Information Visualization vs. Statistical Graphics

Information Visualization vs. Statistical Graphics

Information Visualization vs. Statistical Graphics

Information Visualization shares part of its history and some techniques with statistical graphics. The two fields differ in their approaches though, and in the expectations people have of what they will gain from a visual representation. In two articles, Andrew Gelman and I have written about what we think visualization is, and our points of view could hardly be more different.

Martin Theus invited Andrew Gelman and me to write articles for the Statistical Computing and Graphics Newsletter he is co-editing. The idea was to write about visualization from two perspectives: Gelman from the statistical graphics side, and I from the information visualization (infovis) side.

The resulting articles can be found in the current issue of the newsletter (PDF), and I’m sure you’ll find them interesting. My article is largely based on a critique of one of Gelman’s postings titled That Puzzle Solving Feeling and its accompanying slides, though it’s not actually mentioned in the text. But if you find the veiled references and the last sentence confusing, that is what they are referring to.

Read Also:
Data from 130 million commuters reveal US ‘megaregions’

What has been interesting is not just the debate itself, but also the reaction and comparing the different points of view of two different communities. In response to the articles, Andrew Gelman and Kaiser Fung (author of Junk Charts) have written further postings, and I’m having an interesting email exchange with Stephen Few about the merits of spirals for finding periodicity. The latter will be the topic for another posting, but the Gelman and Fung pieces are a good illustration why we need this kind of debate.

Gelman’s posting essentially rehashes the argument that he made before about how the goal in infovis is drawing the reader in with nice-looking visuals, but not delivering much new information. He calls that the Chris Rock Effect:

I call this the Chris Rock effect. Chris Rock says things we all know are true. But he says it so well that we get a shock of recognition, the joy of relearning what we already know, but hearing it in a new way that makes us think more deeply about all sorts of related topics. Sure, you might have already known that Denver is not near any other large city–but seeing it on this map of phone calls brings this fact to life in a way that maybe never happened in your previous experiences looking at U.S. maps.

Read Also:
Data silos holding back healthcare breakthroughs, outcomes

That is clearly not what information visualization is about. The problem is not that Gelman misrepresents infovis on purpose, he simply has a skewed picture of what it is. Within a few days, he wrote another posting making the same point with different examples.

 



Data Innovation Summit 2017

30
Mar
2017
Data Innovation Summit 2017

30% off with code 7wData

Read Also:
Good models + Bad data = Bad analysis

Big Data Innovation Summit London

30
Mar
2017
Big Data Innovation Summit London

$200 off with code DATA200

Read Also:
Data from 130 million commuters reveal US ‘megaregions’

Enterprise Data World 2017

2
Apr
2017
Enterprise Data World 2017

$200 off with code 7WDATA

Read Also:
Visually Speaking: Patterns for Humane Data Visualization

Data Visualisation Summit San Francisco

19
Apr
2017
Data Visualisation Summit San Francisco

$200 off with code DATA200

Read Also:
5 questions to ask before starting a Big Data project

Chief Analytics Officer Europe

25
Apr
2017
Chief Analytics Officer Europe

15% off with code 7WDCAO17

Read Also:
Good models + Bad data = Bad analysis
Read Also:
Visually Speaking: Patterns for Humane Data Visualization

Leave a Reply

Your email address will not be published. Required fields are marked *