Thursday, July 11, 2013

A nice interactive graphic by The Washington Post

The Washington Post: U.S. life expectancy
Good news graphics tell stories. Excellent news graphics tell stories, prompt you to explore their nuances, give you access to the complexity and the data behind them, and offer clues to put them in context. Excellent news graphics usually avoid gimmicks and hollow creativity. They are elegant, restrained, and also a joy to navigate. Their design enhances understanding, rather than diverting attention; design may even become invisible.

Those were my thoughts when I visited  a recent interactive visualization by The Washington Post's graphics desk (Twitter.) Everything works nicely for me here: The layout, the composition, made of different kinds of charts and maps —choropleth map, time-series graph, a truncated slopegraph on the bottom-right,— the copy, the visual style, and the simple interface.

As I've explained before, one of the main exercises that I ask my students to do at the beginning of each semester is to comment on published infographics based on their efficiency, functionality, design, the insights that they facilitate, and other criteria. This is easy to do if the instructor chooses deeply flawed graphics from an infographics mill, but it's much trickier when the task is to review an excellent project.

So here's some summer homework for you: Explore the interactive, read the story —otherwise, you won't understand what the focus is— and think: a) Are there other ways to display these data? b) Are we missing variables which could make the core messages clearer? c) How should those variables (both the ones included and the ones that we could potentially add) be visually compared and related to each other? I don't have answers to those questions myself —well, perhaps I do, but that's the point of the exercise, isn't it?

1 comment:

  1. In the current status (not the change), I find that there are too few classes. The dividing points betweeen the colors in any color coded information are arbitrary, and with only five classes, we hide a lot of the information lying underneath it. Here's an example:
    Male life expectancy has the same color almost everywhere. Teton County in Wyoming (most northern dark blue county) draws the reader's attention very quickly. Life expectancy there is 80.9 as opposed to e.g Park County (northeast of Teton) with 78.2. Not a big difference, but another color. Fremont (southeast) has 72.8. Quite different, but in exact the same color as Park Country.
    So I'd rather see these maps with a color range that is much more differentiated.
    Berry

    ReplyDelete