This week’s Data Darts and Laurels features newcomer Storyline, a just-launched section of Washingtonpost.com “dedicated to the power of stories to help us understand complicated, critical things,” according to Jim Tankersley, who is heading the project. Storyline treats each individual piece as part of a larger story thread. For example, so far there have been several stories written about the “uneven recovery” from the recession and the “immigration stalemate.”
Data journalism plays a big part in Storyline’s storytelling so far. Staff writer Jeff Guo was effective in his use of data to explain the “uneven recovery,” showing how a decline in birth rate is correlated with the weak economy in recent years. He was less effective in a piece that presents a graphic breaking down by country how many people registered with the Deferred Action for Childhood Arrivals program, which allows younger undocumented immigrants to avoid deportation by coming forward. There isn’t much of a conclusion in this specific piece, other than that a lot of Mexicans are participating. “Some speculate that Asian communities might have fewer resources for the undocumented, or might be more suspicious of such programs,” Guo writes, linking to a New York Times story on Asian adoption of the program. The Storyline piece would’ve been better if it had explored Asian reluctance to the program rather than leaving it to vague speculation. As it stands, the story ends in a pretty unsatisfying way. However the way Storyline is set up, anything left unexplained can and will be revisited at a later time. For understanding that data can explain, but not entirely encapsulate a complex story, Storyline tentatively gets a LAUREL.
Earlier this month, the Pew Research Center released an insightful study about how the number of journalists covering the 50 statehouses have dwindled in recent years. In its analysis piece, Pew established that larger states like Texas, California, and New York have the largest numbers. This week, Pew ran a short piece that mapped the data by calculating the ratio of statehouse reporters to the state’s entire population. That map implied that states like Vermont, Wyoming, and Alaska — in other words, states with the smallest populations — had the most robust statehouse press corps. The map and additional analysis was featured in a short piece by The Washington Post’s Chris Cillizza that had the headline: “Vermont is the best covered state in the country. California is the worst.” That’s not true, as Pew established in its first piece. Furthermore, the ratio of reporters to population is irrelevant—a more substantive way to conclude if a statehouse is well covered is the number of reporters to legislators. For finding significance in meaningless calculations, both Pew and Cillizza get DARTS.
One of FiveThirtyEight’s best pieces of data journalism this week involved no number-crunching. Freelancer Clare Malone interviewed Sarit Michaeli, spokeswoman of B’Tselem, an Israeli organization that, among other things, works to count the number of people killed in the longrunning Israel-Palestine conflict. The Q&A piece gives insight on how the organization gathers and vets information by checking first-person accounts, death certificates, and even photos of dead bodies against the counts of other organizations and media. “We check and double-check everything before we publish it,” she says. Michaeli acknowledges some issues they face, such as Palestinians potentially lying about the number of civilians killed or the difficulty of counting the dead during an artillery attack. For helping explain how data is generated in the ongoing conflict, where counting bodies is extremely difficult, this piece gets a LAUREL.
Following one of the other big international stories in the past week — the crash of Malaysia Airlines Flight 17 — FiveThirtyEight’s Nate Silver tried answering the question: “Should Travelers Avoid Flying Airlines That Have Had Crashes in the Past?” His conclusion makes sense: Airlines from richer countries tend to be safer and flying, as a whole, is very safe. But the way he gets to that point get a DARTS, as it is a bit convoluted and arbitrary. His calculations are based on 30 years of safety data, broken down into two 15-year blocks, which is then measured against distance flown per passenger rather than simply the amount of flights flown. A simpler way of approaching this data may have been just calculating the time and number of flights between incidents for a given airline. The conclusion would probably be the same, but the process of getting there would be more meaningful and easier for readers to grasp.