Friday, April 27, 2012

How to Present Data, Poorly

I found a 2010 Computer article on "emotional and cognitive overload" (ECO) pretty interesting, at least in part because this is a sort of overload I often feel. The authors point out that one source of ECO is the need to deal with ever-changing web interfaces, best exemplified in my recent experience with the nonsensical changes to Gmail, Google Docs, Blogger, etc., though Gnome is another good example.

I read the article with great interest until the authors started presenting their data. One of their graphs appears to the right. It purports to show the effects of ECO as a function of age. It appears to actually show that the prevalance of feeling of being overloaded and of not being overloaded decrease and then increase together. This seems odd, but the figure is not explained in the text. So then the obvious question is what the units to the left are. This is also not explained in the article. So it appears that they tossed a meaningless graphic in that neither supports nor contradicts their contention that "both young and old suffer from ECO with information technology."

I am not saying the authors are wrong, but rather that this glitch in their presentation should never have gotten past the reviewers or the editors. Computer is usually better than this.

The article is A. Rutkowski and C. Saunders, "Growing Pains with Information Overload," Computer, June 2010.

No comments: