Cartographic Skills: Data Classification

This week we learned about Data Classification for properly displaying data within a map. There are many methods of displaying data. We looked at four methods of classifications and built a comparison map of each classification. We used 2010 Census Bureau data and preformed classification methods on the percentage and total population of ages 65 and older, within census tracts for Miami Dade County Florida. Below is a map using four methods of classification; Natural Break, Quantile, Equal Interval and Standard Deviation.

The map above displays four comparison methods of classification using the percent of population ages 65 and up for census tracts in Miami Dade County, Florida. The classification methods used are; Natural Breaks, Quantile, Equal Interval and Standard Deviation. Each method of classification displays data differently.

Natural Break classification groups data “naturally” into classes by using an algorithm to find spikes or clusters where the data is found within the range of data distribution. This type of classification works good for the percentage of age 65 and up map as it finds breaks in the data where there might be larger groupings and separates the percentages into appropriate views. However, map readers may not understand the legend very well in determining how the data is broken up into each class.

The Quantile method of classification divides the total distribution of data into equal numbers of observations. Each class contains a range value obtained by determining a midpoint between the highest and lowest value for each class. I believe the Quantile method is the best method of classification for displaying the percentage of population ages 65 and up. The data range shows the highest percent of values placed in the last class. Showing where the highest values exist, map readers can compare against the lower values and easily understand the breakdown of data distribution. Also the legend is easily understood in how the ranges are broken up. 

Equal Interval classification shows the data within equal ranges. Basically the data is equally split up into breaks or classes. Ranges for the data are determined by dividing the total data range by the number of classes used. This type of classification does not display data very well for the percent of population ages 65 and up map. Since the percent values are equally spread across the number line and a class of ranges are equally divided, the data on the map looks very even and is hard to compare neighboring values because the data is equally dispersed.

The Standard Deviation considers how data is distributed along the number line. Classes are formed from a mean deviation and broken up by adding or subtracting values. This classification does not show the data very well for the percent of population map. It’s very confusing to understand and hard to portray the data appropriately as the map reader must have an understanding of statistical analysis.