Journal Archive > 2002 > February

Election outcomes

Ballot design does influence voters, study finds

The design of a ballot can directly affect the choices voters deliberately make at the polls, according to a new Tufts study.

"We already knew that ballot design can change how people vote by accident—as was so clearly illustrated in Florida's 'butterfly' ballot fiasco during the last presidential election," said James Glaser, associate professor and chairman of political science, who conducted the study. "But in this study, we've found that ballot design can influence how people purposely vote, as well."

The Tufts study, which examined a 1991 vote in Jackson, Miss., showed the ballot's design influenced voters to pass a school bond issue for the first time in 25 years. Glaser's study was published in the January issue of The American Journal of Political Science.

The study also suggests that ballot design can be influential enough to counter what might be considered racially divisive political issues. The Mississippi school bond vote was momentous because the city's population is 50 percent white, while its public school population is overwhelmingly black. Historically, school bond issues had been entangled in racial conflict, Glaser said.

Concerned that the racial divide would prove too deep to pass the bond issues after several similar votes had been rejected, Jackson, Miss., school officials chose to present the ballot as a checklist of separate and specific school projects, rather than listing the vote as one general "omnibus" bond measure.

The ballot's individual "mini-bonds" included financing for air conditioning and general renovations, new computer equipment and new athletic facilities, among other initiatives. As a result of the way the ballot was presented to voters, three of the seven checklist bond issues passed, rather than the entire combined bond package being defeated, as it had been in the past.

Glaser's study used public opinion experimentation to simulate the election and confirmed that the "checklist ballot" did, in fact, have a profound effect, especially on white voters.

The study, based on 1,400 randomly selected Mississippi voters, not only proved that the checklist worked, but that the outcome was related to the restructuring of the ballot so that voters had more choices and felt a sense of control over how resources were to be allocated.

Glaser's study illustrates that white voters are more likely to support specific and meaningful initiatives such as air conditioning and computers in predominantly black schools, rather than a general bond that appears more costly and potentially wasteful.

"The study shows how racial hostility generated by competition over the allocation of resources can be overcome, and progress can be made based on how politics are practiced," Glaser said.