Our discussion today stemmed from the Tufte essay on the principles of information graphics.
Misleading and Misrepresenting
“All information graphics reduce.” Since any sort of graphic or interpretation is a designer’s take on the material, it must leave something out (or else it would simply be a regurgitation of the original data). There is a necessary trade-off between accessibility to a layperson and an unbiased presentation.
Since there’s data omitted, it’s easy to misrepresent things, especially to support your own conclusions. For example, temperature was only one of many causes of death in the Minard piece we discussed, but the graphic makes it seem like that’s the only one. Ethics in design is a topic that most designers will come across at some point, and is often a complex and interesting topic.
Even something as simple as Olympic medal counts can be skewed: do you count gold medals or the total medal count?
Mistaking correlation for causation is a real danger when interpreting data, and the unscrupulous designer can easily exploit this to push a conclusion.
Tufte counters this with a call for documentation when creating information graphics. Other elements of the process that can ensure objectivity are checks and balances on information usage, and openly displaying meta information about data quality and source integrity.
Another aspect we talked about in class was Tufte’s demand for putting as much information as possible into a graphic; i.e., adding in as many axes as you can (multiple variables). However, as we discussed, this is another trade-off: multiple axes can add more information, but sometime at the cost of usability.
Omission is a legitimate technique, for both the artistic and utilitarian sides of data presentation. We noted the playwright Harold Pinter’s deft usage of leaving the endings of his plays ambiguous. A more modern example is the movie Doubt which leaves the guilt or lack therof of a character up to the viewer.
Miscellany and Links
- Individual responsibility for graphics are important to ensuring their integrity. When a single person is held accountable for a work, she is more likely to present relatively unbiased conclusions (since it can be traced back to her).
- Content counts most of all, says Tufte. This sentiment is echoed across the internet, often in the form of the credo “Content is king.” Jeffrey Zeldman, a man so involved in shaping the internet that he is often called the “king of web standards”, wrote a very influential article on the fact that no matter how flashy the package is, content is still essential to any graphic or website.
- And this awesome periodic table of visualizations.