Module 6 - Data Classification

An interesting and mathematically mind-numbing lecture and lab this week as we discussed data classification methods. We also touched on symbolization in our textbook readings, some of which made sense, like how to use different visual variables, but some was new to me, like all the different kinds of modeling for maps. For our lab assignment, we created a map displaying four types of classification method: Equal Interval, Quantile, Standard Deviation, and Natural Breaks, using a data category of percentage of people over the age of 65 in Escambia County, FL-according to the 2010 US Census.

The map itself was a breeze to create. Add the data, assign a classification method, choose color ramp, do layout-done. What took a bit longer was trying to understand each different classification map and determine which one best represented the data. I used my textbook as a starting point for how each method breaks up data into classes, and the disadvantages or advantages inherent in each one. I then went back and forth between the distribution graphs and visual output of each method, trying to...well "get it." Needing more help, I pulled in my mathematically-intelligent husband and went over with him what I thought each one meant and getting some background background knowledge. Standard deviation? I think I learned that, about 20 years ago. In the end, I just had to go with what I think I am seeing and understanding at this point. Each map has it's advantages and disadvantages, in my opinion though, the Natural Breaks method captured this particular data set the most accurately. I liked that I did begin to see ways the data could be represented better, like an additional class, or a few tweaks to the class break choices. More time with data and distribution and math is all I need to really get it. I even found it kind of fun (shh, don't tell my husband I said anything with numbers was fun).

Map of four different classification methods used to compare and contrast data presentation in regards to accuracy and user interpretation.