Cartographic Skills Module 6: Data Classification

This week was all about data classification, or how to categorize your data in order to display it on a map. For the lab assignment, we worked with census data for Miami-Dade County, FL to map two related demographic variables using four different classification methods. As you can see, the classification method makes a big difference in how the data is ultimately visualized, which is why it's important to fully consider the nature of the dataset and the purpose of the map before choosing a classification method.

Pictured here are the equal interval, quantile, standard deviation, and natural breaks methods of classification. All four maps use the same dataset and show the number of senior citizens per square mile in each census tract (the other set of maps from this lab illustrated the percentage of the population over 65). In each case the data classification was performed automatically within ArcMap. In equal interval classification, the data is divided into classes that each cover the same range of data values, which in this case shows us that most of the county has a low density of people over 65, but doesn't show a lot of variation. The quantile method divides data into classes that each have an equal number of data observations, which definitely brings out the subtle variation in this dataset, although you have to look closely at the legend and realize that most of the classes are pretty close together in terms of the values they contain. The natural breaks method tries to group like values and make each class as different from the others as possible, which can lead to some weird ranges of values for those classes, but does bring out some variation in the data without overstating it. Standard deviation is a bit different. It classifies data values according to how far they fall from the mean (i.e. within one or two standard deviations above or below the mean). In this case, that shows us that the density for most of the county is below the mean, with a cluster of tracts in the northeast that are around or above the mean.