A precinct warden had told our volunteer that she wouldn’t need to observe the count after polls closed, because the poll workers didn’t need to count the votes, just run them through the machine and then send them to city hall. It wasn’t the first such call. Volunteers from at least six precincts called us in the hours before 8pm, worried because local elections officials had told them the precinct would not hand-count the vote. I patiently confirmed our volunteers’ understanding: yes, poll workers needed to hand count the write in votes after polls closed. In one case, the warden insisted the machines count count write-in votes. For a couple of them, we had to call Boston’s elections department, who called the polling place to tell them they had to count the votes. One precinct didn’t change their mind and decide to count until nearly a half hour after they closed. And one warden wanted to ask us how to count the votes.
Getting the tallies on election night is always an emotional rollercoaster, as the staff tries to extrapolate from each return. “Oh no, we’re hosed” when a particularly disappointing result from a supportive precinct comes in; then “wow! now that’s a margin!” a bit later when a much better than expected tally comes in from another precinct. That’s normal, but this election night featured an extra element of confusion. With almost every tally report, I got another story.
In Public View?
The most common trouble was not being allowed to observe the count. The law describing how to hand-count votes says “the ballot box shall be opened by the presiding officer and the ballots taken therefrom and audibly counted in public view, one by one” [M.G.L. 54-105] Although the law regarding precincts where machines are used (M.G.L. 54-105A) is not specific about how to hand-count, the intent in section 105 makes sense: people should be allowed to watch the hand count, to see how it’s conducted, to ask questions, and to point out ballots for reconsideration if they think something was missed.
We had designated one volunteer each in most of the polling places to do this, but many were unable to. One was told he had to sit on the opposite side of the auditorium; several were told to wait outside the doors. Many were allowed to watch, but not stand close enough to see what was on the ballots. One volunteer who I sent to drive to another polling place in the final half hour because we had nobody else there to observe the count, was able to watch at first, but when he asked questions, he was kicked out. Poll workers, he said, were very accomodating, welcomed his questions, and tried to do a good job, but a police officer took exception and forced him to leave the premises.
By What Standard?
From the polling places where our volunteers were able to stay inside for the count, we got reports of a wide variety of counting practices. One polling place, for example, called in two sets of tallies, some time apart. At first, they had only counted ballots where voters had filled in the oval next to the write-in spot for state senator, but they realized there were a number of ballots with names or stickers in the write-in spot but without that oval filled in, so they decided to count again including those. It was the right thing to do, but how many other precincts never counted the ballots without filled-in ovals?
We’ve had some discussions about how Boston might count ambiguous write-in votes. The standard should be: can you determine the intent of the voter by looking at the ballot? If you can, count it; if you cannot, don’t. Well, some polling places had a novel approach to the problem: pass the buck. The report from one polling place, for example, was that they had tallied the ballots that were obviously correct for any of the candidates – sticker affixed in the right place, oval filled in – but set the rest aside in a separate pile. Rather than try to determine which candidate those ballots were votes for, they simple sealed them and sent them to city hall uncounted. There were approximately 50 ballots in that pile.
In meetings before the election, the Chang-Diaz campaign tells me, we were told that a series of trainings was planned for poll wardens. But on election night, wardens didn’t just have different interpretations on the more esoteric matters, but also on simple things like allowing observers to watch the count. And many weren’t even clear on the most basic matter of all: the need to actually count the votes! The next day, Bonifaz called the elections division to ask what standard poll workers were instructed to use when counting votes. Boston has not yet provided the answer – not because they don’t want to tell us, but because they need to figure out what the answer is so they can give us a truthful answer. Were all the poll workers and wardens trained to handle this election? I can’t make myself believe that.
If I had to guess the average number of questionable ballots per precinct – ballots not counted, ballots counted by incorrect standards – I would estimate at the very least ten, and possibly as high as 40 or 50. With 73 precincts, that’s a lot of variance. As things stand now, more than 24 hours after the end of the election, Sonia Chang-Diaz and Diane Wilkerson are separated by just 141 votes out of about 11,500 tallied. When all the votes are counted, I would not be surprised if the margin were 1 vote… or 1,000 – for either candidate.
Now is the time for Boston, and Massachusetts, to do the right thing: Count all the votes, using the “intent of the voter” standard uniformly, in a process open to public view, to give us a result both trusted and verified, so we know who was legitimately elected. Count every vote.
(crossposted at johnbonifaz.com)
From How to Request a Recount, on the Secretary of State’s website:
<
p>
which is also consistent with this:
<
p>
<
p>
If this rule was not followed, then the votes were not correctly counted.
The precinct that realized their mistake and counted again, including those without ovals filled in, did the right thing, but others did not. The “intent of the voter” standard is a good general rule to help resolve situations like this even if they’re not clearly written down. If a voter put a candidate’s sticker on the ballot, then obviously that voter intended to vote for that candidate (unless they also put on a competing candidate’s sticker) – you don’t need to see a filled in oval to make that judgement.
<
p>
If this rule was not followed,
<
p>
At some precincts it was, at others it wasn’t.
<
p>
then the votes were not correctly counted.
<
p>
This I can say for sure: the votes have not yet been correctly counted.
I was at a polling place while the write-in votes were counted. Though the poll workers were trying to do their best, it was pretty clear they didn’t have instructions on how they were to go about counting the votes. The warden didn’t let the observers view the actual ballots while counting, instead they had us sit on the side of the room. They agreed to read the names out loud after they had been counted by two workers performing the “official” tally. While two workers did the “offical” count silently another worker read out the name on each ballot and separated the ballots into piles for each candidate. The tally of the two workers was different than the tally of the ballots read aloud.
<
p>
Some other workers then found a few more stacks of votes that needed to be counted and included those in the piles. It was unclear where those ballots had been or why they weren’t included in the initial count or even why they had been separated in the first place. They were counted and then added to the total that had been tallied from reading the ballots aloud, not the number that the two workers doing the “official” count had tallied!
<
p>
One of the workers, who appeared to have some experience, said something along the lines of don’t worry if it isn’t right, you’ll drive yourself crazy trying to get it exactly right.
<
p>
At our polling place, 448 ballots were cast. The computer reported that 218 of those ballots had a write-in vote, while 230 did not. The final tally was 83 for Wilkerson, 111 for Chang-Diaz, and 12 for other candidates. That totals 204, where the other 10 votes went is unclear.
<
p>
Overall, the counting was confused and demonstrably inaccurate. While the difference in tallies were small, that discrepancy leads to larger implications. The fact that there was any inaccuracy at all calls into question the results of the this particular polling place. The inaccuracy of the results from this polling place call into question the results from across the district. It would be irresponsible for the results of this election be certified without a recount.
they are usually kept separate and tallied in the end
<
p>
of course, nothing is usual here, it all starts with the necessity of even having a writein campaign in the first place
ballots from the “left” or “right” side of the machine. For some reason, the machines place ballots in two stacks at the end. Poll workers at my precinct (5-7) didn’t appear to know what if any was the distinction, and they in fact counted the two stacks separately and only merged them at the end.
Do they, for instance, all receive a copy of How to Request a Recount? Is this document available at all polling places while counting is done?
They’re getting about $minimum wage. They work the following:
<
p>
Presidential Primary (every four years)
State Primary (every two years)
State General (every two years)
Muni Elections (either every two years or very year if a town)
<
p>
So that come up to about FIVE state elections in FOUR years
AND two to four elections on the muni side.
<
p>
Long hours, low pay, lucky to book 30 to 40 hours of work a year!
<
p>
You aren’t going to get people interested in the ins and outs of the law or the procedures beyond the taask at hand. Maybe the local Election Warden is versed, but most of the workers have simple repetitive tasks. Even if they were issued that book, they don’t have time to read it on the job AND no one would pay them to take it home and learn it. PLUS there is no guarantee you are even called back the next election! (no job security).
The exact same thing happened in 12,8 and 9 and 11, 3. The machine tally was much less than the actual vote tally (we later discovered). Plus, the warden would only let Wilkerson’s counters see the numbers – refusing to show us.
There are also many more specific procedures in Massachusetts CMR.
<
p>
While everything you say is correct, you are overlooking what is known as SOP. (Standard Operating Procedure, no not SOB).
<
p>
The reason machines were “invented” was to speed up the count and thereby AVOID the long process of hand-counting paper ballots.
<
p>
SO most of the places that are using “machine read” paper ballots, still as part of SOP, do the election night count with the machine.
<
p>
<
p>
What that implies, is that you need to understand that in sticker campaigns, many votes will remain uncounted, UNTIL a re-count has been requested. If you request a machine re-count, not much is going to change. If you request a hand re-count, then you ALL sides get to pick up votes, based on “the intent on the voter.”
<
p>
The paradox here is that the candidate that has the people LEAST LIKELY to follow voting procedures will pick up the most votes. That is because these are the people that negelected to fill in the oval election night.
<
p>
Since I’m not sure the types of machines involved in these precincts, I’m not sure what the SOP of Boston Elections is.
<
p>
But one thing is clear to me. Boston Elections KNEW what the SD was, and should not have gone home until they had RESULTS from each and every precinct in the SD on a TALLY SHEET.
<
p>
I know from experience, that unfortuately election workers are in a very powerful position, noo matter how much intelligence they have,m they strongly RESIST any assistance in trying to HELP solve potential problems during the election and count.
<
p>
The bottom line: Low pay, low levels of training, very low frequency of work, not enough qualified supervision, what do you expect? I hope you’re not trying to pin this on Bill Galvin, because even “State Elections” are run locally by municipalities…
Where did that quoted section (“As far as I know, there is NO mechanism…”) come from? Link to the source?
And I’ll stand by it, there is no way a hand “re-count” can be requested the night of the election. Ballots, including absentee ballots are fed into the machine, precinct by precinct and totalled up.
<
p>
First off, even the campaigns themselves, don’t have standing to be able to force re-counts except by the provisions of MGL (Laws) and/or CMR (Regs).
<
p>
Were the election run in “East Overshoe” the race would be on paper and of course they would be counted by hand.
The real question is, why count these ballots by machine at all, when they are guaranteed not to contain any machine-readable votes (for this office)?
The same ballots contain machine-readable votes for all the other offices, so it does make sense to run them all through the machines first. After that, of course, they should all have been hand counted, for write-in votes. The fact that there was any debate or confusion on this point really boggles me.
The idea was to machine count the Gov’s race first, and then handcount the Senate race?
<
p>
Bigger Question
<
p>
….Why in an election where THERE INHERENTLY MUST BE CONFUSION — an extremely rare scenario combining ALL WRITE-ins plus TWO DIAZ’s — is there not a system to PRESCRIBE THE RULES publicly, with a SINGLE STANDARD, and in advance?
<
p>
Voter intent is obvious on Wilkinson, Wilks, Deb Wilkerberg, etc. No one will contest those.
<
p>
But almost anything that is not PRECISELY “Sonia Chang-Diaz” can be challenged as intent for Samiyah Diaz, no?
<
p>
I.e., why not have a sheet in advance that says:
<
p>
<
p>
Or whatever. The list above would disadvantage Samiyah Diaz. But whatever the decisions, so much better to have them made in advance than having dozens of different standards determined by some random poll workers.
<
p>
If only Galvin clucked on, sorry, CLICKED on BMG a few weeks back. You would think David’s chicken cartoon would have egged him on.
<
p>
to wit: determine the intent of the voter, if possible. Count it if you can, otherwise don’t.
<
p>
That’s it. The suggested mechanical rules (don’t count x ever) would only cloud that standard and could lead to counterintuitive results.
<
p>
The intent of the voter has been the psephological gold standard for more than a century (though the Republicans trashed it in 2000).
simply means someone has to decide.
<
p>
right now, that means lots of people decide. somia diaz at one precinct counts. somia diaz at another does not count.
<
p>
this was the rare situation where a mechanical decision done in advance would have helped.
A mechanical rule in this case might mean that a ballot marked “S. Diaz” does not count even if the words “Chang Diaz” are clearly written elsewhere on the ballot.
<
p>
You can refine the rule to rule out that outcome, of course, but you can’t anticipate every possible thing. (Who could anticipate “Somia Diaz?” Do you have a mechanical rule for every possible combination of alphabetical characters before the word “Diaz?”)
<
p>
So a qualitative standard–intent–is superior to a strictly empirical one.
<
p>
The remedy for conflicting interpretations of intent is procedural: there should be an appeals process that puts all the “Sonia Diaz” ballots before the same judge.
Writins are always going to be problematic in any system. That is why you need to get the signatures and get on the ballot. It is not a difficult thing to do.
<
p>
One person was obviously incompetent and couldnt get her signatures and the other took that opportunity to capitalize.
<
p>
This is what they should have expected.
Forget about “feeling bad for either candidate” – the reason votes need to be counted is that we live in a democracy and democracy is the system we want to preserve. Counting the votes is a critical piece of that. The voters voted and they should get the candidate they legitimately elected.
<
p>
<
p>
Most emphatically NO. “Could” have expected, perhaps (though really, this is far worse than we could have expected). But we should expect, and demand, an accurate count of all the votes.
<
p>
(Leaving aside that by “they” you probably mean “the candidates” – that’s missing the point. I mean the people of Massachusetts.)
Because right now some people are giving their own time, energy, even money, to lead on the issue of election reform–in a situation that is not theoretical but very real. I suggest you take a moment and pick your category.
(This comment about my experience with vote counting in the Second Suffolk race is reposted from an earlier diary on this topic, in which a blogger asked whether the 8 uncounted precincts in the State Senator’s race could really have been a good faith clerical error as Sec. Galvin described it. I monitored the poll closing for the Chang-Diaz campaign at Boston precinct 5-7, and also witnessed some of the closing at 5-8 which was at the same location, the Public Library at Copley.)
<
p>
…”clerical error” might not be totally inaccurate. Let me tell you that in the precincts I watched close, which were Boston 5-7 and 5-8, the performance and the level of knowledge of the poll workers was very disturbing. Those I spoke with seemed to have very little understanding of the nature of the race, for starts. It took a long time for it to sink in that there were no regular votes in the race, only write-ins. No one had apparently briefed them on this. Even worse, there was no procedure in place whatsoever as to how to count the ballots. I am not talking about how to judge the questionable ballots (“S Diaz” counts for whom, and so forth); I am talking about a basic procedure for how to collect all the write-in ballots and count them out for each candidate. The poll workers at 5-7 had to do this twice, because they went through the count once and then concluded they had done something wrong, so they did a lot of head-scratching and ended up recounting everything.
Finally…and the reason I say the 8 precincts of uncounted ballots could well be a clerical error … is that the poll workers at 5-7 were expressly saying regarding some of the write-in ballots, “We are not going to figure this out; they will do it at City Hall.” As best I was able to observe it, they made no effort to evaluate any ballots whose marking deviated from the norm (sticker in the wrong place; candidate name written unclearly; etc). They were expressing the sentiment that they would send the ballots to City Hall, where a more thorough count would happen. They repeated this so many times, I left with the impression that the counts at the precinct were in fact preliminary. I now believe that in fact they would be final, unless someone called for a recount. Yet, the poll workers at my precinct were not even remotely implementing the procedures for judging write-in votes, due to the apparently completely incorrect belief that “City Hall would figure it out”.
<
p>
Therefore, if another precinct (or several) had decided not to count the write-ins at all, because “that will be figured out at City Hall” or something like that, I would not be surprised. Based on the confusion I observed, I’d say such an error could well have been possible due to a lack of education about the race and proper procedures provided to the poll workers.
here are the latest numbers from the recount.
<
p>
http://www.boston.co…
Please remember that this is not a “recount” – these are votes that haven’t been counted in the first place. Also, here’s something very troubling:
Actually, no, this count of the 8 precincts will not determine the winner. We still need to count the votes from the other 65 precincts that did submit tallies, but used a variety of standards, and in many cases deliberately did not count some of the ballots, and/or came up with tallies that didn’t add up.
<
p>
It’s important for us to counter the idea going around the press that somehow once these 8 precincts get counted, we’ll know the result. We absolutely won’t, and it’ll be tragic if the city thinks their job is done and the public thinks the votes were counted when they haven’t been.
<
p>
We still have 65 more precincts to count and recount.
Uh, Cos, remember, there’s actually a contested race in November for a change. The kicker question is whether Sonia’s and John’s supporters will cast a protest vote in favor of Samiyah to dislodge Dianne from the seat and essentially open the seat to new challengers two years hence (or is it “thence”?). No way Samiyah keeps the seat even if she wins in November, simply not possible. Ironic, that voting for Samiyah might be the only way to get a forward thinking Democrat like Sonia (or John Bonifaz who also lives in the district) into the Legislature. About half of Democratic primary voters opted to retire Dianne from the Senate. If those numbers hold, I wonder how the Independents and Republicans will vote in November?
it will be hades freezing over before sonia’s supporters vote for a republican. they will just stay home, or more precisely, skip that line on the ballot.
The November election will bring a lot of voters to the polls in the contest between Deval Patrick, Kerry Healey, Christy Mihos, and Grace Ross. Some of them will have strong opinions about Wilkerson & Chang-Diaz, some won’t. But that district is heavily heavily Democratic (and I include a lot of “unenrolled” voters in that – just because they don’t register as Democrats doesn’t mean they don’t vote that way). It’s a general election and the party label will be on the ballot: a Democratic nominee vs. a Republican nominee. The Democratic nominee will win by a wide margin, partly helped by a lot of votes coming from people who are voting for the party rather than the candidate, or who are ambivalent about the candidates so use party affiliation as the heuristic to make their choice.
Just as a reminder, we’re talking about an area where our politicians are people like Maura Hennigan and John Bonifaz. My local city councilor is in the Green-Rainbow Party. And Kerry Healey got a whopping two votes in my precinct in the primary.
<
p>
Samiyah Diaz’s chances of winning in November are roughly as good as mine and I’m not even running.
http://www.boston.co…
Even though it’s looking less likely that you will win, I hope you still press ahead for the full recount. Once you collect the required signatures, does it cost the Chang-Diaz campaign anything to request one?