I See What You Mean

The charts over at GraphJam aren't exactly data-heavy or rigorous, but they are often amusing: Graph by weegee64, via the GraphJam builder . Graph by oliver.wolf, via the GraphJam builder . OK, I'm slightly biased here because for years and years I was a paid observer of pop culture (aka journalist) , so naturally I appreciate the GraphJammers' mockery of rock songs and movies. But this stuff pleases me on a professional level too: People who make charts and graphs out of heretofore unchartable (or at least uncharted) cultural artifacts show themselves to be comfortable with graphical renderings. They know how to create them and they know how to read them. And that's nothing but good news for Synoptical Charts and our fellows in the infoviz biz. The more people speak our (visual) language, the more uses they will find for it, and the more they will eventually find themselves relying on it... I hope.

SEED magazine's recent rundown of the pluses and minuses of charting is well worth a read, especially if you haven't spent hours and hours considering which graphical formats are most effective and why. To wit: Many psychoperceptual studies have explored the human mind’s aptitude for gleaning information from pictures. Unfortunately, the pie chart incorporates tasks that we humans systematically fail to perform accurately, all those exercises that come at the bottom of the hierarchy of perceptual tasks... So although we’re good at comparing linear distances along a scale — judging which of two lines is longer, a task used in bar graphs — and we’re even better at judging the position of points along a scale, pie charts don’t bring those skills to bear. They do ask us compare angles, but we tend to underestimate acute angles, overestimate obtuse angles, and take horizontally bisected angles as much larger than their vertical counterparts. The problems worsen when we’re asked to ...

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 ]

John Cole gets his infoviz on and illustrates the actual quantitative difference "between socialism and capitalist nirvana." His bar graph demonstrates that, political rhetoric notwithstanding, the proposed increase for top-bracket taxpayers doesn't exactly warrant the dreaded S word.

Over at American Scientist , Robert Kosara of Eager Eyes weighs in on the difference between SciVis and InfoVis (rivers of blood, people, rivers of blood!): Visualization is often valued for producing pretty pictures for publications. But in scientific disciplines that work in nonspatial realms (bioinformatics, chemistry, the social sciences and so on), visualizing data is useful very early in the process of discovery. Turning numbers into pictures enables scientists to use their human prowess with reading visual data to spot patterns, trends and outliers. There is a historic distinction in the field of visualization between scientific visualization (SciVis) and information visualization (InfoVis). In SciVis, spatial information is almost always a given, coming from measured or simulated three-dimensional objects—photographic images of sorts. In InfoVis, researchers choose the most appropriate and informative layout. He then illustrates some ways in which the farmer and the cowman ca...

I was surprised to read this week that putatively educated Americans ( a Louisiana lawyer, a Colorado dentist , an ABC News reporter ) don't understand the concept of marginal tax rates. Because of this lack of comprehension, per ABC, the lawyer and dentist are vowing to keep their taxable income below $250,000 to avoid President Obama's proposed tax increase: "I've put thought into how to get under $250,000," said [the ill-informed dentist]. "It would mean working fewer days which means having fewer employees, seeing fewer patients and taking time off." Apparently some clarification is called for . Below is a US federal income tax table for 2009 ( source ). This does not mean that if you bring in more than $372,951, every single dollar in your entire pile of money is taxed at 35%. Only Dollar #372,952 (plus whatever additional money you may earn) is taxed at that rate. Dollar #372,950 is taxed at 33%. Meanwhile, Dollar #1 is taxed at 10%. Hence the ter...

San Diego digital artist Tim Schwartz tracked the prevalence of certain words in 158 years of the New York Times and then, rather than create animation or static graphs (too mainstream?), he used that data to power his amusingly retro Command Center (above). Check out those analog gauges and the LED time display! Even the biomorphic shape harks back to a low-tech era. Lovely work, Tim. Playful projects like this give me tremendous hope that soon data visualization will be understood, accepted and even adopted by average folks.