In the weeks leading up to the 2016 election, headlines about “rigged” voting have raised questions about the legitimacy of election results. However, a new paper from Rice University psychologists suggests that a bigger problem may exist within America’s voting systems: poor ballot design.
“Voting is fundamental to democracy, and election integrity is one of the primary concerns of a democracy,” said Philip Kortum, an assistant professor of psychology at Rice and the study’s lead author. “Elections must reflect the will of the voters, with every vote being cast and counted as each voter intended. However, despite speculation from current news headlines, some of the biggest threats to election integrity are not from unseen forces with nefarious intent, but rather from poorly designed voting systems that fail to account for the perceptual and cognitive limitations of voters themselves.”
Kortum and Michael Byrne, a professor of psychology and computer science at Rice and the study’s co-author, provided multiple examples in the paper of how poor ballot design may have impacted election results, including the most high-profile example: Florida’s confusing “butterfly ballot” in the presidential election of 2000. In this situation, the arrangement of names on the ballot caused many voters to believe that they were voting for Al Gore, whose name was on the left side of the ballot, when in fact they were casting their ballots for a candidate listed on the right side of the ballot, Pat Buchanan.
Voters who did so were using a reasonable strategy: Holes and arrows do not always line up perfectly on butterfly ballots, so many voters simply count from the top of the ballot down to determine where to mark their vote. The researchers said that according to a 2001 research paper on the race, the number of votes that were miscast as a result of this voter confusion in Florida was greater than the margin of Bush’s victory, and almost certainly cost Al Gore the presidency.
Other examples in the paper of how poor ballot design may have impacted elections include:
- Sarasota, Fla., 2006: Every race on the ballot with the exception of one was presented on its own virtual page, but on one page, two races were presented. This design caused a large number of voters to fail to cast a vote (a phenomenon called undervoting) in a congressional race presented at the top of the page. In fact, the rate of undervoting for this race was six times as high in Sarasota relative to other jurisdictions that had different ballot designs. The authors said that it is highly probable that the outcome of the race would have been reversed if the 18,000 voters who undervoted in this race had actually indicated their preference, as the margin of victory was smaller than 400 votes.
- California gubernatorial recall election, 2003: This election’s ballot featured 135 candidates, which made it extremely difficult for voters to find a candidate. The long list of candidates also exacerbated a number of known selection biases, such as ballot order and name recognition, which had the potential to adversely impact election results, the authors said.
The authors said that poorly designed ballots can make voters’ intentions unclear and may even be so difficult that the voters unintentionally select too many candidates, the wrong candidate or fail to select a candidate at all, effectively nullifying their vote. The authors said that although there are several prominent examples of how poor ballot design may have impacted election, it is difficult to know just how many elections have been impacted due to poor design.
“Following the fallout from the 2000 presidential election, it became clear that we did not fully understand how our voting systems were operating from a psychological perspective, and that this lack of understanding could have a significant impact on voting outcomes,” Kortum said. “Only recently have we begun to grasp the difficulties people can have with the simple act of voting because of the voting system design.”
And for those who believe the shift from paper to electronic ballots will solve design issues, think again, the authors said.
“The widespread distribution of electronic voting machines over the last 10 years has not addressed this issue, and at times has made it worse,” Byrne said. “This is because these systems do not include everything we know regarding the psychology of design and human error. The chance to make things better with these systems missed the mark.”
Unlike other technologies and interfaces, people interact with voting systems infrequently and the systems often change from one election to the next, which means voters don’t have the advantage of learning how to use the systems through repeated use, the authors wrote.
“The most surprising thing we continue to find in examples such as the 2000 election is that even highly motivated people who take the time to register to vote, find their polling place and then make the effort to actually vote on Election Day can make voting errors so severe that their votes don’t reflect their intent,” Kortum said.
This is why it is so critical for voting systems, whether paper or electronic, to be designed with the voter in mind, the authors said.
“Understanding how psychological processes like memory, attention and perception shape voters’ interactions with these systems can inform the design process and make certain that all votes are cast correctly,” they said.
The authors added that now is the ideal time to use behavior science to address voting system design issues, as the systems purchased in the aftermath of the controversial 2000 election are nearing the end of their usable lives.
Kortum and Byrne are part of a team of computer scientists and other voting experts working with election officials in Texas and California to build systems that are not only secure, but quicker, easier and accurate — even for first-time voters. They hope that this paper will raise awareness, particularly with the people who administer elections, about the importance of ballot design. They also hope it will encourage other researchers to study the problem.
The study was funded by the National Science Foundation.
“The Importance of Psychological Science in a Voter’s Ability To Cast a Vote” will appear in the December edition of Current Directions in Psychological Science. To request a copy of the study, contact Amy McCaig, senior media relations specialist at Rice, at 713-348-6777 or firstname.lastname@example.org.