Love the message, hate the graphic

Meg McLain tells a great story about the relative risk of being killed by terrorists in the US.  Unfortunately, she comes up with this baffling graphic which appears to use the sort of number scales beloved of President Obama’s budget speechwriters:

Sure, there’s a scale problem, when the multipliers range from 6 to 17,600, but generations of scientists and engineers have handled that with a logarithmic scale:

This still doesn’t give the compressed range that MM’s chart shows.  Aha!  Perhaps she’s using the little-known log-log scale (beloved by statisticians who deal with generalized linear models)–let’s see:

Pretty close.  But how would any reasonable person expect a layperson to understand this exotic measurement scale?

Tip from the Knowledge Problem.  And from Thnik Again!

Update (28 September).  In her comments section, Dr Kiesling admonishes me not to “go all Tufte” and tosses out the phrase “mathematically pedant.”    How flattering!  I bet LK was the kind of girl who slugged guys in junior high to get their attention.

Bump charts get renamed as SLOPEGRAPHS

Charlie Park has a nice post describing Tufte’s slopegraphs (old chart, new name).

Kaiser Fung likes these a lot; he’s been calling them Bump charts.

I introduce these to my undergrads when we discuss the paired t test.

Tip from kottke.org.
Update (16 July).  James Kierstead publishes an R implementation.

Citation Networks, visualized

Check out this great visualization of the citation linkages in the sciences.

Tip from R-Bloggers.

Where the melting happened in the Melting Pot

The NY Times does total information overload about immigrant communities in the US–a fabulous display of a very dense data set.

Tip from Jonah Goldberg at The Corner.

R bubblecharts

Nice R bubblechart tutorial at Flowing Data.  Since ggplot2 is all the rage, I thought I’d give it a spin.  Here’s the result

qplot(data=crime, x=murder, y=burglary, 
	main="2008 State Crime Data", xlab="murders per 1000", ylab="burglaries per 1000") + 
	geom_text(aes(label=state), size=2.25, angle=45, hjust=1, vjust=1, color="blue") +
	geom_point(aes(size=population/1000), color="orange", alpha=0.7) +
	scale_size(to=c(3,15))