Data visualization of the day, unemployment edition

By Felix Salmon
April 15, 2009

Chris Wilson has the best graph of the day: an interactive map of US job losses since the beginning of 2007. Play the whole thing through once, and then take a closer look at what happens from month to month — especially from April 2008 to February 2009. And pity poor Detroit.

I do have a question about the scale, though: the “50,000 jobs” circle seems as though it has a much larger area than five of the “10,000 jobs” circles would have. But I don’t think they’re being silly and just working on the radius or the diameter, either. I wonder what they used for that.

Update: Chris Wilson provides graphical proof that the areas are right!


More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
2 comments so far

Hey Felix, thanks much for the link. To answer your question about scale: The circles are proportional in area to the number of jobs lost, so the radius (or diameter) will scale by the square root of the ratio between the larger and smaller number. So the 50,000 circle, which represents five times as many jobs as the 10,000 circle, is about 2.2 times wider (that is, sqrt of 5). This makes sense visually, I think, if you picture how many 10,000 dots put side-by-side would fit into the 50,000 circle — just over two. Five would fit perfectly if we could lump them together like putty, though this gets hard to visualize (at least for me).


Humans aren\’t very good are perceiving differences in areas so cartographers have often resorted to perceptual scaling to compensate.

More than you ever wanted to know here

@Chris: Great map–we need more dynamic time-series displays on the web…


Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see