GIS 4006, Data Classification, Lab 4

In this lab, we experimented with displaying one population dataset in eight different ways -- by natural breaks, equal interval, standard deviation, or quantile classifications, both as a percent of population per census tract area, and normalized to absolute numbers per square mile. This was super interesting, because it displayed different patterns that emerged, and showed what was obscured or highlighted by different classifications. There are inherent tradeoffs in how you present your data, and no individual classification or display style will tell the full story or work in every situation. In the discussion questions, I ruminated on the relative pros and cons of displaying info about seniors as a percent of a census tract or normalized by square mile -- very large tracts could wash out seniors (a modifiable unit area problem), high concentration as a percent could obscure small populations, etc. Also, the population data alone doesn't tell the full story if you're using this data for real world problems. An affluent retirement community has different needs than rural seniors with transportation issues, and other trends like socioeconomic status, existing support resources, and individual support networks aren't displayed.