How Usable is YOUR World?

I attended a Workshop (at Webstock 2010) called Data Visualization that was very insightful. Wikipedia defines data visualization as “the study of the visual representation of data“. It’s incredible how you can present data in different ways and how it can affect the level of reader’s understanding. In fact, data visualization is used to make a point, to communicate raw / unfamiliar data.

One of Toby Segaran’s projects was to show openings of new WalMart stores from 1962 to 2007 with the first one in Arkansas. It’s quite funny to see how it spread. By watching his video, it’s much more easier to understand the evolution of WalMart in USA than by just looking at a table showing new store locations by year and by city.

He mentioned two tools that can be used freely to make your own data visualization:

• yFiles – The one he prefers.
• Graphviz

I haven’t tried these yet, but I’m looking forward to it. It’s important to be able to translate raw (dull) data into something meaningful to the reader.

He also talked about good and bad visualization, for example, a 3D pie chart is usually not a good idea if you want to compare figures. Let’s see with the following example of a 3D pie chart that shows 5 companies and their market share, which one from A and C have the biggest market share?

The answer is both companies (A and C) have a 20% of the market. In that case, a 2D bar chart might be a better idea if you have to compare these data. Some readers might say that you only have to add percentages to the pie charts, yes it would help but still, visually, there will be a favorable bias toward company C which seems to have a much bigger pie chart.

Presenting data in a way that is easier to understand is important but be careful to not create a bias in your graph.

Filed under:
Conference by Olivier Lorrain