Data visualization is one of those phrases that is frequently heard these days. It’s a very interesting field; done properly, data visualization allows you to use charts, graphs or other visuals to put statistics into context far more easily than if they were in tabular format. The flip side is that if not done properly, data visualizations can be confusing or, even worse, misleading (as illustrated by this chart).
A recent article in the Harvard Business Review by Scott Berinato on “Visualizations that Really Work” talks about how visualizations enable us to use data to make decisions; too much data makes it impossible to pick out what is relevant. When designing data visualizations, Berinato suggests first you ask yourself two questions: first, are you visualizing qualitative information (ideas) or quantitative information (statistics)? Second, what are you trying to do with the visualization: communicate information or use it to figure something out? If you combine the answers to these questions together, you end up with four types of visual communication which Berinato describes as follows:
- idea illustration (communicating an idea)
- idea generation
- visual discovery
- everyday visualizations (communicating statistical information)
The end result should be to “project the idea that you’re showing a reflection of human activity, of things people did to make a line go up and down.” (Berinato, 94)
For librarians, there are two main reasons to use data visualization. The first is to convey information about the management of the library, most usually information related to budget and spending (“everyday visualizations”), and the second is to convey legal information (“idea illustration”).
Data visualization can be very useful to visualize library spending, For example, David Whelan has produced a graph showing the ever growing costs of loose-leafs: https://twitter.com/davidpwhelan/status/461511623350304768. Other places where data visualization can be useful include illustrating usage statistics for databases (e.g. what databases are being used and by whom) and cost recovery.
Another example of where visual representation of data can be very effective in illustrating a point is David Whelan’s chart showing comparative availability of resources at http://ofaolain.com/blog/2014/07/04/law-libraries-and-legal-malpractice/. Not only does the chart show the various resources needed for legal research, but it also highlights their comparative costs and what resources you would be left with if you didn’t have a library.
Given that legal information tends to be about words, rather than statistics, a use for data visualization in legal research may not seem immediately obvious. However there are a number of areas where data visualization can be used to communicate legal information. One example would be a graphical history of a case; WestlawNext Canada includes this as part of its KeyCite function. Data visualization can also be used to illustrate the coming into force provisions of certain pieces of legislation. And you may not need a graphic to understand how the Canadian court system is organized, but what about the British system?
Data visualization is becoming more common in legal databases. For example, when you KeyCite a case in WestlawNext, it uses depth-of-treatment bars to indicate how thoroughly your case is discussed by another case. Lexis Advance uses data visualization to illustrate a user’s research trail.
Although not strictly speaking “data visualization”, colour can be used as a way of adding information. For example, Justis, a British online service, uses colour to indicate when text in a case has been cited by other cases; the darker the colour, the more frequently that particular text has been quoted.
Data visualization, done properly, is a useful tool to summarize or make information clearer. Lawyers have lots of information thrown at them, so the faster information is to comprehend the better, making data visualization a useful tool. However, data visualization is not a solution for all problems; as Nathan Yau notes in his book Data points: Visualization That Means Something, “Sometimes a table is better. […] You don’t need to visualize your data just because you have it.”