M4: Data Classification

 For the fourth lab in Cartography the assignment was to display 2010 Miami Dade census data as a choropleth map with four different data classification methods. When we choose to use a color gradient, there are a few ways that the data can be broken up into different color ranges. The way that the data is classified into different ranges for the choropleth can considerably change how the data is interpreted, and the "main idea" of the map. This assignment provided a great comparison of the different classifications, and practice normalizing data for area or sample size - something that is crucial to make choropleth maps that are not misleading. 

For the map I used the cartographic principles we had reviewed in other lessons to make the data stand out. 

The first map layout I created was one that looked at the percentage of seniors (65 and above) in census tracts. This layout was composed of four maps, one that classified the data with Natural Breaks, one with Quantile, one with Standard Deviation, and one with Equal Interval method. Though these are each mathematical ways for classifying the data, they were each already familiar to me with previous GIS work, and can be calculated as options in the symbology pane. I manipulated the color ramp and tone, as well as added a background to the layout to make my data stand out and be easy to read. 

This map shows data that has been presented with a normalization, however it could have benefitted from additional class numbers; perhaps 8-10 instead of 5. 

The second map layout was structured the same as the first, but instead of a percentage this set of maps displayed senior citizens as counts of individuals, normalized (adjusted for) the size of the census tract they were in. This gives a better "big picture view" statistically speaking, for the distribution of seniors through the general population of Miami Dade county. For this map layout I was unsure if I could use a custom number of classes for the data, I feel that some of the classes were swallowed up a bit by only having 5 classes, with most of the maps appearing largely light blue or periwinkle. However, the instructions were not clear on whether or not I could add classes to improve the display. The normalized data showed a belt of areas in the middle of the county where there were higher concentrations of seniors, while the percentage data highlighted an outlier census tract, along with (generally) a similar belt. The least accurate in the displays was the use of Equal Interval for the percentage data, as it makes it appear seniors are only found in on census tract in Miami-Dade County.