By Mark Schroeder
Throughout human history, stories have helped us make sense of sequences of events in our lives, infer cause and effect relationships, and share them with others. Just as our own memories are fallible and retelling stories can shape how we remember events, data can be fallible too. Its value is shaped by the process used to collect it and can be incomplete, incorrect, or biased in some fashion. How can we use data to gain true insights about the world and share them despite these challenges?
Data Driven Storytelling: a Deep Dive into Visualization Techniques, a workshop held at the headquarters of the American Association for the Advancement of Science (AAAS) in Washington, DC, on July 14th, 2017, brought together academics, federal staff, policy analysts, journalists, and science communicators to answer this question using their skills and experiences telling stories with data. The Big Data Affinity Group of the AAAS Science and Technology Policy Fellowships, the South Big Data Innovation Hub, and the West Big Data Innovation Hub hosted the event with support from the National Consortium for Data Science and the Renaissance Computing Institute (RENCI) at the University of North Carolina at Chapel Hill.
Multimedia has expanded the set of tools we have to convey experience, and pictures and video often come close to sharing an actual experience. Increasingly, data has become an important tool used to record experiences and convey meaning, allowing people to go beyond anecdotes and provide context that communicates how typical or remarkable an event is. Charts and other types of visualizations provide the potential to summarize a range of experience and can be an effective way to make a point or share an insight. However, charts, graphs, videos and other formats we use to tell stories with data must tie back to the experiences of individuals and the emotions these experiences invoke to make this data meaningful and relevant to the audience.
In the morning, we learned that by integrating data into the decision-making process, it competes with other forms of information and must be accurate and usable for the purpose. For example, car collision data recorded at the nearest intersection does not help us understand which sections of roads are the most dangerous so we can improve them. Data can seem objective, but we must understand its limitations so our decisions are not affected by biases, missing data, or other inaccuracies. When dealing with large and complex data sets, it can be more effective to use simple color schemes and consistent designs across a series of visualizations to effectively communicate key points to the audience. In the early afternoon people had a chance to learn how to visualize data by taking a training in either D3 or Tableau.
In the late afternoon we learned the key components of storytelling: A story takes the audience on a journey that reveals insights about people, places, or their view of the world. There are different models of what a story is: a news story, the epic journey of a hero, or a recounting of a sequence of events. Words are an extremely powerful medium for summarizing the message. Pictures can quickly draw attention and connect the audience to an experience. Video draws on both sight and sound to present a sequence that can hold a whole story. Stories must be relatable and engage emotion to capture the audience’s interest and make them care about the points you intend to make. Importantly, a story can leave room for interpretation so the audience can draw their own insights based on how they relate to the content.
The key take-away from the event was that storytelling can be learned. Today, we have a growing number of tools and a wealth of data to help us understand our world. Those who learn the skills of storytelling and data visualization now will increase the chance that their voices will stand out among the noise.
Mark Schroeder is a AAAS Science and Technology Policy Fellow at the National Science Foundation.
The organizers would like to thank many people for their help with this event. The speakers: Marie Whittaker, Alberto Cuadra, John Muyskens, Michael Pack, Aaron Milner, Shannon McKeen, Christy Steele, and Jon Schwabish; The moderators: Paul Tanger and Elizabeth Zeitler; The Tableau trainers: Tim Pritchard and Anna Lee; John Muyskens for doing a D3 training on top of speaking; Chinonye Nnakwe and Barbara Natalizio for the lunchtime Design Thinking workshop; Sarah Davis for her unflagging efforts on logistics and event coordination; AAAS for use of the facilities; and the National Consortium for Data Science for funding the catering. Joanna Chan deserves a special thank you as the organizer who made sure everything came together in the end. Slides from the speakers are available here. See also the blog post related to Jon Schwabish’s talk.