Study Finds Ballots of Poor Minorities More Likely To Be Tossed

Ballots do not indicate whether a voter is black or white, rich or poor. But a new report from Congress says race and income play a crucial role in determining whether a ballot will be counted or not.

Residents of low-income, high-minority districts are three times more likely to have their ballots tossed than those who live in wealthier and whiter, areas.

The report , completed by Democratic staffers on the House Government Reform Committee and released Monday (7/9), is the first to analyze the problem on a national level, echoing the findings of state studies in Florida, Illinois and Georgia.

It analyzes 40 congressional districts nationwide: 20 poor, high-minority districts and 20 wealthy, low-minority districts. The report found that in poorer districts, 4 percent of ballots were discarded, compared with 1.2 percent in the 20 affluent districts. It also found the introduction of newer technology, such as optical scan cards that reject spoiled ballots and give voters a chance to fix their mistakes, reduced uncounted votes substantially.

Voting machines, and particularly Florida’s infamous punch cards, were pegged as the early culprits responsible for many of the problems in the 2000 presidential election. As reviews of the election continue around the country, the emphasis has broadened to include poll worker training, sample ballots, registration roll maintenance and voter education.

Sponsors of the report say their findings should shift the emphasis back to machines and show that what happened in Nov. 2000 was not limited to the Sunshine State. “This problem is an urgent national priority,” said Rep. Henry A. Waxman, the committee’s ranking Democrat. “The technology is available to make sure that everyone’s vote counts. It is intolerable to allow the disenfranchisement of poor and minority voters to continue.”

Organizations representing minorities and people with disabilities have been urging state and federal lawmakers to adopt strict national standards for machines. Many state leaders, including election directors, lawmakers and governors, have pushed to keep the historic though not Constitutionally-guaranteed flexibility in the nation’s many election systems and styles.

Where this new report will fit into the debate remains to be seen, said one elections expert.

“Machines are certainly relevant, but what we have learned in the past few months is that there may not be a single panacea here,” said Norm Ornstein, an elections reform expert at the American Enterprise Institute. “What this study certainly suggests is that the machines do matter whether legitimate votes are counted and suggests that it is true even if one accounts for the nature of the precinct or voting area. Changing a machine makes a difference, but it certainly doesn’t make all the difference.”

Another group of researchers from the California Institute of Technology and the Massachusetts Institute of Technology found race and income alone cannot account for an area’s rate of ballot spoilage.

Stephen Ansolabehere, an MIT professor leading the Voting Technology Project , said the data produced by the House staffers should be “judged carefully.””There is a big difference in [spoiled votes] in rich counties and poor counties, but there are other variables too, such as whether the county has a lot of elderly voters or very few elderly voters,” Ansolabehere said. “You’ll find difference in all of those indicators.”

What the analysts do agree upon, however, is that optical scanning or touch-screen technology that gives voters a chance to fix their mistakes can eliminate many of the disparities in voting across the country.

In Alabama’s 7th Congressional District, for example, 31 percent of the population lives below the poverty line, but only 0.3 percent of ballots cast for president in the 2000 election went uncounted. The House report also noted in Louisiana’s 2nd Congressional District, another low-income, high-minority area, a touch-screen voting machine reduced voter error to 0.5 percent.