UXD: User Experience Design

Pictures that hurt

A picture is worth a thousand words.  But images can also be misleading.  One of the ways that images can mislead is when they are used to simplify real data for quick scanning.  Since they have such a strong impact on us, are so easy to find on a visual plain, and are so easy to remember, content producers sometimes find it useful to replace numeric data with imagery.  However, if they are not selected carefully, there is a risk that they will oversimplify complex or nuanced data, and therefore mislead the intended audience.

Take the weather forecast, for example.  We’ve all seen the images that the appear near the end of the daily weather forecast on the local news, or in the paper—the little pictures of the smiley face on the sun, or the dark cloud with the lightning coming out of it, or the sun with the rain drops juxtaposed on top of it.  Those ubiquitous images are so popular because they serve an important purpose—we’re all in a rush, and we want to be able to get a quick and accurate view of the weather in the next few days. 

Picture_1_2
Since I live in Chicago, the local weather in the summer is crucial. My weekend plans are based entirely on what Saturday and Sunday will look like.  I’ve come to rely on my weather forecast widget (Mac OS 10.4 widgets), which features the same type of images in the familiar 7-day forecast.

I’m noticing, though, that the weather is almost always more, how should I say, nuanced than the little icon on my trusty widget.  I wouldn’t say that the data is incorrect, but the very act of squeezing the projected weather patterns for the day into a 25 x 25 picture is going to involve excessive simplification, even over-simplification.   What happens, though, is that the my brain, which craves simplicity, responds quite well to the weather widget, and the image sticks in my head and gets associated with the day it represents.  I’m conditioning myself to look beyond the widget, because I’m frequently disappointed with the way the weekend turns out (more so lately, what’s up with all the rain?). 

Which brings me back to my original point.  In this instance, real data—not doubt data that exists, like percent chance of rain, humidity level, sunlight consistency, small differences in temperature—are sacrificed for the sake of brevity, and as a result the image is misleading.  In general, where there is a lot of data available, content producers are doing a disservice to their audience by removing it and opting instead for the simple image.

Comments: 5 so far

  1. It is a really hard balancing act to get the right amount of data shown to the user without oversimplifying it into uselessness…

    It’s almost like signal theory where you get aliasing and harmonics if you simplify or sample it at the wrong points…

    Heh, and think yourself lucky! I’ve got rain coming out of the sun today (not sure what that means… The last time I saw that was on the Flash Gordon movie I think), with thunder tomorrow… Think it’s a good weekend for staying in ;)

    Comment by bloid, Friday, July 6, 2007 @ 2:55 pm

  2. Here in the UK, there is http://www.metcheck.com who have obviously thought quite hard about their icons (type eg ‘London’ into the location field on the front page and press Enter to see examples). They don’t always get it right but they seem to have a greater variety to draw on than most sites IMO.

    Comment by Kit, Monday, July 9, 2007 @ 3:45 am

  3. On a somewhat related note, I saw a panel of neurologists on TV recently who cited a study that showed that people believe they understand something to a greater degree when they see an image.

    This is actually dangerous in science, as it can lend credence to an idea that isn’t necessarily true.

    Comment by M Easter, Monday, July 9, 2007 @ 7:16 am

  4. Edward Tufte’s “sparklines” are a great example of how a lot of information can be crammed into a small space without compromising clarity:

    http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001OR

    Comment by Gary, Tuesday, July 10, 2007 @ 8:55 am

  5. Google uses sparklines in their web analytics tool, and Yahoo uses it in their financial pages. Although both with flaws. In Google’s case (google.com/analytics) the data being displayed would actually be more easily scanned and processed with a bar chart.
    In Yahoo’s case (finance.yahoo.com)–and I see this in just about every financial site–the baseline (y=0) isn’t included, instead, the bottom edge of the chart represents the security’s low point for the interval represented on the x axis, and so the stock price looks much more volatile than it actually is.

    Comment by Sholom Sandalow, Saturday, July 21, 2007 @ 1:33 am

Leave a comment

Powered by WP Hashcash

About Pathfinder

  • We design and build extraordinary applications for companies looking to make the next great idea a reality.
  • learn more

Topics

WordPress

Comments about this site: info@pathf.com