifications in a GIS text. Other considerations when classifying data include whether or not to normalize the data and whether or not the data might be suitable for classification by spatial proximity.

*However, when using finer resolution data, we’ve found that these values may not be applicable.

This entry was posted on January 25, 2012, 1:43 pm and is filed under Analysis, Best Practices. You can follow any responses to this entry through RSS 2.0. Both comments and pings are currently closed.

  1. #1 by @dropstones on January 25, 2012 - 1:47 pm

    Nice chart of different methods for reference: RT @PetersonGIS: Just wrote up an Introduction to Classifying Map Data http://t.co/gDCclRhk

  2. #2 by Jon Yungkans on January 25, 2012 - 3:18 pm

    Clear, concise and easy to follow. In short, excellent.

  3. #3 by John Nelson on January 26, 2012 - 10:26 am

    Great chart! I’ve been looking for a clear and concise description of these tradeoffs. Thanks Gretchen!

  4. #4 by Gretchen on January 26, 2012 - 10:31 am

    Thanks John – I’d like to point out to readers that John has a fantastic visualization of the quantile, standard deviation, and equal interval methods here.

Comments are closed.