Griffin opened the line of speakers with a short talk about why statistical reasoning skills are essential to successful data visualization. He discussed the struggles journalists face in dealing with people who don't trust scientific information and in trying to present a large amount of information in a a simple, informative manner. He referenced a study in which different ways of showing margin of error espoused different levels of trust in the data. It all came down to how much the researchers or journalists were leaving up to the reader to interpret. In this case, the less, the better.
Griffin's talk was interesting because it showed that there is a morality issue in data visualization, just as there is in other parts of reporting. How the journalist chooses to portray their data (with or without a margin of error or a denominator) has an incredible impact on the way people respond to information. Griffin used the example of how the story of two people getting Ebola is hyped up because there is no context to the number "two". Two out of how many thousands? It is true that science and data have a great impact on readers and must be used carefully.
| || |
Horvit spoke next. As a journalist and executive director of Investigative Reporters and Editors (IRE), he had a variety of examples where data visualization had gone very well or poorly (one graphic had percentages that added up to more than 100).
Each of the examples, like the "Ghost Factories" or "Stadium Food" stories each showed a new way in which journalists have presented information that has changed lives. Such work requires large amounts of research and tabulation, and Horvit explained that many times the real newsworthy story is found in the data and not what the journalist had envisioned for the project originally.
Interestingly, when Horvit asked the best data crunchers he knew to send him their best work, the examples he received were very simple and easy to look at rather than crazy points and lines everywhere.
The two tools Horvit suggested for visualizing data were Batchgeo and Tableau, both of which can be used for free.
"Design is more than just picking colors or making it look pretty. Its a way to present problems," she said. Her presentation of people's problems allows them to look at their challenges in a way that is manageable.
As a journalist, my goal is not only to present a problem, but also to inspire a solution. Each of the speakers presented the potential of numbers when used correctly. As Horvit pointed out, in the case of the Milwaukee Journal Sentinel's "Deadly Delays", data is literally saving lives. As a science writer, I will be doing a lot of research and synthesizing large amounts of complex information. I now have the guideline to do so effectively so I can enhance my storytelling skills and my readers' responses to those stories.