Skip to content
 

An improved time-series graph instead of that notorious “spiraling down the drain” spiderweb

Jon Peltier saw this horrible graph that I’d discussed earlier:

CXM946.gif

Peltier writes:

Well, this is an eye-catching chart. It seems to show an inward spiral, but the overall trend is really not very clear. It also looks distorted, too tall and not wide enough, but I examined axis settings several times, then even physically measured the lengths of the January-July and April-October spokes, and everything lined up. This optical illusion was caused by plotting the data in a radar chart: April and October were the two largest numbers in 1929, stretching the curve vertically. The first step to improving this chart is to cut it between December and January, and unroll it. . . .

In the chart below, we can easily see the downward trend which started around the time of the stock market crash of October 1929. The trend was well underway even before the Smoot-Hawley Act was enacted in June 1930, because many international trade partners had instituted preemptive retaliatory tariffs of their own. By the middle of 1932, the volume of international trade had effectively plateaued at one-third of its high of 1929:

plughole-timeline.png

That’s what I’m talking about.

P.S. Jason Roos writes:

I think Jon’s graph is far more informative than the original, but it lacks drama. Please see the attached, which I think incorporates the strengths of each.

plughole-timeline-better.png

6 Comments

  1. Jon Peltier says:

    Ha, just goes to show that, as a bad chart can be improved, so can a good chart.

  2. Xi'an says:

    As a coincidence, the December issue of Significance (pp. 181-182) discusses the introduction by Florence Nightingale of such graphs for making a case about more soldiers dying from infection than from wounds during the Crimea war and their shortcomings (area versus radius, superimposed causes of death versus stakes causes, etc.).

  3. Andrew Gelman says:

    Xi'an: See here. I recommended a line plot for that example also.

  4. Xi'an says:

    Thanks, Andrew, I should have checked the earlier link! It is surprising to see how close the paper in Significance is to the one in Science News, down to the quote of Farr.

  5. While I agree that the first graph is a misleading way to present the data, there is one extra bit of information that it conveys over the traditional line plot: the cyclical nature of the decent.

    The usual line plot shows world trade decreasing over time, but the decrease is not monotone. So, while the overall trend is clear, what's missing is that each month in subsequent years, trade was down over the same month in the past year. Or at least that detail doesn't jump out at you as in the spiral plot.

    I wonder if there's a way to convey that information in an obvious way without using an the overall misleading spiral representation.

  6. Barry says:

    You know, I've never seen that graph continued through 1933, to see how much of it recovered how fast. Do you have access to that data set?

    Thanks!