Acquire – Obtain the data, whether from a file on a disk or a source over a network.
Parse – Provide some structure for the data’s meaning, and order it into categories.
Filter – Remove all but the data of interest.
Mine – Apply methods from statistics or data mining as a way to discern patterns or place the data in mathematical context.
HOW TO LIE WITH STATISTICS by Darrell Huff – My mom gave me this book in high school and I still find it useful. It covers the basics of statistics as well as common ways people manipulate charts, graphs, and numbers in order to disguise data and mislead their audience. Some of the key takeaways include some simple (but still ignored!) concepts:
– Correlation does not imply causation
– Use random sampling
– Don’t truncate the scale of a line or bar chart
– Don’t cherry pick data in a time series
– Don’t represent one-dimensional quantities with two- or three-dimensional objects
Buy it. Read it. Send a copy over to the folks at Fox News.
Represent – Choose a basic visual model, such as a bar graph, list, or tree.
THE VISUAL DISPLAY OF QUANTITATIVE INFORMATION, by Edward Tufte – There’s a reason why Edward Tufte is regarded as one of the pioneers of modern information design and this is the book that started it all. It includes examples of the best and worst data visualizations in history and introduces concepts like data ink, data density, and chartjunk, a term Tufte invented to describe unnecessary graphic elements that obscure the interpretation of data. The book also helps explain the thinking behind graphic design decisions and introduces the reader to several unique charts and diagrams. The cover illustration even inspired one of my own creations.
INFORMATION DASHBOARD DESIGN: THE EFFECTIVE VISUAL COMMUNICATION OF DATA, by Stephen Few – Through his blog, Perceptual Edge, Stephen Few has long championed a clear and simple approach to the display of information. With this book, Stephen Few helps bring good dashboard design to the masses … along with great arguments against the use of pie charts, word clouds and other gimmicks.
Refine – Improve the basic representation to make it clearer and more visually engaging.
RICHARD SCARRY’S CARS AND TRUCKS AND THINGS THAT GO, by Richard Scarry – What better way to learn how to engage your audience than by studying a few of the most compelling children’s stories ever written? Author/illustrator Richard Scarry created some of my favorites, including Cars and Trucks and Things That Go, What Do People do All Day?, and Best Word Book Ever. More than just linear tales of Lowly Worm and Huckle the Cat, Scarry’s books are rich visualizations that invite exploration and repeat visits.
Interact – Add methods for manipulating the data or controlling what features are visible.
THE DESIGN OF EVERYDAY THINGS, by Donald Norman – My mom checked this book out of the library once when I was a kid and it ended up having a profound effect on me. The book’s main premise is that users of computer programs or equipment (or charts) should be able to figure out how something works through design clues and logical mental models (“knowledge in the world”) instead of having to remember fussy steps or obscure details (“knowledge in the head”). In other words, good design should be intuitive, particularly for critical items or objects that people don’t use every day. This concept places a higher burden on the designer, who must anticipate problems and ensure that any system operates “as expected.” It also explains why user experience (UX) is one of the biggest challenges facing software developers today.