By Tim Chartier, Davidson College

Our world produces a stunning amount of data and each year we produce data at an increasing rate. In 2010, the world is estimated to have created 2 zettabytes of data, which grew to approximately 64 zettabytes by 2020. If each gigabyte in one zettabyte were a brick, 258 Great Walls of China could be built, which could reach from the Earth to the moon and back 7 times. Visualization is an important part of working and communicating with such data.

A leader in the field of visualization is Alberto Cairo, a journalist and designer, and the Knight Chair in Visual Journalism at the School of Communication of the University of Miami (UM). He is also the director of the visualization program at UM’s Center for Computational Science. Cairo consults with companies and institutions like Google and the Congressional Budget Office. He’s also authored the books “How Charts Lie: Getting Smarter About Visual Information,” “The Truthful Art: data, charts, and maps for communication,” and “The Functional Art: an introduction to information graphics and visualization.”

Here is an example of his work:

Cairo has also been featured in The New York Times with articles such as this one: “Those Hurricane Maps Don’t Mean What You Think They Mean”.

Tim Chartier: What advice do you have for being a careful, discerning viewer of data visualizations?

Alberto Cairo: The first thing to consider is that we should never assume as readers of visualizations that we understand the visualization just by quickly looking at it. It is necessary to approach a visualization as if it were a piece of text. You cannot assume that you understand text if you don't read it. In the same way, you cannot assume that you understand a visualization if you don't read it carefully. So never assume that a visualization is just an image.

Visualizations are an argument made visual. It's also important to remember that the visualization is usually not the end of the story. A visualization is usually part of a broader argument that involves the data and sometimes other visualizations. So it is also important to always look beyond the visualization. Sometimes we can access the original data that the graphic is based on or we can look for other sources. This advice essentially boils down to: be attentive. Don't just quickly look at a visualization.

Read it. What is pressing in the visualization? What is absent from the visualization?

Chartier: Suppose I have data and want to visualize it, what tips do you have for moving from data to a visualization?

Cairo: Whenever you deal with a dataset and you're about to transform it into some sort of graphic or map, always take the purpose of the visualization into account. What is it that you want to communicate with the visualization? More importantly, what do you want people to be able to do with that visualization? For instance, as a very basic example, if you want people to be able to compare certain numbers to each other, usually a bar graph is great. On the other hand, if you want to see the change of a variable over time, a line graph may be better. If you want to see spatial patterns in the dataset then usually a map is best. These are very simple examples, but there are many, many other types of visualizations that we could consider. Each one of those types of graphic forms are appropriate or not appropriate depending on the task and purpose of the visualization. Keeping the purpose of a visualization in mind is the most important thing.

Chartier: What tools do you recommend for producing data visualizations?

Cairo: There are many nowadays from off-the-shelf tools like Excel.  There are many other tools out there, too. There are tools to create maps from Geographic Information Systems (GIS) data. Some are free like Quantum GIS. There are programming languages that you can use to create visualizations. Python and R programming languages have many visualization libraries.

There is a wealth of tools nowadays. A simple Google search will reveal many tools that we can use. And how do we choose which tool to use?  Most important to remember is that the tool is not the most important component of the data visualization process.  It is the thinking behind the visualization that really matters.

Chartier: Suppose I’ve made my data visualization. What tips do you have for gauging its effectiveness and for critiquing my own work?

Cairo: Show it to other people. That's always the best test to assess whether a visualization is working or not. List the things that you want to communicate with the visualization. Then, show the visualization to other people and don't tell them what the visualization is about. Have them read the visualization and then get their responses and reactions.

Chartier: You wrote an entire book on “How Charts Lie.” How does one take care in creating an infographic or data visualization so we don’t unintentionally lie in our own work?

Cairo: This involves a lot of different factors. The most important one is really understanding the information to visualize. We must try to be ethical in our representation of the data and try to represent the data in a way that we would like it to be represented to us. We must also try to visualize our best understanding of the messages that the data may hide.

Chartier: There is a lot of interest in colleges and universities in data analytics. There is also a lot of interest in students to learn such skills. What advice do you have about teaching and learning data visualization?

Cairo: Obviously, reading books that introduce data visualization can take you a long way. Join an organization which connects you with other people. For example, the Data Visualization Society is great. It is a great place to have discussions, get recommendations of software, and ask questions to people who are already in the field.

But nothing really substitutes for practice. Pose challenges to yourself. Get data sets you would like to see represented graphically and create those graphics. Put them out there so others can critique them. Let the world see your work and get responses from the world. Learn by doing things.

I also believe in learning by copying other people, not in the sense of plagiarizing anybody.  Expose yourself to the work of as many people as possible, and then borrow ideas from those people.

Tim Chartier is the 2021-22 Distinguished Visiting Professor for the Public Dissemination of Mathematics for the National Museum of Mathematics. Dr. Chartier works in data analytics, specializing in the area of sports analytics.