Underallocation of Voting Machines Suppresses Votes: Franklin Co. OH, 2004

The "bottleneck" effect is an inherent property of computerized voting machines that enforce serial, one-voter-at-a-time voting. This is most particularly so for touchscreen DRE terminals. This unadvertised feature can and will continue to be exploited by partisan election officials wherever they can get away with it, with the perpetual after-the-fact excuse that the voting machine shortage was an "oversight," an "unintended consequence" of "circumstances that could not have been foreseen."

Kathy Dopp, who originally published Liddle's study for USCountVotes.org, writes:

"Franklin County, OH was also one of the counties where the election administrators underallocated voting machines to Democratic districts and withheld voting machines in warehouses that they knew were needed during the election.
Here is a short study we did on it back then, but I know there have been others."

http://uscountvotes.org/ucvAnalysis/OH/FranklinCountyReport_v2.pdf
Click here to download and read the report.

Ohio election activist and researcher Rady Ananda writes:

"I recall this - and I witnessed it when I served as a numbers runner for the Dems on 11-2-04. I went into four precincts on Election Day: three predominantly Democrat and one predominantly Republican. At the Republican precinct (in white, wealthy Westerville, OH), voters waited maybe 10 minutes to vote. Volunteers served free coffee and pastries. At the Democratic precincts, the lines went out the door (into the rain) and those voters waited at least an hour, often longer. No refreshments were offered.

Matt Damschroder withheld close to 100 voting machines, and a lawsuit filed by the Democratic Party (at 5 pm) failed to get them deployed. This was a successful vote suppression strategy. Matt serves as a "consultant" to the Franklin BOE thru the end of this year, which means his brilliant strategies will undoubtedly be used in the next stolen election."

Introduction to the Study

An article by Harvey Wasserman in The Free Press ¹ alleged that long lines on Election Day, 2nd November 2004 deprived some voters in Franklin county, Ohio, of the right to vote, and that this problem was greatest in the most strongly Democratic precincts. If these allegations are well-founded, they should be apparent in the election data. I therefore analyzed data from Franklin County supplied by Cliff Arnebeck ².

The results indicate that in precincts where the number of “active voters” (voters who have voted at least once in the last two election cycles) per voting machine was high, turnout was significantly depressed as compared with turnout in precincts where the number of “active voters” per machine was low, supporting Wasserman’s claim that turnout was indeed depressed by inadequate provision of machines.

The results also indicate that the number of “active voters” required to share a machine in a precinct was strongly and positively correlated with the proportion of that precinct’s vote for John Kerry, again supporting Wasserman’s claim of selective under-provision of machines to strongly Democratic precincts. Statistical tests of both these effects indicate that they are highly significant, and thus extremely unlikely to have arisen by chance.

Two things are clear to the naked eye from the data. First is the finding that where the number of active voters per machine in a precinct exceeded a certain level, turnout (as a proportion of potential voters) dropped off (Figure 1).

To download and read the entire study with graphs, click here.