To Standardize or Not?
A very important consideration in thematic mapping is whether you want to present your data as raw counts / totals (e.g., the population of each country) or if you want to standardize your data to create rates / ratios (e.g., number of people in a each country divided by the area of that country = number of people per square mile/km). In this population example, the raw counts tell you how many people exist in each country, and the ratio tells you how tightly packed-in those people are. One of the primary reasons to standardize (or normalize) your data is to allow your readers to compare places that are very different. For example, you can directly compare a large place (like Canada) with a smaller place (like Switzerland). Although Canada has more people than Switzerland, it has a far lower population density; without standardization, that fact might not be obvious.
NOTE: If you want to make a choropleth map, you MUST map only standardized data.
Bottom line: If you want your map readers to understand magnitudes (how much of something exists), use map totals/counts. If you want your map readers to understand relative differences (that account for things like size differences), use standardized data.
Are My Data Already Standardized?
The might be! If the numerical data you are about to map are listed as “x per square mile/km/…” or “x per capita” or “percent” or “ratio of x / y” then you can skip the data standardization step.
How to Standardize Your Data
The two most common ways to standardize our data are to divide the data by (1) the area of the enumeration units—creating “x per square mile/km” data—or (2) by the number of people within those places—creating “x per capita” or “x as a % of the total population” data.