RELEASE:
Tiered Election Audits: An Improved Approach to Accurate Election Results
A recent report released by the National Election Data Archive entitled "Tiered Election Audits" suggests a tiered tabulation system for audits may be the solution to making sure votes counted more closely match the votes cast.
A bill sponsored by Rush Holt (D-NJ) dealing with election reform is possibly being released on the House floor this week. In light of the increase in contested elections since 2000; the 18,000 missing votes in Sarasota FL in 2006 <1>; the introduction of new voting technology that many reputable scientists insist is flawed <2>; the fact that the testing lab that certified a majority of US voting machines was recently decertified by the U.S. Election Assistance Commission <3>, and that two election officials in Ohio were convicted on January 24, 2006 of rigging the 2004 presidential recount <4>, concerned citizens are questioning the integrity of our voting systems.
According to Kathy Dopp, President of National Election Data Archive (NEDA),
citizen oversight of manual election audits are needed to make sure that votes are counted accurately. * Dopp says that "If the goal of elections is to ensure the will of the voters, then election audits should be designed to ensure that voters determine who represents them. However, the US Congress may be planning to require election audits that are designed instead to ensure that a certain percentage of votes are accurately counted. For instance, if Congress requires audits to ensure that 95% of votes are accurately counted, this would allow 5% of votes to be switched wrongly from one candidate to another, leaving any race with less than 10% margins between candidates wide open to vote fraud."
How many ballots have to be hand counted to detect vote counting errors that are big enough to change the outcome of an election? When the margin between candidates is smaller in close elections, a smaller amount of vote miscount can wrongly alter the outcome. Central to the concerns regarding accurate vote counts is whether or not manual counts of a flat 2% of vote counts are sufficient to ensure the integrity of election outcomes or whether higher audit rates are needed when races are close.
According to Dopp, it is not enough to specify election audit percentages. A minimum number of vote counts must be audited because some election races involve fewer total number of vote counts. For example, if one-in-20 vote counts were corrupt, then at least 20 vote counts must be sampled to detect at least one of the corrupt counts. Although a 10% audit of 500 vote counts would sample 50 counts and be sufficient, a 10% audit of 20 vote counts would sample only 2 counts and have little chance of detecting the corrupt vote count.
On January 26th, the National Election Data Archive (NEDA) released a new paper, "Tiered Election Audits" that provides a table to look up the margin between the leading candidates and find a percentage and a minimum audit amount needed to ensure that election outcomes are accurate.
document in it's entirety is here:
http://electionarchive.org/ucvInfo/release/Release-TieredElectionAudits.pdf(emphasis, mine)
* On citizen oversight and comments by Land Shark here:
http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=465777&mesg_id=465777All members welcome and encouraged to participate.Please post Election Reform, Fraud, & Related News on this thread. :hi:
If you can:
1. Post stories and announcements you find on the web.
2. Re-post stories and announcements you find on DU, providing a link to the original thread with thanks to the Original Poster, too.
3. If you have information from an election reform activist organization outside of DU feel free to post (local or national)
4. Start a discussion thread by re-posting a story you see on this thread.
If you want to know how to post "News Banners" or other images, go here:
http://www.democraticunderground.com/forums/faq.html#image