Alberto Cairo: “Data journalism will change this profession for the better”

Alberto Cairo during a conference

Alberto Cairo during a conference

Alberto Cairo is a professor of Information Graphics and Visualization at the School of Communication at the University of Miami, where he is also the director of the Visualization Program at the UM’s Center for Computational Science. He has written two books: Infografía 2.0: Visualización interactiva de información en prensa (Alamut, Spain, 2008) and The Functional Art: an Introduction to Information Graphics and Visualization (PeachPit Press, 2012). In 2000, Cairo led the creation of the Interactive Infographics Department at El Mundo, the second largest newspaper in Spain, and one of the pioneers in embracing data visualization.

Why did you decide to become an expert in data journalism?

My knowledge of data comes mainly from studying on my own a little bit of statistics and the scientific method, but I don’t usually call myself a data journalist. I feel that goes beyond my area of expertise, although somehow my work is related to data journalism, but I’m not so good at using databases, extracting stories from databases or dealing with complex mathematical operations. I know my stats, but I’m not an expert.

I became interested in information graphics and visualization because I studied Journalism in Spain a number of years ago, and then I got an internship at a local newspaper there which was looking for people who had a background in Journalism but who could also draw a little bit. I’m not a great artist, but I know how to sketch ideas, so they hired me and then I got in love with the possibilities of this field, the possibilities of conveying ideas and complex information by means of diagrams, charts, graphs and maps, etc. I am basically selft-trained like almost everybody else in this industry.

How would you say the rise of data journalism in the era of digital media has affected to journalism?

I could prefer to call it data-driven journalism or evidence-based journalism. The latter is the label that I have coined and I am trying to promote lately. Many people would think that this is a contradiction, that all journalism is based on evidence, but I would disagree with that. Many journalists out there do not know what evidence means. They are not able to use statistical methods of analysis, for instance. They do not even know that the scientific method is and that is a huge issue, so even if they say that all their work is based on evidence, that is not actually true. They are being affected by things like cognitive biases or their own political and cultural biases, even if they are not very aware of that, and they are not able to double-check or to put their preconceived ideas under control by looking at data and evidence.

The current popularity of people like [the American statistician] Nate Silver or other famous data journalists is actually extremely positive to journalism, because it is prompting a change in the curricula of journalism schools and it is also driving changes in newsrooms. More and more newsrooms are creating small teams of data journalists, statisticians, and information & visualization designers that will somehow in the future make the whole profession of journalism change for the better. They will increase the awareness of how important statistic and analytical methods are for journalism.

Which are the characteristics that a good data visualization should accomplish?

Any infographic or visualization should be truthful. I am not talking about “the TRUTH”, in all caps. THE TRUTH is unattainable even through the most sophisticated analysis techniques. We communicators should be humble enough to drop the word “truth” and use the word “truthful” instead. Being truthful implies conveying your best understanding of what the truth behind a phenomenon or a data set is. This smaller “truth” is not what you would like the truth to be, it is not what benefits your clients or employers, and it is not what can help you advance your political ideals. It is an evidence-driven, provisional, understanding of the truth.
Second, a great visualization should be functional. This means that the visual shapes you will make your data adopt should not depend on your personal aesthetic preferences alone. Quite the contrary. The choice of shapes to encode your data should be tied to the tasks you predict your audience will try to to undertake using your visualization as a tool to gain knowledge. Function should not dictate form, but function must constraint form, for sure. I believe that many designers still don’t embrace this common sense notion.
Third of all, a great visualization should be beautiful. I know that “beauty” is a slippery, subjective, and context-dependent term, but still, I think that we can all agree that a visualization that looks good, that is well designed, that uses type and color in a visually pleasant way, and that is surprising in some sense, will be more effective and efficient than another one that sticks to, say, Excel’s defaults. The next feature is, in my opinion, one of the most important ones, and it’s one that we visualization designers often overlook.
Great infographics and visualizations are always insightful. That is, they don’t just throw tons of data at you, or present just predictable messages, but force you to discover things that are surprising, unexpected, counterintuitive and, ultimately, useful. And to do that, in most cases they take you by the hand and guide you, rather than leaving you alone to let figure stuff out by yourself.
Finally, data visualizations should be enlightening. At their core, visualization and infographics should always be about increasing understanding about the world, and about changing people’s minds for the better based on an evidence-driven approach to information.
Embracing these ideals is, in my opinion, what will define the future of visualization for communication much more than technological or conceptual innovation.

Which software would you recommend to journalists starting to embrace data journalism skills?

It depends on what you want to focus on. If you are going to focus just on writing , it is obvious that you will need to get acquainted with software related to databases, so learn jQuery, how to use Google Fusion Tables, Access and Excel or their open source alternatives. Those are going to be extremely important. But the tools are not enough. If you take a look at the handbooks about data journalism you will see that they focus too much on the tools, and they do not focus enough, in my opinion, on the concepts. So if you want to get into data journalism you do need to get solid training on statistics. You can be self-trained in stats and quantitative methods of analysis, there are many resources out there, like the massive online courses that you can get at Coursera and tons of books that can give you the foundations that you need. Those are much more important than the software tools.

If you are interested in data visualization or information graphics there are quite a lot of tools to create effective graphs: Google graphic tools are quite good, Tableau Public, programming languages such as Python or Javascript, plus the library called D3. I also discuss tools in my blog regularly.

According to your opinion, which is the most common mistake among data journalists?

News editors are one of the most reactionary species I’ve ever met. They believe that too many of us, visualization designers, are too focused on being creative rather than on being clear. Having worked in newsrooms for many years, I have observed the following attitude many times: “Oh, I like this new, cool graphic form, let’s use it; Oh, I like this typeface or this color palette…”, rather than thinking about if that graphic form, typeface, or color palette fits the purpose of your visualization. The desire to be creative, as respectful and admirable as it might be, doesn’t free us from the obligation of respecting what the human visual brain can or can’t do. Any innovation must happen within certain constrains imposed by human visual perception and cognition. If the goal of the graphic is to let readers compare, shape your data in a way that allows for accurate comparisons.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s