## Some of the Neatest Statistics You’ve Never Seen

By Sean Rule, Math

 Figure 1: from Michael Blastland’s “Go Figure” column, March 11, 2009

Before you read any further, please take a look at the figure above for a few moments.  Pretty, isn’t it?  Now look deeper…what’s it showing you?  (resist the urge to click on the hyperlink).  For you youngsters out there who don’t remember the blockbuster 1997 movie, two hints: 1) big boat; 2) iceberg.

What strikes me about this graph is not just its beauty, although that sure is a large part of what made me say, “Wow!” when I first saw it.  No, combined with its beauty is its function.  It represents data in a non – traditional way.  To contrast, here’s how some similar data is presented in Wikipedia (among other places):

 Category Number aboard Number of survivors Percentage  survived Number lost Percentage lost First class 329 199 60.5 % 130 39.5 % Second class 285 119 41.7 % 166 58.3 % Third class 710 174 24.5 % 536 75.5 % Crew 899 214 23.8 % 685 76.2 % Total 2,223 706 31.8 % 1,517 68.2 %

Figure 2: From Wikipedia

Both represent similar ideas, but I feel the graph in Figure 1 (using parallel sets) reveals a richer data experience.  I also think it contains a “Wow!” factor that might be lost by simply looking at a data table (like figure 2) or histogram.

Here’s another example with (in my opinion) a high “wow” factor:

Figure 3: From Grande Reportagem’s “Meet the World” campaign

(Note:  in case you can’t read it in this image, the colors of the flag represent world statistics:  red = working 14 years old; yellow = studying 14 years old)

Again, these are statistics that could just as easily be presented in a pie chart or table.  But ask yourself: would a pie chart or table grab your attention like the flag did?  And, wasn’t it neat that you didn’t know what you were supposed to get from it right away?  When I use the flag images in MTH 244, I relish the pregnant pause that comes immediately after the students see them.  Then, a moment later, the class lets out a collective, “whoa” when they connect with the representation.  Stellar.

Let’s talk population, shall we?  What grabs your attention more, the following table…

 Rank City State Population 1. New York New York 8,214,426 2. Los Angeles California 3,849,378 3. Chicago Illinois 2,833,321 4. Houston Texas 2,144,491 5. Phoenix Arizona 1,512,986 6. Philadelphia Pennsylvania 1,448,394 7. San Antonio Texas 1,296,682 8. San Diego California 1,256,951 9. Dallas Texas 1,232,940 10. San Jose City California 929,936 11. Detroit Michigan 871,121 12. Jacksonville Florida 794,555 13. Indianapolis Indiana 785,597 14. San Francisco California 744,041 15. Columbus Ohio 733,203 16. Austin Texas 709,893 17. Memphis Tennessee 670,902 18. Fort Worth Texas 653,320 19. Baltimore Maryland 631,366 20. Charlotte North Carolina 630,478

Figure 4a: Populations of US cities, 2006 (US Census Bureau)

…or the image that Time Magazine published in October, 2006, showing the same information?

 Figure 4b: From Time’s America by the Numbers  (click on the link for a clearer display of image)

It never ceases to amaze me when an image like the one above stops me in my tracks (I also like that Alaska is sized properly).  However, it’s easy to realize why there aren’t more like it around.  Aren’t blah – blah tables easier to create?  Sure.  But I also want an answer to this: if we can build a better data construct, that reaches out to more people in more ways…shouldn’t we?

I’d like to close with some examples form one of my favorite artists.  Check out the following:

 Figure 5a: From Chris Jordan’s “Running the Numbers” series

Oh!  I get it!  It’s impressionistic, right?  Mmmmm…some French guy?  Well, sort of…look more closely at the ladies with the parasol in the middle of the piece:

 Figure 5b: …a closer look…

Now closer still…

 Figure 5c: …A – ha!

What you eventually realize is that the picture is made up entirely of soft – drink cans…106,000 to be exact.  Chris tells us that the US uses 106,000 such cans every thirty seconds. Stop and think about that for a second or two, OK?  Somehow, the idea “106,000 cans” and the image above, while conveying the same message, could not be more different.

What’s neat, also, about Chris’ art is its scale; this piece, Cans Seurat, is 5 feet high and 8 feet wide, a size he uses frequently.  However, that’s one of his smaller pieces.  Consider his piece Building Blocks, which depicts 9 million toy blocks, equivalent to the number of children without health care coverage in 2007:

 Figure 4a: Building Blocks

As you get closer, you begin to realize that each of those blocks in the original piece actually consists of smaller blocks:

 Figure 4b: Building Blocks zooms

Now, what’s really striking about this piece is its actual size…16 feet high by 32 feet wide:

 Figure 4c: Building Blocks with reference

I believe that Chris Jordan’s Running the Numbers series blurs the line between art and statistics. He’s also amazingly passionate about his work.  If you have time, please spend 11 minutes to watch a TED video where he presents some of the elements of Running.

Please realize I’m only expressing my opinion here when I speak of the gravity of these images.  After all, they’re only showing data…or are they?  In this age of ubiquitous, not – always – researched  – nor – refereed  information (and atrocities like CNN “text in your answer” polls), we are inundated with statistics.  Not all are meaningful; I’ll even go out on a limb here and say that most data with which you are confronted on a daily basis in media is biased so badly it’s meaningless.  However, if and when we get good data, I believe it’s our responsibility to provide that data in 1) as correct a manner as possible, and 2) in a way that makes the reader stop and think, not from our viewpoint, but from their own.  I believe the examples above do that, and much, much more.  I hope you get as much out of them as I do.