Progress and retrogress
Statistical science fiction

Cram it like Koby

You have to gradually build up your gut by eating larger and larger amounts of food, and then be sure to work it all off so body fat doesn't put a squeeze on the expansion of your stomach in competition  -- Takeru Kobayashi, six-time champion of the Coney Island hot dog eating contest

Kobayashi is a phenom.  He can stuff 60 hot dogs or 100 burgers in ten or twelve minutes and show no consequences.  Ordinary people can't hope to emulate these feats.

Junk Charts sees Kobayashi as a hero; an anti-hero really.  We are ordinary people; we can't hope to cram it like Koby.  A message we keep repeating here is: too much data sinks a chart.

Econ_anglosaxon Not long after this chart showed up in the Economist, several readers urged us to take a look.  It's a well-nourished chart indeed, one to challenge Kobayashi, but for all that it contains, the reader has to try very hard to find insights.  What with the multiple colors, iron-fisted gridlines, above-and-below boxes, dotted and solid lines, and a legend with nine pieces split in two spots?  Besides, the U.S. boxes grab all the attention by virtue of them being wider (country being more partisan).

The key to unraveling this chart is to identify the relevant comparisons:

  • UK average vs US average
  • UK left vs US left
  • UK right vs US right
  • UK independent vs US independent

And then for the gluttonous:

  • UK right vs US left
  • UK left vs independent vs right
  • US left vs independent vs right

In the junkchart version, we address these comparisons sequentially.

Redo_anglosaxon1a
(Apologies for the tiny font.)

We are again using a small multiples approach that places four comparisons next to each other: average, left, independent, right. Consistently, the British is to the left of Americans.  The only places where the two cultures meet are where liberals agree on "ideology" and "military action".

Also note that we use a symmetric horizontal scale centered at 0.  There are too many charts out there where the center is not at the center!

A similar presentation addresses the other three comparisons.  Democrats in the U.S. are miles to the right of Tories in terms of "religion".  In the UK, Labor and Tories are not much different except on "ideology".  In the US, Independents lean closer to Democrats.

Redo_anglosaxon2a

Joining the lines (I hear the grumbles) helps bring out the gap between the groups being compared.  Without lines, the chart would look like this.

Redo_anglosaxon3a

It is often hard to keep track of which dot is which as they trade order from issue to issue.

PS. Anyone knows what is being measured on the horizontal axis?  The original graph mysteriously stated "respondents' views".


References: 

Eric Talmadge: "Pigout champion Kobayashi limbers up for hot dog gold" June 25, 2004

"Anglo-Saxon Attitudes", Economist, Mar 27 2008.

Comments

Jon Peltier

'PS. Anyone knows what is being measured on the horizontal axis? The original graph mysteriously stated "respondents' views".'

I spent a while looking at the original chart and the redrawn charts trying to figure out what was being measured. It wasn't until later I noticed your question, and realized that you don't know either. I went to the Economist article, and it wasn't too clear either:

'The data are derived by subtracting left-wing answers from right-wing ones, for each country and for each main political grouping within each country. A net minus rating suggests predominantly left-wing views and a positive rating suggests a preponderance of right-wing views.'

Clear as mud. What are left- and right-wing answers? Sometimes it's easy, like 'Do you believe there is a God?', which has either a Yes or No answer. Sometimes it's subject to more interpretation, or at least interpolation, like 'The profit motive is the best spur to job-creation', which has a spectrum of responses between strongly agree and strongly disagree.

Keith Watson

Setting aside Jon Peltier's important point about the concept of what the heck is being measured on the horizontal axis, the junk charts version is a great improvement. I like the version with lines because it allows my eye to follow which group is which as I scan vertically.

Stephen Hampshire

While the Junkcharts version is much clearer, I do feel you've lost the representation of the size of the gap between "left" and "right".

The Economist chart could be clarified, and would be very good with 2 parties. Representing all 3 is what makes it hard work I think.

Alan De Smet

What's with the lines in the junk art version? Lines between points suggests a continuum, which this obviously isn't; it's six distinct points. For example, it's not meaningful to examine the point between Ideology and Military action. If it's just there to help identify related points, surely there is a better way?

Alan De Smet

Sorry, obviously ignore the above. I didn't see the final notes in the post when I made my point. Perhaps it was added after I composed it, and then accidentally let it sit for a while. Or perhaps I just failed and overlooked it.

Andreas

One information that the junkchart versions do not convey as clear - I think - is that the US voteres are much more divided then the UK voter. So maybe the observation that
"the U.S. boxes grab all the attention by virtue of them being wider (country being more partisan)."
is the whole point?

Kaiser

Andreas: you pretty much summed up the issue in different words! If the point was to show the difference between US and UK spreads, then all the extra information about 3 parties and 2 averages only serves to confuse things. The horizontal bars highlight the differences at the level of issues; they don't provide a good overview of the average difference.

Bob

It always strikes me as odd how much white space the junk charts "improved" graphs have. Very little of the space available is used to represent the relevant information, leaving the graphs looking undernourished, straining the eyes looking for clear conclusions from emaciated lines. The other improvements seem great as a general rule -- but only after working harder than seems necessary to find them.

Kaiser

Bob: good point. I think the fear of white space is what leads to a lot of cluttered, confusing charts. This is no different from creating slide presentations: a common advice is to focus on one key idea on each slide, and not to cram the slide with text.

Bob

Cramming it with text certainly won't help. Making the lines thick enough to make the main points easily visible, using color conservatively to differentiate conditions... this can make it easier for the eye to catch the key idea without any added clutter.

J

No offense, but I tend to agree with Bob's points above. The Economist chart is not great, but the redesign doesn't seem like much of an improvement.

The Junk Charts version would take up about three times the space of the original chart, but it removes information (the country ranges are lost in the small multiples, so it's no longer visually obvious that the U.S. is more partisan), abbreviates the labels, crams them in so that they run into each other, and adds meaningless connecting lines. Beyond those serious technical problems, the charts look completely unfinished, as if they were built in Excel in about ten minutes.

The redesigned charts would look like a printing error if they were to actually run in the Economist, so it's not really clear what the point of the exercise is. Assuming you were actually given three times the space in the magazine, is this really your ideal redesign, what you'd like to see on the page? Like many of the charts on this site, the redesign seems to completely ignore the space limitations of the printed page, along with legibility, color and typography, all of which are essential to conveying information. The only thing the redesign does seem to show clearly is a strong preference for scatterplots.

Kaiser

J: you are entitled to your opinion. For me, I do not believe that the purpose of the original chart was to show that "the US is more partisan". That is what our eyes pick up; it only shows the chart's weakness.

In fact, the title of the piece was "Anglo-Saxon Attitudes" and the subtitle was "Our polls show the two may have less in common than they think."

Further, the solution to limited space is not to cram more data onto the chart. The solution is to simplify the message, clean up the chart so that the message stands out.

J

Yes, I agree that's the solution. My point is that for once it would be nice to see this blog actually try to make a chart that is complete, that has been cleaned up, that has been executed well, and that does have a legible message that stands out without paragraphs of text next to it explaining what the message is -- in other words, a redesigned chart that could actually stand in the place of the original in a real publication. That would be nice to see, but I have yet to see it on this blog.

The comments to this entry are closed.