Making Maps in R by Kevin Johnson.
from the post:
I make a lot of maps in my line of work. R is not the easiest way to create maps, but it is convenient and it allows for full control of what the map looks like. There are tons of different ways to create maps, even just within R. In this post I’ll talk about the method I use most of the time. I will assume you are proficient in R and have some level of familiarity with the ggplot2 package.
The American Community Survey provides data on almost any topic imaginable for various geographic levels in the US. For this example I will look at the 2012 5-year estimates of the percent of people without health insurance by census tract in the state of Georgia (obtained from the US Census FactFinder). Shapefiles were obtained from the US Census TIGER database. I generally use the cartographic boundary files since they are simplified representations of the boundaries, which saves a lot of space and processing time.
…
Occurs to me that getting students to make maps of their home states with a short list of data options (for a class), could be an illustration of testing whether results are “likely” or not. Reasoning that students are likely to have some sense of demographic distributions for their home states (or should).
I first saw this in a tweet by Neil Saunders.