When you think of elections as a design problem, you see a plethora of opportunities in everything from the wayfinding of polling stations, poll workers training workshops, to the ballot. We looked at the Rhode Island ballot and discovered several areas of improvement based on user insights. The current ballot is daunting to say the least and is not intuitive by any means. In a very short 5 week period, we sought to change that. We were driven by the question, how can we make the ballot so intuitive that it doesn't need any instructions?
Current State of the Rhode Island Ballot
Observations and interviews at the Woonsocket polling station
Version A: The version that wouldn't scare off the Board of Elections
For version A, we stayed within the constraints of the existing ballot, meaning that the famililar marking style was kept the same and the layout was also kept the same so the scanning machine would read the markings. We paid attention to the layout of the instructions so that it wouldn't look as daunting as the original did. The numbers in front of each kind of voting was also subtracted as it previously looked like a series of steps to take rather than different kinds of voting one could do (Voting, Write-in Voting, Straight Party Voting).
Version B: The version that would scare off the Board of Elections
For Version B, we disregarded the existing ballot constraints. We switched the strange arrow mark-making with the more intuitive bubbles. We sectioned off the straight party voting section in a more distinct way so that people would not overvote like they previously did (the original ballot made it seem like straight party voting was just another question to answer on the page).
Based on our user testing, this version was preferred by 80% of the participants over the original ballot. In fact, two users literally refused to take the original ballot when we presented it to them. It was visually daunting for them. However, when we presented Version B to the same users, they agreed to fill it out. There was a very distinct reaction for each ballot.
Preliminary User Testing
Conclusion
Our user insights and reimagined ballots were presented to the Rhode Island Board of Elections as well as representatives from the Secretary of State's office. They were very receptive to our iterations and have been in discussion about implementing the changes in the long run. More detailed user insights can be seen in the presentation slides below.
