Timelines are generally considered the lowest form of data visualization, because displaying data chronologically doesn’t tend to provide much journalist value. But The New York Times recently upended that theory with a timeline-cum-scattergraph of driving safety records. This timeline works because it’s not actually a timeline—rather than running in a linear fashion, years have been plotted as dots, which are then set in a matrix whose axis measure the numbers of car crashes and miles driven. The focus of the piece is not on the historical angle of the story, but rather on the relationship between miles traveled and deaths on the road.
Hannah Fairfield, the Times’s senior graphics editor, said her inspiration for this piece came from reading Steven Levitt and Stephen Dubner’s Super Freakonomics. Their schtick is that they don’t always take data sets and trend reports at face value. In the book, the authors discuss the fatality analysis reporting system, a database of reports on factors behind road accidents.
This discussion sparked Fairfield’s interest in looking at previously unexplored trends in driving safety records. She noticed that data about fatality records is almost always examined through the lens of miles driven every year. Thinking freakonomically, Fairfield started to probe further into the numbers.
Fairfield said she took figures relating to deaths per mile driven over time and “tore them apart.” This meant charting deaths per 100,000 people against miles driven over time and plotting the results as a scattergraph. Each dot represents the annual figure of how many car-related deaths there were in relation to miles driven. The outcome was that, as the American population grew, driving miles increased drastically.
The production of this piece came together in what Fairfield described as a process similar to reporting any other story. The bulk of her time was spent gathering data from various sources, including the National Highway Traffic Administration (NHTSA) database, and then analyzing it all. After charting the numbers in various ways, she discovered the central element to her story: the plateau-drop pattern. That is, the number of road fatalities remained relatively steady during the 1970s and then took a drastic plunge; interrogating the data revealed a number of reasons for this, mainly technology advances and tighter road rules. In order to verify her hypothesis, she spoke to auto safety experts and the NHTSA.
The design of this piece is simple and subtle. The topic of driving of course lends itself to movement, so the slightly curved and wonky lines of the graph are in keeping with the thematic tone without ostentatiously incorporating images of cars and roads. The six text blocks add necessary historical context needed to explain the drastic variations in the data. Despite the simplicity and lack of interactivity in this piece, it feels very animated. The graph snakes around the text, incorporating it into the piece and inviting the reader to read it. The reader looks at it and sees that as time has passed, auto fatalities have decreased while miles driven have increased.
Making the annotations clear and readable was a big consideration for Fairfield. “For most people, it’s not an intuitive chart, since readers are more used to seeing time on the horizontal axis rather than as dots on a scatterplot, so the design needs to guide readers through it,” she said.Anna Codrea-Rado is a digital media associate at the Tow Center for Digital Journalism at the Columbia University Graduate School of Journalism. Follow her on Twitter @annacod.