Data Visualization doesn't need to be biased

At the FlowingData blog, data visualization commentator and Visualize This author Nathan Yau lists 5 misconceptions about visualization:

• Software does everything (Nathan notes "Personally, I use a lot of R and have a lot of fun in Illustrator", but uses a lot of other tools as well.)
• Visualization is for making data flashy
• The more information in a single graphic, the better
• It has to be exact
• Visualization is too biased to be useful

I agree completely with Nathan's comments on the last point above:

There's a certain amount of subjectivity that goes into any visualization as you choose what data to show and how to show it. By focusing on one part of the data, you might inadvertently obscure another. However, if you're careful, get to know the data that you're dealing with, and stay true to what's there, then it should be easier to overcome bias.

After all, statistics is somewhat subjective, too. You choose what you analyze, what methods to use, and pick what to point out in reports.

News organizations, for example, have to do this all the time. They get a dataset, decide what story they want to tell (or find what story the data has to tell). Browse through graphics by The New York Times, and you can see how you can add a layer of information that objectively describes what the data is about.

This stands in contrast to the presentation I saw today at the Strata conference from Alex Lundry, Chart Wars: The Political Power of Data Visualization. (You can see a shorter version of his talk online). It was an entertaining talk, but his main point was to encourage data visualization partitioners to actively insert a point of view into the presentation of data. For example, he encourages more charts like the one on the right below, rather than the one on the left.

(Images from Nigel Holmes' the paper, Useful Junk? The Effects of Visual Embellishment on Comprehension and Memorability of Charts by Scott Bateman et al.)

Lundry's take is that because the image on the right is more easily recalled by those who have seen it, it's naturally better. I disagree. My objection to the chart on the right isn't just that uses chartjunk, nor that the teeth are disporoportionately sized to the values, nor even that the X "axis" is slanted upwards to exaggerate the rise. My objection is the chart on the right is that it actively pushes an analysis upon the viewer. As Nathan notes, there's always an element of bias in what data is selected to be presented, and the way it's presented. But good charts merely present data, and leave the analysis (obvious though it may be) to the viewer. When a chart takes on the burden of analysis for the viewer, that's when it strays from data visualization into propaganda.

Update Sep 26: Corrected attribution of paper with images above.

FlowingData: 5 misconceptions about visualization 

Comments

[Alex Lundry]

Thanks for the comments Dave!

I think the key point of the talk I'd like your readers to understand is that we have to start parsing our ideas, attitudes and best practices about data visualization between dataviz for analytical purposes and dataviz for communicative purposes. I think that analytical visualizations can and should strive for objectivity, but if my goal is to communicate a piece of information efficiently in a crowded information space, then we should use the tools of design to increase the visualization's effectiveness.

So long as communicative visualizations are done transparently and ethically, I see no reason not to encourage the use of visual flourishes.

Visualizations that purposefully obfuscate or are intentionally dishonest should be derided for what they are: propoganda. But a visualization adorned with meaningful and transparent visual flourishes is, in my opinion, okay. Rhetoric wrapped words have given us some of our most uplifting political speeches...why can't we do the same with our visualizations?

[Carlisle Rainey]

David and Alex,

I am sympathetic to both of your positions. I think some of the disagreement lies in what you want your graphics to do. Alex wants graphs to make a point in an authoritative fashion, such as how an expert tells non-experts how something works. David wants graphs to allows readers to see the interesting relationships for themselves. When I make graphs, I definitely follow David's approach.

I don't care if graphs push analyses on readers. Plotting relevant comparisons is itself pushing an analysis on the reader, although to a lesser degree than illustrated above.

My objection to the chart on the right above is that it pushes an analysis

at the expense of clear comparisons

. This is a property of the chartjunk though, not of the intent of the creator. Whether chartjunk is intended to promote a conclusion or simply make a graph prettier, it is counterproductive to my goals when it gets in the way of clear comparisons. More here http://wp.me/p1RDVT-4S

[Robert Young]

Well, never forget: "figures don't lie, but liars figure". In the politically charged atmosphere of The Great Recession, being straightforward will be viewed as self-inflicted punishment. IOW, Social Darwinism takes the lead.

[Rick Wicklin]

"Good charts ...present data, and leave the analysis ... to the viewer." Well said. Furthermore, there is a great advantage to the reader to see something that looks familiar. We, as analysts, should not force the reader to figure out some new graphic if a simple, familiar graphic suffices. To do so is to put the spotlight on ourselves ("Look at how clever I am!") rather than on the data.

[Cody L. Custis]

Point of order. The image on the right is originally by Nigel Holmes, but the paper is about Nigel Holmes by authors at the University of Saskatewan.

[David Smith]

Thanks Cody and others for pointing out that paper wasn't by Nigel Holmes. I've corrected the post above.