Appearance

A Scholarly Discussion with Andrew Gelman and Anthony Unwin ​

This is how scholarly exchanges used to work: Scientist A publishes a result, Scientist B then writes an angry letter saying that Scientist A is full of it, to which A responds with more insults, etc., and all that published in a fine scholarly journal. I was recently asked to respond to a piece Andrew Gelman and Anthony Unwin had written about visualization for the Journal of Computational and Graphical Statistics, which had some issues.

In addition to yours truly, the journal's editor Richard Levine also asked Hadley Wickham, Stephen Few, and Paul Murrell for comment. The original article, responses, and Gelman and Unwin's rejoinder are only going to appear in the March 2013 issue of the journal. However, since Gelman has already put both of his articles online and blogged about them, it makes sense to also make my part of the conversation available; in particular since Gelman and Unwin extensively refer to my piece in their rejoinder.

This was an interesting exercise. First, this is a rather unusual way of having a conversation, and reminiscent of how scholarly debates used to work 100 years ago or so (and apparently still do in old-fashioned fields like statistics). It's also a bit like a blog discussion, only much slower. And finally, there was a lot to discuss in Gelman and Unwin's original article.

Without spoiling the fun of reading the original article, they took a number of examples from FlowingData and then generalized from those to all of information visualization. There are many, many statements in that article that just ask to be debunked, and my original draft tediously went through them one by one. I soon realized, however, that I was writing a book rather than an article, and a rather boring one, too. So I ended up writing a short response that mostly points to the big picture of what InfoVis really is, and that gives some examples of the many things they missed.

While the original article is rather infuriating, the rejoinder is a great example of why this kind of conversation is so valuable. Gelman and Unwin respond very thoughtfully to the comments, seem to have a much more accurate view of information visualization than they used to, and make some good points in response.

Here are the links to the articles:

The other responses are not available online yet, from what I can tell. I will link to them if/when I find them, or once the complete set is published in the journal.

Posted by Robert Kosara on November 29, 2012.