Visualizing Populations

INFO VISUALIZATION: I found this visualization in the Times showing the difference between the topology of large cities in the States and China.







The point is well taken: China has a whole lot of new big cities.



However, I wonder if the representation of the U.S. is skewed due to the erratic nature of city, county, and state boundaries here. Yes, there are only 9 cities with a population of 1 million or more, but if you look at urban areas without regard to arbitrary political boundaries, the census reports 37 U.S. "cities" with populations larger then a million. That's four times the number represented in this graph. It can make a huge difference when you look at the whole picture. For example, New York's urban population doubles from around 8 million to just under 18 million (or 21 million according to this source) when you look at natural rather then political boundaries.



This visualization also gives the impression that there are 10 or so New York sized cities in China. That's a little too amazing to sound true. Here again if we look at cities without regard to political boundaries, the three largest urban areas are in North America:

1 New York (NY) 21,199,000

2 Mexico City 20,267,000

3 Los Angeles (CA) 16,373,000

In fact China doesn't even come into play until rank 17.

16 Istanbul 8,506,000

17 Shanghai 8,214,000

18 Washington (DC) 7,608,000

So, interesting as it is, the graph is a bit skewed in its representation. It would be very interesting to see this sort of display using a city's natural boundaries.