Design principles for ranked choice voting

Best practices for designing ballots, voter education, and election results

Voters deserve information and ballot design that helps ensure they can vote as intended and understand how their vote is counted. The Center for Civic Design partnered with the Ranked Choice Voting Resource Center and FairVote to research best practices for ranked choice ballot design, voter education, and presentations of the election results. We have completed several projects, each looking at different aspects of ranked choice voting, creating best practice samples for each.

Best practices for ranked choice voting ballots and other materials
The first project looked at ballots along with the instructions and other information for voters. W

Ranked choice voting election results – interactive demonstration
Understanding the results is as important as knowing how to vote. Embedding a narrative of the counting process in the results display so helps voters envision how their ballot was counted and how all votes contributed to the final winning tally. The interactive demonstration lets you explore several different elections.

Ranked choice voting for runoff elections for overseas voters
Five states use ranked choice voting to allow overseas voters to participate in runoff elections by ranking their preferences for a runoff if their first choice is not a winner. Helping them do so successfully required explaining runoffs in addition to how to mark their ballot.

Designing an accessible ranked choice voting ballot
We used the Anywhere Ballot interface as the basis for research into how to support voters who are blind or very low vision, have limited or no use of their hands, or have cognitive or attention disabilities.

Recommendations for introducing ranked choice voting ballots
In our research, when ballot design uses best practices, voters make few if any mechanical errors marking their ballots and can focus their attention on marking their selections. Participants in our testing were confident using both grid-style and optical scan columns. Those new to ranked choice voting preferred ranking up to 6 candidates.

Summary of the best practices

Each of the best practice guidelines looks at one element of voter materials to show how it contributes to helping people participate in ranked choice voting as informed voters.

The examples are a toolkit of designs that worked well in our testing, which election officials can use in creating their materials. We hope that this provides a balance between wanting to know the “best” design and the need to create election materials that must adapt to the specifics of local election law, current procedures, and voting systems.

Principles for designing voter information

Give voters the information they need to prepare to vote.  Voters need voter education materials that  explain the concept of ranked choice voting clearly so they can make confident choices on their ballots.

  • Briefly explain the benefits and basics, starting with a simple definition.
  • Answer voters’ questions about how to make their choices and how their ballot is counted: how many do they need to rank, how ranking affects their first choice, and when additional choices are counted.
  • Show the connection between how the voter’s own ballot is counted and the overall counting process.

Present all information in clear, simple language.  Ranked choice voting can seem complicated, so making the information simple is especially important. Write for voters of different levels of reading and civic literacy.

  • Keep information short and to the point, using everyday words.
  • Use illustrations and visual explanations to help explain how ranked-choice voting works.
  • Use icons sparingly to draw attention to different types of information.

Repeat essential messages across different voting materials.  Voter education and ballot instructions support each other.

  • Provide general instructions about ranked choice voting in the polling place or vote-by-mail materials in an appropriate format.
  • On a digital ballot, make this information easy to find with an always visible “help” button on all screens.

Best practices for ballot design

Use best practices for ballot design.  Use best practices for ballot design from the Election Assistance Commission and Voluntary Voting System Guidelines requirements for accessibility as a starting point for ballot layouts.

  • Use typography to make the ballot easy to read.
  • Use visual elements to attract attention and separate contests.
  • On a digital ballot, only one contest per page.

Include clear, concise instructions that will help voters avoid errors and rank candidates as they intend. Instructions on paper and digital ballots should be easy to find and easy to follow.

  • Put instructions where voters need them: before the first ranked-choice voting contest.
  • Use informational icons and typography for emphasis.
  • Illustrate the correct way to marking a paper ballot

Best practices for ranked choice voting for runoff elections

We developed two ballot layouts – a split grid and a two-page design, building on the best practices. Both designs worked well to communicate how to use ranking to  vote in a runoff election, offering elections offices a choice in their approach.

Explain what a runoff election is, and the benefits of using ranked choice in this context. Voters needed to know not just how to mark the ballot but why.

  • Be clear that the ballot is for two elections.
  • Provide explicit information about how this ballot works.
  • Break the voting process into two steps to help voters understand the process.
  • Be clear that voters don’t have to mark all choices.

The research report includes both ballot designs, highlighting how information about runoffs and how to rank their choices were incorporated into each version.

Best practices for accessible digital ballot design

This research built on the early explorations of a digital ballot design. Even working with mockups, participants were very consistent about wanting control of the process and for the interaction to be as simple as possible.

Give voters control of all interactions. Any sense that the system is making decisions made participants trust it less.

  • Have visible controls for all actions to show what actions are possible-and for accessibility.
  • Do not automatically re-order candidates while voters are in the middle of ranking.
  • Allow voters to leave candidates out of the ranking. Candidates not ranked are shown at the end of a rank order list.
  • Provide a review screen where voters can review their rankings and make changes or add additional selections.

In the prototype, voters select candidates in the order they want to rank them. At any point, they can put the list of candidates into a ranked order and edit the rankings. This simple interaction worked well to help voters discover how ranking works and (for many) encouraged them to rank more candidates.

In the audio design, create a structure for efficient listening. Consistency and clear cues helped give participants the confidence to move quickly through the ballot.

  • Use a consistent syntax of “To [do something] press [key name]”
  • Order the list of instructions so ordered so the most specific prompt is first, any general instructions follow, and the generic instruction on how to continue is last.
  • Use short pauses to separate chunks of text and allow a moment to absorb the more important information.
  • Use longer pauses to allow ‘space’ for the voter to interrupt the audio.
  • Add an explicit cue when moving from one area of the screen to another (for example, contest information,  the list of candidates, or general ballot navigation).
  • Make the transition from contests to the review screen as obvious in audio as it is visually.
  • Present only the list of selected candidates and their rankings on the review screen, with a prompt for the number of unranked candidates.

For a polling place voting system designed with little ability to customize the interacion, the biggest design challenge is the different combinations of accessibility needs. For example, voters may need the presentation in  visual, audio, or both. They might use touch or tactile controls with 2, 3 or 5 keys. Although our goal was to find a combination that is usable in all modes, having a setup function or a way to detect the interaction mode in use might allow the system to optimize the experience for voters.

Best practices for displaying election results

Make election results transparent.  Show the process of counting a ranked choice voting election when reporting election results will help voters understand how the winner is determined and build confidence in the voting system.

  • When reporting results, show the winner before explaining the counting process.
  • First, describe what happens in each round, then follow with a visual display.
  • Show all rounds of counting. Include vote totals for each candidate, the number of votes removed or added, the number of inactive ballots, and a “goal line” for the winners.
  • Make it easy to see the number of votes transferred to each candidate during each round.
  • Show inactive ballots in the results list. Differentiate inactive ballots from the active candidates.
  • Make it easy for users to navigate both forward and back to see the process of counting.

The interactive demonstration, based on real elections, lets you explore both single and multi-winner contests with a range of numbers of candidates and election outcomes.

How we worked on this project

We started with a review of the wide variety of materials already in use around the country from Portland, Maine to San Francisco, California, to get to know the range of ballot designs and approached to voter education. We then created prototype materials—ballots, voter education, and election results —which we tested with voters in California, New Jersey, and Minneapolis. Between cities sessions, we refined the prototypes as we worked to identify broad principles and the best practices report, released in December 2016.

The following year, we conducted two final usability tests to compare ballot layouts:

  • A large scale study compared experiences with 3 ballot layouts. In this study, conducted in Denver, almost all of the 112 participants felt confident their vote would be counted accurately all of the ballot layouts confirming that best design practices and clear instructions worked well to support voters. We also looked at how many candidates they ranked and found that the ballot style influenced this decision.
  • A study compared 3 different sizes of grid-style ballots. We intercepted 22 people leaving polling places in Maine, where they had just voted in a ranked-choice voting election. We wanted to understand to explore attitudes about different size grids. Working with voters in the midst of an election allowed us to build on their voting experience, helping us understand the difference between “power rankers” and more tentative novice rankers.

In 2020 we continued working on ranked choice voting, which focused on two specific contexts: overseas voters and an accessible ballot design.

The study of overseas voters and runoff elections had three components:

Our work on an accessible ranked choice, was done just as stay-at-home orders were being put into place, so the sessions with voters were conducted online. Despite the logistical challenges, we were able to work with 15 participants including voters with no use of their hands, autism or other attention and cognitive disabilities, and 6 blind voters.

We started with an interactive prototype forked from the ElectionGuard Github repository adding a new contest type and review-screen display. We used a static mockup for the printed ballot. As in all of our work, we made changes to the design between groups of sessions.

The work on the audio format was challenging because we wanted to be able to experiment with different phrasing fluidly, even trying alternatives during a session. We borrowed a research method from the Los Angeles County VSAP research team and used a human to be the voice of the voting system. The participant listened to the audio and simply spoke the name of the button they would press on the keypad. One of the researchers “drove” the interface so the moderator and observer could follow the interaction. It worked so well that one participant did not realize the audio was not digital.