[ skip to content ]

DOES EVERY VOTE COUNT? IT DEPENDS ON WHERE YOU LIVE, SAYS PROFESSOR

Voters in Florida know many of their ballots didn't count in November's presidential election. Voters in Virginia should know the same thing, and it all comes down to the type of voting system that's used, according to a new report by Old Dominion University's Stephen K. Medvic.

"Optical scan and, to a lesser extent, punch-card systems are clearly problematic," said Medvic, an assistant professor of political science.

"Anyone wanting to implement op-scan voting systems statewide ought to rethink that course of action," Medvic argued.
On the other hand, electronic (or touch-screen) voting systems had low rejection rates. "The problem there is the high cost of such systems," said Medvic. "But our voting systems ought to be nearly error-proof, and electronic systems are as close as we can get right now."

According to the report, more than 21 percent of Virginia voters use electronic voting systems; an equal number use an op-scan system. (An op-scan system is similar to what is frequently used to record the results on standardized tests, for which a student uses a number-two pencil to color in circles. The forms are then fed into a machine and scanned.)

Using data from the State Board of Elections, Medvic's study determined "rejection rates" for each locality in Virginia. The "rejection rate" is the difference between the total number of ballots cast in a jurisdiction and the number of votes counted in a given race, divided by the total number of ballots cast in that jurisdiction.

While much was made of "undervotes" in the Florida balloting, Medvic's study looked at both "undervotes" and "overvotes" - the combination of which he terms "rejected votes."

The key findings of the study note that:

· The statewide rejection rate was 2.41 percent. The lowest rate was .33 percent (Albemarle County), and the highest was 9.77 percent (Brunswick County).
· The best performing system was the electronic voting system (e.g., touch-screens), which had an average rejection rate of .9 percent.
· Mechanical (i.e., lever) systems had an average rejection rate of 2.2 percent.
· Punch-card systems had an average rejection rate of 2.3 percent.
· The worst performing system was the optical scan system, which had an average rejection rate of 3.2 percent.
· Optical scan systems were significantly worse (statistically speaking) than both electronic and mechanical systems.
· While punch cards did not perform worse than electronic or mechanical systems, this method was no better than the optical scan system (again, statistically speaking).
· With respect to demographic factors, those jurisdictions with higher numbers of college graduates and with higher median household incomes had significantly lower rejection rates; the influence of the racial make-up of the locality depended upon how "racial make-up" was measured (under one measure race had no influence, under another the percentage of whites in a locality had a negative influence on the rejection rate); and, there was no significant pattern of vote rejection based on the partisan leanings of the locality.
· When voting systems and demographic factors were analyzed simultaneously, the use of the optical scan system and the wealth of the jurisdiction were the two most significant predictors of the rejection rate. Race and the use of punch-cards were also predictors, but at a lower level of statistical confidence.
· There were no clear patterns of voting system use, save for the use of mechanical systems in areas with relatively low median incomes and few college graduates.

The conclusion of the report is that optical scan systems should be avoided (with the possible exception of those where votes were counted at the precinct as opposed to a central counting location), as should punch-card systems, particularly in jurisdictions with low incomes and few college graduates. Instead, the report says Virginia localities should begin to adopt electronic voting systems.

A breakdown of voting systems by locality can be found on the State Board of Elections Web site: www.sbe.state.va.us/Election/Voting_In_Va/virginia_voting_systems.htm.

This article was posted on: April 6, 2001

Old Dominion University
Office of University Relations

Room 100 Koch Hall Norfolk, Virginia 23529-0018
Telephone: 757-683-3114
http://www.odu.edu/news

Old Dominion University is an equal opportunity, affirmative action institution.