M4 Data Classification

 

This week’s lab is about data classification. Using census data from Census.gov and UWF GIS we analyzed the data of the over-65 age group in Miami-Dade County in south Florida. The classification methods for performing this task are equal interval, quantile, standard deviation, and natural break (Jenks) data. After the data are added to the map ArcGIS Pro does all the work with classification breakdown in the symbology pane. Rainbow color schemes were not used and should not be used when classifying this type of data, they should be a low to high intensity (light to dark).

We created one map for each classification method with two different layouts for a total of eight individual classification method maps. It is interesting to see how the methods compare to each other for each map. The first set of maps, layout 1, was data for the census tract and not normalized. The colors used for each one aid in presenting the data and display how skewed the information can be conveyed in maps. It brings to mind Monmonier’s book How To Lie With Maps, it can be easy to do unknowingly, or purposefully if one does not understand the data and the methods used to classify it. 

In creating these maps I found the standard deviation best defines the data. Because it emphasizes the variations above and below the mean it is ideal for population density. It might prove to be a slight challenge to an audience unfamiliar with standard deviation and the mean but in this case, I believe it is the best choice when compared to the other data classification methods.

In map 2 we normalized the data by population per square mile. In the normalized maps each classification method might offer a possibility to be a final choice in presenting the data to an audience. However, this is where understanding the methods and how they represent the data on the maps is crucial. Admittedly I need to understand more about all of these methods by creating maps with other types of data. Standard deviation seems to be the best choice because of how data relates to the average, whether above or below. It offers better statistical analysis calculated through ArcGIS Pro which makes it easier to display and analyze.