Using Cartograms to Visualize Population Normalized Big-Data Sets

A map is often a useful way to visualize big-data sets that vary by region. For example, voting patterns by region, income levels by region, or tweet frequency by location are just some examples of data that benefit from being placed on a map. However, measuring any regional activity and placing it on a map is usually disappointing. Typically, whenever a bar chart or pie chart is placed on a map, it either covers something else up or visually disappoints in other ways. A heat-map overlaid on top of a map is better, but it tends to show areas of high activity but gives no way of highlighting areas between average levels and below average levels of activity. In this paper we show several examples of cartograms that solve this problem. A cartogram is a map in which a regional variable - such as population, Senate representation, income, patents issued, or tweet activity - is substituted for land area. The geometry or space of the map is distorted to convey the information of the regional variable in a much more realistic and visually persuasive manner.