Visualizing Spatial Relationships - Data Points: Visualization That Means Something - page 306

FIGurE 8-29 Focusing on change
If you were to download urban population counts from the World Bank,
you'd have similar data to the previous example using access to improved
water. each row is a country, and each column is a year. However, the
urban population data is raw counts for an estimated number of people
in the country living in urban areas. A choropleth map of these counts
would inevitably highlight larger countries because they of course have
larger populations in general. Two maps to show the difference in urban
population between 2005 and 2009 wouldn't be useful unless you changed
the values to proportions. To do that, you'd have to download population
data for 2005 and 2009 in all countries and then do some simple math. It's
not all that hard to do that, but it's an extra step. Plus, if the changes are
subtle, they'll be hard to see across multiple maps.
Instead, you can take the difference and show it in a single map. You can
easily calculate this in excel or modify the previous Python script, and
then make a single map, as shown in Figure 8-30.
It's easy to see which countries changed the most and which ones changed
the least when you visualize the differences. In contrast, Figure 8-31
shows the proportion of each country's total population that lived in an
urban area in 2005.