Study of 1996 election shows racial disparities

By Thomas Hargrove
Scripps Howard News Service

Votes cast in predominantly Hispanic and black communities are more than twice as likely to miss being counted in presidential elections as are ballots cast in the rest of the United States.

The apparent culprits are voting systems with a higher rate of inaccuracy, like punch-card balloting, and a general failure of election officials to adequately assist less-educated voters who live in communities with large numbers of racial and ethnic minorities.

These findings are based upon a new study of voting in the 1996 presidential election, conducted by Scripps Howard News Service, that examined the so-called voting drop-off rate: the difference between the number of ballots cast and the number of votes actually counted in major elections.

The study examined official returns in 1,825 counties and voting precincts nationwide.

The median drop-off rate four years ago was 2.08 percent. But the study found it rose to 4.21 percent in counties and precincts in which more than half of the voters are black. It was 4.41 percent in majority Hispanic areas.

"This differential in uncounted ballots is something we just did not know about," said Curtis Gans, director of the Committee for the Study of the American Electorate. "We've known about intimidation of the minority voters in the past, but not about this."

Some of the drop-off rate resulted from a conscious decision by some voters to ignore the presidential race and participate only in local or state elections.

But Gans said "there is absolutely no reason" to believe that minorities are less interested in the presidential election than are non-Hispanic white voters.

NAACP President Kweisi Mfume last month asked for a Justice Department investigation into the alleged roadblocks that minorities face in casting ballots and having their votes counted. "It is our moral obligation ... to insist that all voters be allowed to cast an unfettered ballot," Mfume said following the disputed presidential election in Florida.

The Scripps Howard study found that the drop-off rate generally rises in areas with higher-than-average levels of poverty and illiteracy, especially in southern and southwestern states.

"This is a problem of a lack of education and a problem of new voters who have a harder time learning how to use the voting equipment," said Kimball Brace, president of the Election Data Services consulting firm that provided some of the information used in the study.

"There are some jurisdictions that do a better job than others with outreach programs to voters," Brace said.

"In some areas people are out in the shopping malls demonstrating how to vote or providing literature on the method of voting in their counties. But we don't see a lot of this in many places."

The study found that different election machines can produce enormous variations in the numbers of votes that are apparently discarded, either because voters become confused and vote for too many candidates or because the equipment fails to register an intended vote.

The median drop-off rate was only 2.27 percent in counties with large black populations using mechanical lever voting machines first made popular in the 1950s. It rose to 5.03 percent in counties using punch-card balloting, to 7.68 percent in counties using the multiple-card system known as Datavote, and to 10.69 percent in counties using hand-tabulated paper ballots.

In communities in which at least a quarter of the voters are Hispanic, different voting machines also produced different drop-off rates. The newest generation optically scanned balloting systems produced the lowest drop-off of 1.31 percent. Punch cards resulted in 3.78 percent drop-off and electronic voting machines had a 4.26 percent drop-off.

"Not only did the problems this year in Florida open our eyes, but, because of it, we are going to do something about it," Gans said. He said members of Congress are making preparations to upgrade America's election equipment.