I am a big fan of dot maps because I think that they help show concentrations of population indicators, prevalence of disease, etc. better than other styles of maps. Compelling examples are: New York Times covering Election 2004 and Census 2010 as well as Eric Fischer’s color coded race and ethnicity maps of major cities from Census 2000 data, based on Bill Rankin’s model.
This map by John Nelson is an excellent example of dot mapping for vote counts from county-level data. It is much prettier than the NYTimes election 2004 map, but these dot maps of vote counts give a wonderfully more accurate picture of how the country votes. No more red states or blue states, what matters is concentrations of people in specific regions. The “urban” vote has come up as a marker from this election, what if we could see precinct level voting data?
HT @JohnNelsonDIV & @AWoodruff