After the South Carolina primary on February 29th, the Washington Post published an article on the changes in turnout from the 2016 primaries. Turnout is not the most common electoral statistic to see reporting on, and it certainly tells a different story than vote margins. The Post uses turnout in this article to suggest that Biden is, “(attracting) different types of energized Democratic voters” and by disaggregating turnout by race, that these voters are largely white.
Lenny Bronner, a contributor to this article and a data scientist at the Post, is a main contributor to a new blog, PostCode, dedicated to telling the backend stories of the Washington Post’s articles. In the past, Bronner has blogged on how the methods used by the Post to predict turnout on election nights. Bronner explains that the usual metric, precincts reporting, “can mislead readers.” In a recent post, Bronner dives into a method the Post is using to estimate primary elections. This method, looking at voter flow from one primary cycle to the next, estimates how voter preferences change from one primary to the next. The Post, then, is not only making an effort to diversify statistical methods used to tell political stories but explain those methods to readers. Bronner has not yet responded for comment to this story.
On the audience side, it is still challenging to tell how methods in political data journalism are interpreted. For consumers of data journalism, it can be difficult to parse through the intricacies of different statistical methods used for reporting. Especially for organizations using data journalism to inform their work, correctly interpreting what a set of visualizations or metrics says can be crucial, especially come Super Tuesday. Both the W&M Young Democrats and the Williamsburg James City County Democrats have not yet responded for comment to this story.
 Bronner did eventually get back to me, but we have not yet had the time to formally chat.