Wired Magazine has an interesting article titled “Election ’08: Vote by TiVo” discussing the push for electronic voting, citing how internet voting has been used in the US and is used in various countries in Europe.

CNN has the story …

Democrat Joe Courtney’s lead over Republican Rep. Rob Simmons grew to 109 votes Monday, after dipping to as low as 66 votes earlier in the day.

The roller-coaster recount in one of the closest congressional races in the nation has uncovered significant errors in two communities. Preliminary Election Day returns had Courtney winning by 167 votes out of nearly 250,000 ballots cast in the 2nd Congressional District.

Officials in Lebanon discovered Monday that Courtney was given an extra 100 votes over Simmons. “It was human error,” said Lebanon election moderator John Bendoraitis. “It was strictly misreading one number on one machine.”

Computerworld has a story this morning on “vote-flipping”, where a voter selects a candidate on a voting machine but the machine seems to record a vote for another candidate. The bottom line here is that we really need to do some serious research to figure out how frequently this occurs, and what the underlying causes of this phenomenon are.

Count me as another voice in favor of non partisan election administration, as Rick Hasen recently opined in the NY Times, and as Bob Bauer wrote in agreement.

As Bob notes, partisan influence need not be conscious or even present in order to lead one to support this important reform.

Here in the state of Oregon, as in many states, the Secretary of State, Bill Bradbury, is elected on a partisan ballot and runs state elections. I’ve posted many times regarding the excellent election administration in Oregon, what Rick called “squeaky clean” in his op-ed.

But even Bradbury has been accused of partisanship, most frequently when his office has to make a determination over whether a referendum or initiative has sufficient signatures to qualify for the ballot. (Unlike in a vote by mail election, where every signature is checked, the SoS checks a random sample of signatures on petitions.)

There is really no good reason to leave election administration in partisan hands. The depth of partisan divisions, and ongoing public distrust in election outcomes, makes it vital that we pursue this important reform.

I only write for myself here, and not for my collaborators on this blog. But sign me up as one more supporter.

There’s an interesting article in an Elections Canada publication, by Eric C. Bjornlund, “Improviing Vote Count Verifications in Transitional Elections.” Bjornlund is also the author of “Beyond Free and Fair” (Johns Hopkins University Press, 2004), which is a definitive study of international election monitoring. Here’s an abstract of the article:

Despite the increasing popularity of exit polls in international election assistance programs, this article argues that “parallel vote tabulations” (PVTs) or “quick counts” are a more appropriate tool for verifying vote counts in transitional or post-conflict elections. PVTs – often conducted by local monitoring organizations – are generally more reliable than exit polls in political environments characterized by intimidation or uncertainty, as is typical in such elections. Although exit polls can provide insights about voter motivation, the use of exit polls to verify or project results often fails to serve the purpose of true democratization by undercutting the role of domestic organizations. To contribute more effectively to democratic elections, international donors and advisors must better understand these techniques.

Thanks to Susan Hyde for passing along this link!

On the lighter side, the AP is reporting that there might be a rare stamp, from 1918, on an abstentee ballot sitting in a Broward County ballot box … see the Washington Post for details.

By Ted Selker (MIT)

I went to 3 polling places in Cambridge on the day of election. I was awestruck by the organization! In every place clear, huge, and consistent signage led the way to the polling place. In every case, the intake poll book pages were individually laid out without overlapping and taped to a the large intake table, the exit poll book pages were individually laid out without overlapping and taped to the exit table without overlapping. This practice really makes checking off voters easy and transparent. It especially eliminates the ease with which Boston poll workers circumvented the value separate records that make chain voting difficult by not letting a voter leave with a ballot without it being documented in a missing exit poll book entry. In every polling place red white and blue curtains gave voters privacy that is not seen in many polling places. In all places lighting was excellent, in one polling place a bulb hung in each poll booth. Privacy sleeves were used by some voters. The only place I have seen them used consistently was in a polling place in Boston in 2004 where the poll worker handed the voter their ballot in a privacy sleeve.

by Ted Selker (MIT)

During the presidential primary in 2004 a poll worker took the filled out
ballot from my hands to check it (eliminating secrecy in how I voted). I
asked them why they were doing this and they said that they had been told
to turn off the optical scan machine’s over-vote detector so they wanted
to make sure that I hadn’t voted for two candidates. Turning off the over
vote detector eliminates the optical scan system’s approach to second
chance voting. The machine would place such a ballot with two candidates
chosen for president in the bin to be hand read later stopping the voter
from knowing that they should refill out their ballot. I sent the
Secretary of state a complaint note, the over vote detection was turned on
in the next election.

I asked a person at a table on a stage what they were doing. “I am
checking absentee ballots.” I asked have you found any problems today,
they said, I found one and erased it (No one should be in a position to
change another’s ballot at any time during voting).

This year I voted during early voting November 6, and it went smoothly.
While I was there, a man asked to get a ballot to deliver to his son and
bring back by 4PM. His son was a student at Northeastern University, the
poll worker relented and gave the man a ballot hopefully for his son to
fill out (Absentee ballots are thought to be the largest focus of fraud in
this country).

By Ted Selker (MIT)
I watched one polling place in Brookline. The signage was excellent,
keeping canvassers away. The process inside was very organized with
an airport-like public guidance system. Each polling station had integral
lights and the exit/ scanning area was also well marked.

At some point the polling warden took a box of ballots out of the optical
scan machine. They showed me another box behind everything, full of the
nicely organized ballots they had already taken from the scanning machine.

This practice of having an unsecured ballot box that is constantly being
breached during voting day seems standard where precinct counted optical
scan ballots are mixed with scanning of absentee ballots. The absentee
ballots have folds in them which mess up the stacking of the ballots
inside the machine and the machine cannot hold a full day of ballots. I
have yet to see a polling place that has a secure way of transferring the
ballots from the Diebold optical scanners. At this particular polling
place, the warden took the box of ballots to a back room, another
poll-worker then went back into the room. The polling place warden then
came out, got a ballot marking pen and went to the back room again. They
soon came out with the pen. The other person spent at least 15 minutes
arranging the ballots out of site of supervision. The poll-warden visited
the ballot organizer at least twice during this time. I suspect that
everything was as it should be, putting the 200 or so ballots in the same
orientation. I find it extremely troubling that this must be done and I
don’t like it being done without supervision. Without a camera record
made to demonstrate that no marks are added or ballots compromised in such
a process, there is no way to attest that ballots are not being changed.

By Ted Selker (MIT)

This short description documents the variety of practices in Boston that are
worse than surrounding towns and indeed worse than other jurisdictions I have
observed in Nevada, California, Chicago, and New Orleans.

In my next blogs I will describe small problems that can be improved in the
cities of Brookline, Arlington, and Cambridge, Massachusetts

Boston: comparing Nov 2004 and Nov 2006 elections

During the presidential election of 2004, I witnessed active voting at
approximately 30 precincts and watched closing practices at one in Boston.
Observations included a line of 62 people at 7:00AM at the entrance to a
polling place that was actually the entrance to four precincts. The result was
all voters regardless of their particular precinct waited in this line because
one of the polling places checked registration at the entrance, and the other
three precincts had no lines. Although these lines were at the top of at least
a dozen steps, the marking for the polling place was around the back on a door.
We tried to enter the polling place by this more accessible back way and we
found ourselves in a cafeteria; the route to the polling place took us through
an active commercial-style kitchen. After poll workers learned we traversed it,
a poll worker designed and put up signs to help anyone that would vote through
the maze. In doing so, they opened the back door with the polling place sign on
it eliminating any viewable signage of the polling place from the street. The
multiple polling place, single line problem was seen later at other polling
places, with long lines and frustrated voters. Few of the Boston polling places
I visited at that time were accessible. I have detailed this experience in
talks I have given, but will write more about it soon.

During the election of 2006 I witnessed repeats of several problems that I had
witnessed in 2004, many of which were disturbing. At all of the three polling
places I visited, canvassers effectively intimidated voters at the entrance. In
two of them, signs crowded both sides of the entrance gate less than 30 feet
from the polling place entrance surrounded by canvassers. In a third location,
the gate was more seriously crowded by people making the entrance a bit of a
squiggle by them as they shoved literature into your hand. In this polling
place the gate was also the only way to enter the polling place and the
courtyard was about 100 feet walk to the entrance of the building. At one
polling place, I told the canvassers that they were too close, one of them
became angry and told me that she was the past commissioner of elections and
was it bothering me and so on. I approached the polling place warden who asked
me “Is anybody complaining about it” I said I was. The polling place warden
came back from talking to the canvassers and said, “They are upset that you
took their pictures.” The placards were not removed, the people were not
removed and the policeman assigned to the polling place began harassing me as I
stood on the street talking to a colleague in California about details of
electronic voting. The policeman became visibly angry, challenged me to take
his badge number and began accusing me of things. I left feeling very
distressed.

In 2004, I watched one polling place in which a uniformed policeman had taken
over the exit poll book checking operation. I also witnessed anther uniformed
policeman who had taken both intake poll book and exit poll book, placed them
on top of each other and was checking off both books when a voter arrived by
himself without supervision. Certainly this compromises the goal of the poll
book. The specter of having to face a uniformed officer upon entrance to a
polling place to vote is not best practices. I am not certain that the police
actually had training in polling place operations. At all three locations that
I observed in 2006, police officers were responsible for the exit poll book in
Boston; I have never seen this in other jurisdictions around the country.

In 2004 at the end of the polling day, I witnessed two poll workers spend more
than 10 minutes working together to compare the intake poll book and exit poll
book with pencils in hand. They obviously erased and changed various records
with these pencils.

In Boston in 2004 I was struck that the precinct scanner was transported with
its counting module in it, making both the voting machine memory and the
secondary memory in the care of the same person who was transporting the
ballots. I also noted that the ballots were in the same color (manilla)
envelopes as the other election materials. I also was struck that all these
materials were transported by an individual policeman who was part of a strike
that demonstrated in front of the Democratic National Convention a few months
earlier. I have not witnessed such sloppy chain of custody anywhere else in my
poll watching.

The this year I had difficulty in getting in touch with and communicating with
Boston’s election commission. When we couldn?t get responses with email and
phone calls, I visited. The person that came to talk to me began aggressive.
The commissioner did follow up and sent me a letter of introduction so that I
was credentialed through my poll watching for which I am deeply grateful.