Stat Candy

A new site by the Organization for Economic Co-Operation and Development presents statistics in an animated and interactive fashion. Similarly, Gapminder features eye-catching data images that capture everything from the quality of teeth of a country’s citizens to how much oil a country has. After viewing this data candy from, the BBC’s Michael Blastland wonders: is there a point at which sexy design overwhelms the usefulness of the data? [%comments]


the "sexy design" is more likely to intrigue and attract the attention of the masses. Isn't that the premise behind the Freakonomics "movement"?

Fred T.

"Is there a point at which sexy design overwhelms the usefulness of the data?"

When the sexy design collapses the data such that a span of 50 years (1800-1850) is the same width as a span of one year (2006 - 2007), then yes, the sexy design is overwhelming the usefulness of the data.

Jirka Lahvicka

[2] Fred T.

I suppose you are talking about the third link.
Just relabel the horizontal axis with "computing speed (MIPS) per $1000" or "the total amount of stored information" and the scale will be (almost) linear.

Ken Beegle

More worrying is that sexy graphics are likely to obscure bad data and analysis. Professional level graphics and presentation lend credibility to the underlying information and often enhance the conclusions of the author rather than letting the data speak for it's self.

The reality is that we'll see a lot more graphics of data in the coming years because the cost to acquire, analyze and visualize the data is dropping very rapidly. My hope is that we'll become more critical before we accept the conclusions and look at the sources of the data.

David Walbert

@Fred T., agreed about the scale. As designed, it's misleading. I'm equally disturbed by the fact that there is no clear line marking the present (which is essentially the zero on the x-axis) and thus dividing the bare semblance of historical analysis from data that is pure speculation.


is there a point at which sexy design overwhelms the usefulness of the data?

well, both my stats profs certainly thought so. so did my operations profs.


Is that third link even based on actual numbers? Although I think there is a clever truth to it, it does seem like a 'theory' graph rather than a real one.

Tilemahos Efthimiadis

As you said this is "candy". A nice device to lure undergraduate students to do Masters in Economics. Once they enroll in graduate schools, we can unleash the ugliness of raw data.