“Ben / Met Anna / Made a hit / Neglected beard / Ben-Anna split / Burma-Shave”
What, you may ask, do charts have to do with some now defunct shaving cream?
When I was younger (much younger), a great era was just coming to a close. From 1925 until 1963 the Burma Vita company installed a series of small billboards along a roadside spaced for sequential reading by passing motorists, almost always ending with the name of their flagship product Burma-Shave. Over the years, more than 600 unique slogans were developed, but they were always delivered in the same format. The success of the advertising campaign was largely the result of two factors. First, the company chose a format that was well suited to the relatively new passion for the automobile in America. Second, the slogans were designed to draw in passers-by until they could get to the punch line. (For more information on the Burma-Shave ad campaign, visit http://www.eisnermuseum.org/_burma_shave/page1.html.)
If you are using modern software to visualize a large amount of data, it may be wise to keep this old time sloganeering approach in mind. First, the format of a chart should be well chosen to match the data and the audience. Certain types of charts are better suited for different types of data. Bar or column charts are suited for value comparisons across a number of major categories, whereas line or scatter plots are better suited to compare collections of values with independent values in 2 or more dimensions. Pie charts are great for showing category values relative to (as a percent of) the sum of all the categories chosen.
Second, a good chart should draw its viewers in, encouraging them to linger at least until they get to the chart equivalent of the punch line (the “chart line,” perhaps?). Jacques Bertin, one of the founding fathers of modern graphical analysis, says that knowledge is built by the user when interacting with a chart.
There are two main issues to keep in mind here: one is to find the right tool for the job, and the second is to find the balance between the need to correctly display data and the emotional response that helps to keep the audience interested.
With that said, look over these tips, traps, and misconceptions about charts and data visualization from Jorge Camoes’ excellent website “Charts”:
http://charts.jorgecamoes.com/misconceptions-charts-graphs/
And remember,
“A peach / Looks good / With lots of fuzz / But man’s no peach / And never was / Burma-Shave”
Professor Addenhauer
