Election Day 2013 renewed complaints that badly designed ballots are dragging down American democracy. These gripes were particularly loud and widespread in New York City. To some, the city’s ballot was simply difficult to read. To others, shoddy design helped explain why a full quarter of the million-plus New Yorkers who voted for mayor didn’t vote for even one of the six measures on the ballot’s back side.
But what, exactly, makes a ballot good or bad?
We put the question to Dana Chisnell, a ballot usability and design expert who literally wrote the book — or books — on ballot design. She is co-author of a report from the National Institute of Standards and Technology on the use of plain language and creator of a series of Kickstarter-seeded “Field Guides to Ensuring Voter Intent.” When it comes to tricky ballots, New York City is hardly unique. (Palm Beach County, Fla., anyone?) But we asked Chisnell to drill down into its example, and give an expert read on what makes a ballot a designer’s dream or nightmare.
On the overall design:
Chisnell: There are a lot of interesting things about this ballot. For one, the layout of ballots in New York, New Jersey and Connecticut is different from the layout of ballots in the rest of the country. The idea with this ballot design was to mimic the old mechanical lever machine layout, so there are automatically some constraints that other voting systems don’t impose.
Given that, the most remarkable thing is the tiny type. This has been a complaint about New York City ballots for a long time. And contrary to what you might think, using all uppercase is not actually helpful to readability. The space in each of the candidate boxes could be used better. The major constraint here is that the bubbles to fill in must line up with the scanner registration marks on the right side, so the scanner can tell where there should be marks. (Scanner registration marks are those black blocks around the outside of the ballot proper. The tabulators that scan the ballots use those marks to know where to look on the paper for votes.)
But the positioning of the bubbles in the boxes, along with the positioning of the candidates’ names, feels random — the lack of proximity is weird and makes it difficult for voters to be sure they’re marking the ballot exactly the way they intend. Duplicating the party names might appear to help, but it doesn’t. It just adds noise and makes the ballot difficult to navigate.
A little shading on the rows that show the party names might help. Then you could take the party names out of the candidate boxes, but you’d have to play with the size of the heading rows and the weight of the horizontal rules to make sure voters can tell what goes together. It’s a bit weird that there are no instructions for marking the ballot on the front.
I should have started by saying that there is some nice white space overall on the ballot, and the typeface used is pretty good.
On the lack of notice about ballot measures on the back:
Chisnell: That is a classic problem with all paper ballots. In some places, like Minnesota, there’s a poll worker whose job is to tell voters that there is something on the back when they get handed their ballot. In other places they also add an instruction at the bottom on both sides saying to turn the ballot over.
There are any number of reasons they didn’t do that here. There might not be room. There might be legislation that limits what can appear on the ballot. Or they might not have thought of it.
On the design of the ballot measure choices:
Chisnell: Using a ragged right margin rather than fully justified text is great. Having big numbers that correspond to the measure number is nice. And the line length is pretty good. There was an opportunity to make the ballot measures easier to read because there’s so much room. The type didn’t have to be that small. And the headings are indistinguishable from the body text, which isn’t helpful. Headings are one way voters find their way around and identify the things they want to pay attention to and vote on.
On the good side, it looks like the wording is fairly plain and clear, though I’d love to test it on people with limited English proficiency and low literacy.
On how it compares to other ballots:
Chisnell: Some of the best ballots in the country are done in Oregon and Cook County, Ill. They have clear, plain instructions that include illustrations on how to mark the ballot; all mixed case; titles and other information aligned left, rather than centered; shading to separate contests and levels of offices; and instructions where they are needed, at the bottom right of the ballot, to turn it over; and simple language.
On why some jurisdictions ballots are better than others:
Chisnell: For many years it was because the voting systems didn’t actually support good ballot design. You couldn’t make a ballot that was well designed and that the system could read, because they were engineered and designed in the 1990s without thinking about usability for voters.
The second big factor is election legislation. Most of it was written in the 1890s and hasn’t been updated since then. There’s a bill in the New York Assembly sponsored by Brian Kavanagh to reform that. [It’s called the Voter Friendly Ballot Act, and while it has passed the Assembly, it has failed to pass the Senate. – Ed.]
By and large, election officials are everything you want in public servants. They work very, very hard with very little in the way of resources. And the job has changed a lot since 2000, going from a clerk kind of job to really being much more of an IT job. But they don’t have design resources and most don’t have a lot of design literacy. Until recently, most have not felt empowered to change anything related to design. It’s just not something they think about.
This interview was conducted over IM and has been condensed and lightly edited. Thank you to the Brennan Center for Justice for providing the sample ballot. For more on the topic, the American Institute of Graphic Arts’s Design for Democracy group has, on behalf of the U.S. Election Assistance Commission, created a report on ballot design best practices.
Nancy Scola is a Washington, DC-based journalist whose work tends to focus on the intersections of technology, politics, and public policy. Shortly after returning from Havana she started as a tech reporter at POLITICO.