[jimt67]

Given the vast amount of junky infographics, I generally agree with their philosophy. They do, however, seem to ignore some of the research on the value of chartjunk. Like this paper, which won a best paper award at CHI '10: http://hci.usask.ca/publications/view.php?id=173

[cscheid]

And then there's this, about that paper: http://junkcharts.typepad.com/junk_charts/2010/05/8-red-flag...

Edit: And this http://junkcharts.typepad.com/numbersruleyourworld/2010/05/m...

[jimt67]

There are other studies (e.g., http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=6327... ) A Google Scholar search will turn up others...the HCI literature seems to prefer the term visual embellishments to chartjunk.

[danso]

I'm inclined to believe their findings, in the same way that I'd agree putting a picture of Miley Cyrus as the lead art for a story about unemployment rates would make you notice and remember that one story more than other unemployment rate stories.

Quick question on the paper's methodology...is 20 a big enough sample size?

[jimt67]

It may or it may not be...see: http://www.uvm.edu/~bbeckage/Teaching/DataAnalysis/AssignedP...