Psychology of Data Visualization

This is the continuation of the transcript of DM Radio’s program titled “What You See Is What You ‘Get’ – How Data Visualization Conveys Insight,” recorded in September, 2012.

Eric Kavanagh:  Okay folks back here at DM Radio and we have our good friend and colleague Dr. Robin Bloor calling in from Austin, Texas.  Welcome back to the show.

Dr. Robin Bloor:  It’s a pleasure to be here.

Eric Kavanagh:  Sure.  So what do you know about data visualization and maybe the psychology of data visualization?

Dr. Robin Bloor:  Well, I think that's stuff that hasn’t been touched particularly strongly so far. I thought it would probably be a good idea to try.  There’s one problem or one aspect of data visualization that leads to a problem and that is simple education.

With a given visualization, any individual ought to be able to deduce those things. It’s certainly the case that certain visualizations make a lot of things obvious. But the marketing world has been doing data visualization for a long time and normally they were attempting to deceive rather than to inform in terms of the stuff they presented.

construction dashboard

gallery icon
View the Example Gallery of Dashboards and Visualizations.

So there are a lot of arrows that can be made intentionally or unintentionally. It's a fact for instance that 3D visualizations look much better than 2D visualizations. But if you go for bar charts and pie charts, it’s actually easier for the human brain to get information from 2D than 3D visualization.

chart art

Things like that are kind of interesting. I saw some really good stuff on one of the presentations which was animated based on visualization.  It's really, really good but you can't come and stop it and drill down. The point is that various forms of data visualization have very specific context and they carry only a certain amount of information.

And, remember the other thing is that data visualization is now going to the point where, on the one hand, we’re using visualizations in order to better understand data; but, on the other hand, we have a lot of data visualization with poor communication.

And just because I understand a particular way of looking at data doesn't mean the guy that I send it to is going to understand that.  I think there’s an awful lot of work to be done in for people to properly understand what data visualizations options are, what they mean, what you missed, what you get.

Tables are great but as soon as you have several million rows of tables it’s not so great. You pretty much have to visualize it to get any idea of the shape of the data unless you are going to do something simple, such as, find the outliers by sorting it.  You are pretty much lost as soon as you are coming to really big samples of data.

dashboard demo
View a 2-minute demonstration of InetSoft's easy, agile, and robust BI software.