I See What You Mean

At his blog Junk Charts , statistics jock Kaiser Fung cleans up the clunkers. And very well too , I might add. Here's a revamp he did of an eyecatching-but-not-very-useful graphic depicting Americans' changes in religious affiliation (data via Pew ). I like his clever format, with the arrows pointing in all directions. However, at some point a data set is too small to be worth visualizing, per Tufte, and should be displayed as a table instead. Here, ET does a rethink of a statistical table that molds the numbers into an elegant and useful hybrid, a "table-graphic." I wondered if a hybridized display would work well for the Pew data set, so I gave it a try. Usefully simple, or too simplistic? Comments welcomed.

Along with Fast Company (and me ), Business Week asserts that good information graphics and info design have business value. An admiring profile of Edward Tufte — with accompanying slide show , natch — is featured in BW's recent "Voices of Innovation" package: Next to a bad example of a graph, he positions a sublimely clear treatment, often using the same data. Simple as it sounds, the effect has proved to be riveting for a generation of nonprofessional designers. Tufte's work is relevant to anyone who needs to write or present information clearly, from business executives to students.   In dismantling some of the worst habits of two-dimensional design, he has framed new analytical terms that flicker through many design conservations [sic]*. * Conventions? Conversations? Or something else? And, more important, have any readers out there recently had a "design conversation" at work? (Media folks, you're DQ'd, sorry.) Do tell.

Back in 1974, Yale poli-sci professor Edward Tufte published a slim volume called Data Analysis for Politics and Policy (Prentice-Hall, $3.95). The book in its entirety is available for free download (PDFs) at Tufte's website, accompanied by a contemporary review from the Journal of the American Statistical Association . More than 30 years later, the review amuses me with its restrained praise of the perspective that would eventually make Tufte a Major Figure (and a minor fortune ): Tufte puts residual plots to good use to gain understanding of a data set, and he shows how finding outliers gives the analyst hints about the inadequacy of a statistical model... The discussion of graphical techniques in general is quite good... A brief but compelling discussion of the "value of data as evidence ," with regard to the interpretation of nonrandom samples, is presented. If you happen to have a spare 48MB lying about, DAPP 's worth a download. [via Sofa Papa ]

Fast Company 's Cliff Kuang tells his readers that information graphics are all the rage among young graphic designers these days, and for good reason: The discipline—involving cartographers, statisticians, computer programmers, and graphic designers—creates public debate on a level graphic design rarely does. As Manuel Lima, the founder of the information graphics website Visual Complexity , once told me, "One of the big challenges we face now is dealing with all the data around us, and finding ways to make it useful." "Visualizing the Economic Stimulus" — Fast Company

Overnewsed but uninformed-- Stefan Bräutigam The explanation, auto-translated from the German by Google (!) and tidied up slightly by me (!!), reads thus: With access to news possible at any time via a variety of communication channels, a broad media landscape with a myriad of information producers and suppliers has developed. For the viewer it becomes more difficult to determine the authenticity of messages. You are looking for sources that can be trusted. Overnewsed but Uninformed helps in the search. Sequences are described, ownership and dependencies disclosed, and user behavior illustrated. Background information, monitoring and analysis show how the real message — of a bridge fall in Minneapolis [in this case] — can be assessed and categorized. Unfortunately, if you want to see the whole thing up close, you'll have to download the 22MB PDF file here . [Via Visual Complexity ]

As someone who deeply appreciates raw data and admires those who wrangle it, I have to give a shoutout to my fellow travelers out there: Big bunches of ripe bananas to the primates over at Infochimps , "a community to assemble and interconnect a giant free almanac, with tables on everything you can put in a table—things like a century of hourly weather, every major league baseball game, decades of stock prices, or every US patent filing." Check out the (still small but interesting) visualization gallery . Numbrary , whose name is self-explanatory, is "a free online service dedicated to finding, using and sharing numbers on the web." At present its focus seems to be financial and demographic, with info from the U.S. Bureau of Labor Statistics , Board of Governors of the Federal Reserve System , U.S. Bureau of Economic Analysis , U.S. Census Bureau , the SEC EDGAR Database and CIA - The World Factbook . Swivel , which aims to "make it easy for eve

Well, first of all, he's my hero, and has been since I first attended one of his seminars back in 1987. More to the point, I'm far from alone in my appreciation of him , and here's why: I described Edward Tufte as a graphic designer, but that’s not exactly right. His field is almost sui generis, containing bits and pieces of art direction, data-crunching, economics, historical research, and plain old expository writing. It’s often labeled “information architecture,” or “analytic design.” Tufte himself describes it many ways, but one is drawn from a classic piece of science writing: “escaping Flatland,” or using paper’s two dimensions to convey several more. Another, more acidic description: “getting design out of fashion and out of the hands of Microsoft.” His four books have collectively been called a Strunk and White for design. Tufte works by showing both outstanding and horrid graphics he’s found, improving upon the latter, and his principles take on the meditative qual

Interesting how you can lie with statistics even when you set out to tell the truth. Viveka Weiley at Karmanaut has redrawn a recent Washington Post chart purporting to show how Obama's tax plan compares to McCain's. As she notes, the WP original is not to scale; all nine tax brackets are drawn the same size, thereby suggesting that the number of people in each of tax bracket is roughly equivalent, and that there'd be just as many people paying the top marginal rate as there would be paying the lowest rate. In fact, that top echelon (annual income over $2.87 million) comprises only one-tenth of 1% of the population. ( source ) P.S. Weiley herself has no dog in this bloody political fight: She's based in Sydney, Australia.