A Graphic Worth a Thousand Gaffes
Data is unbiased — so why do the visuals communicating it tell a different story?
--
Nobody wants to waste their vote. As the hulking machination that is the United States federal election eats billions of dollars and spits out-of-context quotes, many turn to seemingly impartial sources to simplify and make sense of it all.
More than likely, however, you and yours have been misled — not by a smooth-talking politician, but by subtle choices in data visualization. These choices can craft a particular and less-than-accurate message, with an echo that extends far beyond any single voter.
Quick disclaimer: My criticisms of how candidates are communicated through data visualization should not be construed as support (or condemnation) of a particular candidate and their politics.
Currently, a Google search for the US presidential race brings up a graphic that takes up the entire space above the fold. It shows the race for the Democratic nomination as nowhere near as close as it actually is.
The decision to stack superdelegates on top of pledged delegates unfairly favours Hillary Clinton. Superdelegates can switch their vote at any time, and many likely would if Bernie Sanders’ momentum shifts — remember, this is exactly what happened to Barack Obama in 2008.
The situation was even worse at the beginning of the election. Take a look at February’s version of this chart:
What makes this especially frustrating is that Sanders actually had more delegates than Clinton at this point in the race: 36 compared to 32.
Meanwhile, Bloomberg doesn’t even bother to distinguish between superdelegates and elected delegates:
While the Daily Kos uses a graphic that shows Sanders’ recent surge:
FiveThirtyEight offers the most impartial picture of the current race. It even provides text that explains superdelegates and why they weren’t included in their figures.
The subheader of this image’s source page explains that “tracking a candidate’s progress requires more than straight delegate counts,” and it goes on to showcase how Sanders’ support is starting to snowball via a chart that contextualizes his performance over time.
The Huffington Post suggests Sanders’ growing momentum can mean everything:
I thought Sanders had lost the nomination… But a month later, after many blowout wins by Sanders, and multiple self-inflicting wounds by Clinton, Bernie Sanders proved me wrong. Now more than ever, the light at the end of the tunnel is finally shining through. I thought his momentum was at a halt, but Sanders out-raised Clinton in both February and March (and that’s without Super Pacs).
Not everyone shares that outlook. Nate Silver, founder and editor-in-chief of FiveThirtyEight, isn’t as confident that Sanders’ momentum can carry him all the way to the candidacy:
Frankly, none of it is at all likely. If the remaining states vote based on the same demographic patterns established by the previous ones, Clinton will probably gain further ground on Sanders… But things can change, and polls can be wrong — and so it’s worth doing the math to see what winning 988 more delegates would look like for Sanders.
Facts are fixed: the lens of the writer, or the publication, are what gives a story a slant. It’s how both a positive and a negative outlook can be construed from the same data. The commentary is offered to make sense of those facts and figures, and to forecast potential results.
Then why, depending on your source, does that supposedly impartial data look just as different as these two stances on Sanders?
People increasingly rely on bite-sized ways to ingest vast swaths of data. When it comes to the political process, information gatekeepers have a responsibility to present information fairly and transparently. Misleading users, intentionally or not, can have consequences on a global scale.
And yet, if you settled for that first Google result, you would have no idea how close the race actually is. You’ll simply see big numbers, a bigger graph, and less-than-a-sliver of the actual situation.
