TO: CVF-NEWS FROM: Kim Alexander, CVF President DATE: April 18, 2001 RE: ACLU lawsuit & voting technology performance
This week the American Civil Liberties Union, along with several other organizations, filed a federal lawsuit demanding that the nine California counties currently using the Votomatic and Pollstar voting machines replace those machines before the next statewide election scheduled for March 2002.
The suit makes two claims. First, it says that voters who vote on Pollstar or Votomatic machines stand a greater chance of having their votes discarded than voters using newer, optical scan and touch screen voting machines. Second, the suit claims that the use of Pollstar and Votomatic machines has a "disparate and adverse impact upon African-American, Asian-American and Latino voters" because a higher percentage of these voters reside in counties that use Pollstar and Votomatic machines.
The lawsuit names California Secretary of State Bill Jones as a defendant, and calls on Jones to de-certify the Pollstar and Votomatic punch-card systems and ensure their replacement by more reliable systems, such as optical scan and computerized touch-screen systems. In response, Jones and his staff are calling on the ACLU and other plaintiffs in the lawsuit to join him and Assembly Speaker Bob Hertzberg in pushing for state funding to finance voting technology upgrades across the state.
Next Monday, April 23, three bills dealing with voting technology will be taken up by the Assembly Elections Committee. AB 56/Hertzberg would appropriate $300 million to fund voting technology modernization efforts at the county level. This bill has bipartisan support; it is authored by Assembly Speaker Bob Hertzberg, a Democrat, and supported by Secretary of State Bill Jones, a Republican. It is similar to AB 28/Pacheco, which would also appropriate $300 million for new voting technology, and, like AB 56, requires counties to spend $1 for every $3 the state spends. AB 55/Shelley, was just amended this week and includes a number of provisions dealing with voter education, voter registration notification, pollworker training, and the development of an online voting system. More information about these bills can be accessed via CVF's Voting Technology Resources web page, at http://www.calvoter.org/votingtechnology.html.
Many people have been wondering how much money, if any, the state of California will appropriate for voting technology modernization. The Governor's budget requests $40 million -- far less than Hertzberg and Jones are seeking. Given the state's energy crisis, some are worried that no money will be available for voting technology modernization. The ACLU's lawsuit may help apply some of the pressure needed to make the financial resources available that counties need to modernize voting technology.
Even if funding from the state or federal government is forthcoming, many county election officials say they do not have enough time to change their technology before the March 2002 election. There is also no consensus about what voting technology works best. Though the chorus calling for an end to punch cards grows louder, California's county election officials are of different minds about what should replace them. Some want new touch screen systems while others want optical scan systems.
One reason some election experts are reluctant to move to touch screen is because it is still a relatively new technology and there are many important policy issues that must be considered when we start using computers to vote. Many of these issues are outlined in my "Ten Things to Know About Voting Technology" essay, but briefly, they include the absence of a paper trail, the reduction of transparency in the voting process when we move from paper to computers, and the debate over whether voting technology software should remain proprietary or if the software should instead be public source.
Based on my own research, I have come to the same conclusions that the ACLU has: voters who vote on Votomatic and Pollstar machines stand a better chance of having their votes discarded than voters using other voting systems. This is an issue that I have been concerned with for the past two years, and in fact I raised this problem while I was serving on the California Internet Voting Task Force. Any computerized system precludes overvoting altogether. To use a mix of systems in elections that cross county lines creates a situation where all voters' votes don't stand an equal chance of being counted.
The easiest way to see the difference between performance of different voting systems is to look at the percent of "votes not cast" in a race. This is the combined total of overvotes, or spoiled ballots, and undervotes, which includes intentional nonvotes as well as votes that don't register because of deficiencies in the voting technology.
I've been studying the "votes not cast" in the November 2000 presidential election in California and here is what I've found:
* optical scan systems had an average 1 percent "votes not cast" rate;
* punch card systems had an average 1.6 percent "votes not cast" rate; and
* the one touch screen system used in 2000, in Riverside County, had a .9 percent "votes not cast" rate.
However, I also found a wide range of "votes not cast" within each system. With the Votomatic machines, Los Angeles county had the highest rate of "votes not cast", at 2.7 percent, while Shasta county had the lowest rate, at 1.3 percent. Another punch card system, Datavote, used in 21 counties, differs from Votomatic and Pollstar in that the candidates' names appear on the voting card. With this voting system I also found a wide variety of "votes not cast" -- Colusa county had the worst performance, with 3.2 percent "votes not cast", while El Dorado and Calaveras counties had the best performance, at .6 percent (outperforming even Riverside's touch screen system).
There is one matter that everyone involved with election reform can agree on: voting technology is by no means the only factor playing a role in the reliable performance of a voting system (as is evidenced by the range of performance between counties using the same voting system). Some of the other factors that impact a voting system's performance include:
1) voter education;
2) the age of voting equipment;
3) the maintenance and storage of voting equipment;
4) pollworker training;
5) quality control of ballot cards;
6) ballot design;
7) pre-inspection of ballots before they are inserted into ballot readers;
8) guidelines for determining voter intent; and
9) the utilization of ballot readers at precincts that facilitate
"second-chance" or "do-over" voting to correct overvotes.
Monday's Assembly Elections Committee hearing is the last chance for the three voting technology bills to pass their first legislative hurdle if they are to advance this year. The hearing begins at 4 p.m. in room 444 of the State Capitol. FYI, if you can't make it to the hearing, you should be able to watch the hearing via the Internet through the California Channel's web site, http://www.calchannel.com/daily.htm, which is carrying live feeds from the legislature's in-house television service. (I was thrilled to watch yesterday's redistricting hearing online, though the audio feed was not very clear.)
More information about the ACLU's lawsuit is available from today's Los Angeles Times, http://www.latimes.com/news/state/20010418/t000032826.html,
Contra Costa Times, http://www.contracostatimes.com/news/california/stories/chad_20010418.htm,
and Sacramento Bee, http://www.capitolalert.com/news/capalert04_20010418.html.
The ACLU's complaint is available online in PDF at http://www.aclu.org/court/jones.pdf
That's it for now. Stay tuned....
-- Kim Alexander, California Voter Foundation
firstname.lastname@example.org, http://www.calvoter.org, (916) 325-2120
|Back to CVF-NEWS index|
|Subscribe to CVF-NEWS|
|Contact Kim Alexander|
This page was first published on April 18, 2001 | Last updated on April 23, 2001
copyright 1994 - 2001, California Voter Foundation. All rights reserved.