3 min læsning

Don't drown the data - show it!

Featured Image

Far too often, we seem overly worried that our audience will not find our data interesting and entertaining. So what do we do? We dress up the data with a plethora of appealing and salient visual effects. Effects that, sadly, end up distorting and blurring the very essense of the message we were trying to communicate in the first place.

Taking an interest in data visualization, I took this snapshot of a table shown on a television channel that covered the Tour de France event.
The picture shows the current top 10 in the overall standings at the close of one of the stages of this years race. Actually, an extra rider with particular interest to the Danish audience is shown as well, but that is a minor detail in the context of this article.
TdF_TV2

Snapshot of current top 10 in the overall standing, as shown on Danish TV Channel, TV2, Summer 2013.

No matter whether you take a particular interest in cycling, the above table constitutes a good example of what no to do when communicating data visually. The pictured table, however, is also a good example of the potential for improvement to be found in different forms of visualizations we encounter everyday. So let us delve into that.

Color with color on top

The first thing that comes to mind when looking at the table, is the extravagant use of colors. What do these different colors mean? Obviously, the colored rows are meant to aid the viewer in reading across the table. But then, what is the purpose of the lines - do we really need those as well? And why does each column have its own color - again with lines on top? Overall, do the different colors communicate any important information? Not really. Actually, the only colors that do have a meaning are the numbers 1 and 7. Here, the color is used to inform us of the current position of the yellow and white jersey. Unfortunately, this piece of interesting information drowns in a sea of other, unimportant colors.
Peculiar things also happen with the text. Some parts are written with CAPITALS, others with normal letters. Some are marked bold, others ordinary, and several different font sizes are used. Combined with the many colors, the overall expression becomes a bit inconsistent, cluttered and unsettling.

A camouflaged message

A third observation, which ought to have been the first one, is that the message is a bit unclear. It takes a little concentration to make out the yellow and white jersies (granted, the yellow one is easy to find, but that is because we know exactly where to look, and as such, this information is trivial).
As audience, we would most likely ask questions like ' how far apart are the riders?' and ' is the dispersion in the top 10 great or small?' With this table, we have work to do in order to get those answers. We can read both the time of the fastest rider, and the gap down to each of the following riders. But we can only "read" the information, we cannot really "see" it. The mental picture or pattern that our brain needs to see in order to assess the overall information is simply missing. Only numbers are there. That means that our brain must now work overtime comparing the values in order to create its own mental image of the message.

Above all else - show the data!

The good news is that it is very easy to improve a table like this one!
Let's start by removing all visual effects that do not support our perception of the matter or otherwise distracts our attention. The text styles, the heavy grid structure and backgrounds seem like good places to start. Next, let's preserve colors for the elements that communicate important content. This way, the jersies resurface and visually pop out of the pond. Finally, we'll add to our table a simple graphic (a dot plot) to show the pattern that we missed. Now we can see the actual spread of values and properly assess the status of the top 10.
TdF_Makeover

Remake of current top 10 in the overall standing.

What we just did was a rather simple three step operation:
We removed the clutter, applied color sparingly and strategically to attract attention to what is important, and we provided the missing pattern by means of a simple graphic representation of the data. Keeping it simple - data visualization at its core.