Again, every once in a while I run across neat little ideas that election officials develop and implement, and I found one this morning. Warren Slocum, from San Mateo County in California, has put a neat little flash presentation on his website of the election administration cycle — all of the little things that have to go on for an election to happen. Here is a link to the page and the demo. This is a helpful device to help show folks the complexity of election administration, and now whenever I am asked questions about the election administration cycle, I’ll just send them the link to San Mateo County.

This opens the door for a lot of other interesting ideas, as to how election officials (or those interested in voter education) should think about new, and more interactive, ways to allow voters to learn more about their election process. Another example are the flash demos that were developed for the Buenos Aires e-voting pilot project, which are available at the site we have put together including documentation on the project. This of course brings me back to the general issue of voter education, and it is clear that there are lots of new and interactive ways to give voters the information they need for effective political participation.

We received a call for proposals (RFP) today, for a project run through the George Washington University Graduate School of Political Management, in association with the Pew Charitable Trusts.

Here is the letter we received:

Dear Friends and Colleagues:

I am writing with exciting news…

To build on the young voter turnout success of 2004 and 2005, promote new and creative approaches to get young people to register to vote, and keep the youth vote in the spotlight in 2006, The George Washington University’s Graduate School of Political Management (GSPM) announces a nonpartisan national competition to identify and support innovative and replicable strategies for registering young people ages 18 to 29. Funding for this competition is provided through a grant from The Pew Charitable Trusts.

GSPM invites proposals from nonpartisan 501(c)3 organizations interested in testing voter registration efforts with young voters. We will award grants to a group of eight to ten organizations working with different subsets of the youth population and employing diverse, but strictly nonpartisan, registration methods. Winners of the award will receive grants ranging from $50,000 – $250,000.

Applications must be received by 5:00 pm on January 13th, 2006. Winners will be announced no later than March 1, 2006. Please see the attached RFP for details and pass it along to others that might be interested. Feel free to call with any questions.

Best wishes,
Heather

Heather Smith
Director, Young Voter Strategies
Graduate School of Political Management
The George Washington University
805 21st St, NW, Suite 401, Washington, DC 20052
202-994-5052 (ofc)
720-252-3932 (cell)
hsmith@gwu.edu

There are three documents associated with the RFP, which we have placed here:

1. The RFP itself.
2. Financial forms.
3. The cover sheet.

There are reports today that a tanker truck filled with completed ballots was stopped at the Iran-Iraq border. The New York Times reports that at least three other tanker trucks made it inside Iraq with their loads of ballots. As the paper notes,

The tanker was seized in the evening by agents with the American-trained border protection force at the Iraqi town of Badra, after crossing at Munthirya on the Iraqi border, the official said. According to the Iraqi official, the border police found several thousand partly completed ballots inside.

The official, who spoke on the condition of anonymity because he was not authorized to speak publicly, said the Iranian truck driver told the police under interrogation that at least three other trucks filled with ballots had crossed from Iran at different spots along the border.

The official, who did not attend the interrogation, said he did not know where the driver was headed, or what he intended to do with the ballots.

Clearly, this report undermines the confidence in the election outcome if there are reports in the voting of ballot box stuffing. It also illustrates the difficulty of stopping fraud when paper ballots are used. International elections often use clear ballot boxes, but if the box is stuffed–even knowingly–it is almost impossible to then sort the stuffed ballots from the properly cast ones. As a story in Slate.com noted in 2004, the reason many countries are moving to electronic voting is because it can deter this type of fraud. In India, bandits often take over polls and steal or stuff the ballot boxes. Electronic voting has greatly limited this activity.

The California Secretary of State’s office has just released the presentation materials from the “Voting Systems Testing Summit”, for those presenters who had electronic presentation materials.

Some highlights from my panel include:

1. Henry Brady’s presentation, which began with a critique of the widespread focus on security problems with precinct voting systems. This is detailed in the first few slides from Henry’s talk, where he pointed out that discussions of risk ought to include mention of both vulnerability and threat, not just vulnerability. But the main thrust of Henry’s presentation was to talk about voting system accuracy in California, focusing on a variety of residual vote analyses. One of the methodological issues that arose during Henry’s presentation was his use of data from the recent 2005 special election, and whether his use of that election (which involved only ballot measures on the statewide ballot, and no statewide candidates) to compute residual vote statistics using the lowest residual vote estimate across the eight ballot measures, produced a metric that is indeed comparable to other elections like the 2004 presidential election.
2. David Dill’s discussion (he didn’t use electronic props for his talk), in which he called upon Secretary of State McPherson to take a leadership role in security testing of voting machines.
3. Avi Rubin’s discussion (he also didn’t use electronic props), in which he had the interesting idea to develop a “DARPA” competition for testing the security of voting systems; the idea here would be to develop a contest, with a large monetary reward for the achievement of clearly specified ends, to get individuals and groups interested in thoroughly testing voting systems.
4. Michael Shamos’s presentation, where he first argued that the existing “Independent Testing Authority” process was broken. But the most interesting part of Michael’s presentation came in the second half of his slides (see slide 8 and beyond) where he developed an interesting typology of “voter verified paper audit trail” (VVPAT) systems. I found Michael’s typology to be a very clever way to think about VVPAT systems.

Of course, there were just a few of the things I thought were of interest in our panel; each of the sessions had useful presentations and discussion. My understanding is the the Secretary of State’s staff is preparing a report from this summit; it will be interesting to see what they took from this conference, and what recommendations they make for how California might move ahead into the brave new world of voting system testing.

A headline in today’s electionline.org “Electionline Today” concerns a pact signed by the secretaries of state in four midwestern states — Missouri, Kansas, Iowa, and Nebraska — to create a task force to figure out how they can check voter registration lists across the four states in order to clean up their registration lists. There is a story in the Kansas City Star about this pact, including a few quotes from the relevant actors. The text of the pact is also available.

One obvious requirement for these four states to be able to engage in registration list comparison will be data exchange standards, something that Thad and I have been talking about for a few months now, based largely on the study that we recently published through the IBM Center for the Business of Government, “The Next Big Election Challenge.” The work of these four states might be highly significant, if they are able to successfully develop data exchange standards for their voter registration files; they could provide an early test of the difficulties associated with registration data exchange and comparison — and hopefully provide some guidelines for other states or regions to follow as they work to also exchange voter registration lists.

But there were also four other important components of the pact that have not received as much attention as the registration list exchange proposal. These are:

1. Cooperative training of election officials
2. Cooperative testing of election systems
3. Cooperative improvements for election security processes
4. Establishing uniform protocols for international observers

Given that the these four secretaries of state are trying to implement four reforms (registration list exchange, cross-state processes for testing voting systems, cross-state security planning and testing, and standards for election observers), it almost seems as if they are reading Election Updates!

But seriously, this memorandum of understanding between Missouri, Iowa, Nebraska and Kansas might be a highly significant development in the long run, if they make it work. There is much to be gained if states can work together in these areas, especially voter registration data exchange, and the testing of voting systems. As the latter constitute much of the theme of the recent “Voting Systems Testing Summit” in California — and was something that I speculated on in an earlier essay — I would not be surprised to see other neighboring states in the near future try to develop similar cooperative ventures in these areas.

As attention has focused on election administration, and many outside observers have begun to routinely monitor the election process in the United States, one practice that has begun to receive some attention has been the “remarking” or “enhancement” of ballots by election workers. In some jurisdictions, this is a common practice; depending on the exact voting technology being employed, in some jurisdictions election workers will somehow act to try to insure that the voter’s intention is easily read by a tabulation device.

One jurisdiction that has been proactive with this practice is the Los Angeles City Clerk’s office, and their “remaking” of InkaVote ballots in 2005 raised a minor controversy during the Los Angeles mayoral election. In early observation of election-night activities back in the days the City Clerk was still using prescored punchcards, we observed election workers actively removing chads from the prescored punchcard ballots right after they were coming out of the sealed ballot boxes in their initial examination of the ballots.

There is an interesting story in the Seattle Times about this practice in recent Washington State elections. The story goes into great detail as to the problems that election workers are trying to fix, expecially with paper ballots:

Some mark their ballots with red pens, highlighters or hard-lead pencils the scanners can’t read. Others circle candidates’ names, or make “X’s” or checkmarks next to them instead of darkening the ovals. Still more write in names not listed on the ballot, but they don’t fill in the corresponding oval to identify it as a write-in vote.

Some write editorial comments on their ballots — Logan says the Seattle monorail was a popular topic this year — that can stray into ovals or the tracking marks along the ballots’ edges that scanners must read to register votes correctly.

In Snohomish and other counties where voters mark their choices by drawing a line to connect two halves of an arrow, elections officials say they frequently must duplicate two-sided ballots on which voters have drawn lines so heavily that the ink bleeds through to the other side.

Wendy Mauch, Snohomish County’s elections supervisor, says voters will be asked to use only pencils in future elections.

But one of the more interesting aspects of this particular story is the data provided about the extent to which ballot remarking goes on in Washington State. In a graphic associated with the story, the following counties are listed, with their respective rates of ballot remarking (County, number remarked, percent of ballots):

1. Clark: 14,000 (13.8%)
2. Whatcom: 8,479 (13.3%)
3. Kitsap: 9,218 (11.4%)
4. King: 45,468 (8.3%)
5. Snohomish: 14,085 (8.0%)
6. Thurston: 5,051 (6.6%)
7. Yakima: 2,872 (5.4%)
8. Pierce: 8,739 (4.7%)
9. Spokane: 5,000 (3.7%)

Both the Clark and Spokane numbers are estimates.

These percentages are surprising, and the first time I’ve seen estimates of the rates of ballot remarking. If remarking practices like these are widespread, it does indicate that we need to think about the implications of these administrative practices for our measures of voter error, because if numbers like these hold for other jurisdictions this indicates a much higher rate of voter error for paper ballots than has been found in recent research. But as a proximate matter, we clearly need to gather more information about the extent of such practices in jurisdictions using paper ballots, both the regulations surrounding these practices and how many ballots get remarked.

Another important lesson that I learned during election-day precinct monitoring over the past few years is that there are many good things that occur during any particular election, the result of people thinking through past problems and issues and trying to devise innovative solutions to those problems.

I saw a number of examples of innovations and good thinking during California’s special election, the highlights of which I thought I would present here.

1. Improving polling place access. I’ve already discussed the problems of parking and accessibility, especially at schools that were in session on election day. But there were also people actively trying to alleviate or solve those problems, for example at the Corona Fundamental Intermediate School (Corona, Riverside County, CA). The head precinct judge in this location had brought along his own traffic cones to mark off handicap parking as close as he could to the entrance of his polling location, and had worked with students in the school to develop directional signs to help voters figure out which school entrance to use to get to the precinct. Here was a person who recognized a problem with a polling place, and who worked to mitigate it (he is the guy in the cap in this picture).
2. Improving the use of voting technology. In Riverside County, a jurisdiction employing the Sequoia AVC Edge touchscreen voting device, the local county officials provided to each precinct a handful of “stylus” instruments for voters to use. The “stylus” was just a standard looking No. 2 pencil, but with an eraser at each end. Unfortunately, I was unable to get a photograph of this “stylus” on election day, nor have I been able to locate much information yet about where Riverside County got this idea from or where they procured the “stylus” from. But as a tool to help mitigate calibration issues with their touchscreen voting devices, and to help mitigate the build-up of residue on the touchscreens, this is a good idea. The “stylus” also can help improve voting system accessibility for certain voters who might have trouble touching the screen.

   But there was an interesting twist in the use of the “stylus” in Riverside County. I had never seen this employed before there, and it was clear that the precinct workers were not certain about how they were to deploy this voter aid. In some voting locations in Corona, precinct workers provided the “stylus” to every single voter who used a touchscreen voting device. In other locations in Corona, they provided the “stylus” to voters who the polling place workers thought might need the instrument (typically elderly voters).

   And there was at least one location in Corona where I noticed the polling place workers never gave out the “stylus”, even though there was at least one elderly voter who I observed having great trouble with the touchscreen voting system; after observing the voter in distress, the polling place workers did say verbally to the voter that they could use a “stylus”, but the workers made no effort to physically provide the “stylus” to the voter having difficulty with the touchscreen voting device and did not even get up from their chairs to help this particular voter.

   While an interesting innovation, clearly they need to figure out how to make sure precinct workers deploy it effectively in the future.

3. Improving the voter registration process. In all of my polling place observations over the years, I’ve never had the opportunity to witness the use of a voter registration database, on election day, in precinct voting. I did see, in Los Angeles County, the use of a laptop computer to access the voter registration database in one consolidated voting precinct in Pasadena. According to the precinct workers, it had been very helpful throughout the day, whenever there was a question about a voter’s registration status. This of course is a reform that the Caltech/MIT Voting Technology Project called for in 2001, and is something that I suspect we will see more adoption of as statewide voter registration files become prevalent after January 1, 2006.
4. Improving openness and transparency. As I wrote about just before the special election, Orange County (CA) embarked upon a series of novel uses of new technology to improve openness and transparency of their election process. They used novel webcams, and allowed anyone to use their website to track ballot boxes. An interested individual could use those innovative webpages to see ballot boxes as they arrived in the election headquarters on election night.

   This is a great use of new technology to increase the openness and transparency of the election process. I’d advocate that election officials think seriously about using similar technologies in coming elections, including live coverage (through a webpage) of all stages of the pre- and post-election process. Why not webcast the pre-election logic and accuracy testing of voting devices? Webcast all election-night activities, including ballot box arrival, unsealing, sorting and examination of all material from voting locations, and tabulation? Use a webcam to cover any post-election activities, including examination and tabulation of provisional and absentee ballots, any recount activities (mandatory or not),and the final canvass. And post all of the videoclips of these activities on a website somewhere, so they can be downloaded by anyone in the future for any reasons. This is not a perfect substitute for allowing direct election observation by interested individuals, but it can allow those who for whatever reason could not make it to election headquarters to observe these activities in person.

The bottom line is that while we continue to find things happening on election day that are problematic, we also continue to see many innovations and examples of people working to solve problems. We all need to work to identify these innovations and solutions, and to get the word out when we see new ideas being implemented to solve old problems.

John Mark Hansen’s presentation this morning in the National Academy workshop brought to the table some fundamental questions about election reform — what are the basic normative or democratic principles we are trying to achieve with election reform? In his brief presentation, Mark mentioned three different basic normative principles: fairness, equity and legitimacy. Undoubtedly there are other basic principles we might consider when undertaking election reform.

This is an area that our colleagues in normative political theory can certainly help us with, but I’m not aware of any initiative to engage normative theorists in these discussions. What basic normative principles should we consider when we talk about election reform? Are fairness, equity and legitimacy the only core principles to focus on? Are there others to consider? How (and should) we order these principles? Are there trade-offs on these principles, or can we develop reforms that can achieve positive progress on all normative dimensions simultaneously? If there are trade-offs, how do we make these difficult choices? How can we measure and evaluate how various election reforms perform on any particular normative dimension? These questions about first principles are difficult to assess, but we should seek input from normative political theorists in the ongoing debates about election reform.

Since I am sitting across from Mike at the NAS conference, I too want to blog on the panel I was on, which discussed the role of intermediaries in elections. I am going to focus on the overview comments made by Michael Traugott (Professor, U. Michigan), Ted Selker (Professor, MIT) and then the presentation that I did.

Traugott noted that there are two issues in elections. First, how do candidates get access to the media, given its importance for informing voters? This is a major issue that has been discussed by campaign finance people, who often discuss issues of public financing of elections.

Second, how does the public learn about elections, especially problems in the implementation of an election? Traugott noted that the problems that occur with or in the electoral system are likely to be mediated by a third-party—especially the media—instead of being experienced directly. Consider this: an individual voter has a low probability that she will be affected personally by a problem in polling place. Instead, a voter is likely to experience problems through media reports about election issues and problems. Acknowledging that these problems are learned through a mediated setting increases the importance of the media and their describing the problem correctly. Also, there is good reason to believe that this mediated effort is related to trust in the system.

Traugott notes that there are causal relationships between news and trust, affecting by how the media frame the news. One issue is that reporters have very limited, general skills in dealing with issues of science. In the area of elections, we might ask, for example, what is an error in the election process? How do you explain the occurrence of a problem to the lay public? How do you do this without bias? Traugott argues that elections should be presented as a system, and reporters should be educated about voting, voting processes, and voting technologies. The risks of errors in reporting increase as elections become more complicated with the introduction of electronic voting. Finally, Traugott notes that the blogoshere creates new problems and risks as well. In blogs, information spreads rapidly, raising special issues like the need for fact checkers and truth squads during key times in the electoral process.

Ted Selker discussed the role of intermediaries in elections. For example, poll workers are key intermediaries, as noted in the first picture in his slides, where the poll worker is opening and interacting with the voter-verified paper trail. He also noted that different people spend different amount of time with different forms of media intermediaries: for instance, Asian American men spend 50 percent more time on the Internet compared to all men in the same age category, but spend much less time with radio compared to all men in the category. Ted notes that the intermediaries of the future are likely to be much different than those of today and may change the way people vote. In fact, it is easy to imagine a person today surfing the Internet on a wireless handheld devise to figure out which candidate to vote for in the dog catcher race; you just surf to your local newspaper and find the endorsements.

I focused my discussion on the role of election officials and poll workers as intermediaries (my slides are here). For those of you who read our blog regularly, my talk first focused on the role of information from the election official to the voter, using Larimer County as an example (see here for my blog on this topic). I then talked about how poll workers are critical to the voting process and directly affects the voter’s experience in the election and their trust in government (see this blog for my discussion of the study I have done on this issue). Interestingly, two election officials made comments directly supporting this point. One election official noted that most voter education goes on between the time the voter enters the polling place and the time the voter leaves. A second election official noted that poll workers are “street level lawyers” who interpret the entire election code on the fly for voters.

Clearly, there are a number of assumptions about the role of poll workers in elections and the role of voter education that need to be further tested and studied.

As Doug Chapin from electionline.org has noticed, I am blogging live from the National Academies workshop, “Developing a Sound Analytical Basis for Improving Public Participation and Confidence in 21st Century Elections.” It’s been quite an interesting morning, with a lot of informed and productive discussion.

My panel this morning was on “Fraud prevention and election audits in the new electoral environment.” Fellow panelists were Dan Wallach, Gary Cox, Walter Mebane, and George Gilbert.

My slides are available here, in power point format and in pdf format. Thad’s session is later on this afternoon, on the subject of “Intermediaries” in the election process — and I suspect that once he is done with his session that he’ll put his slides up here as well.

My discussion focused primarily on the issues of making sure that we talk about the entire voting system when we talk about fraud and auditing, to focus some attention on security and contingency planning, and to get us thinking about developing rigorous and thorough protocols for testing of voting systems. These comments parallel and build upon what I talked about last week in Sacramento at the California Secretary of State’s “Voting System Testing Summit.”

During our panel, Walter Mebane presented some interesting results, developing a new method for trying to detect voting device anomalies using data from two Florida counties in the 2004 election (Miami-Dade and Pasco Counties). What Walter does in this analysis is use ballot image data from these two counties, which apparently includes information on the precinct and voting device that recorded each ballot image. Walter than tests to see if the distribution of votes differs across the voting devices used in each precinct, under the assumption that the distribution of votes in a particular race (say the presidential vote distribution) should be the same for each voting device in a particular precinct. If a voting device is systematically malfunctioning, or has been manipulated, it should show up as a deviation from the other voting devices used in the precinct. I’ll talk to Walter and see whether his draft analysis is ready for distribution.

Another interesting thought during the panel discussion came from Dan Wallach, who argued that one important change we could make in the existing testing and certification process would be to alter it from the simple binary “pass or fail” system we now have, to a categorical “grade” format. Of course, voting devices could get a failing grade under Dan’s proposal, but we might get more information about just how close a voting device came to meet certain standards under such a scheme. It might be interesting to consider such a scheme, but where we don’t just get back a single grade for the voting device, but grades for how close the device comes to meeting a whole range of testing standards (imagine a “report card” for the outcome of a particular voting device’s certification process).

More soon …