Mod6 - Data Classification

Data Classification

   This week the focus is on Classification Methods of data sets when rendered in maps. The completed map below is based on the percentage of the population that's above age 65 by census tract for Miami Dade County, FL.

  This map was created in ArcMap using 2010 census data from the United State Census Bureau and other data and instructions from University of West Florida GIS. 

  The five objective of this module are:

1. demonstration of four most commonly used data classification methods.
2. prepare one map with four data frames utilizing ArcMap software.
3. effectively symbolize the map for easy data acquisition/interpretation.
4. compare and contrast classification methods.
5. identify the best suited classification method for spatial data representation.

The classification methods focused on are: Equal Intervals, Natural Breaks, Quantile, and Standard Deviation. I will briefly expatiate on each of the classification methods. 

Equal Interval: The equal interval classification method presents the data by separating it into classes. It divides the total range (Max – Min value) by number of classes to be created. This method is one of the simplest and can be done by hand if necessary. However this method doesn’t take into account how the data falls along a number line and could lead to classes with no values in them. A positive of this method is that there are no gaps in the legend data could lead to confusion.

Natural Break: In this method, data are subjectively broken into classes and as alluded to above, it takes into account where the data is along a number line and tries to group data items based on where they occur most frequently. This tries to keep like values together and unlike values in separate classes.

Quantile : Classifying the data into quantile separates it into numbers of observations per class. This is done through ranking order, e.g. (lowest to highest number) until all data has been equally dispersed among the classes. This does not take into account any data clustering or “Natural Breaks.” Unlike the Equal Interval method there is (are) no chance of an empty class (es) skewing the map.

Standard Deviation : This method of classification does take into account where the numbers lie on a number line. The advantage and disadvantage to this method is that it relies on having normally distributed data. For events with data that follows a Gaussian or standard bell curve this is an excellent choice but, when not dealing with bell curve shaped style of data that contains roughly equal amounts of data points on either side of the mean standard deviation the result will most likely be skewed due to empty classes.

Out of the four classification methods we went over, I prefer Natural Breaks because it best visually represent the statistical data sets. It a fun lab!