| Author |
Message |
    
Bev Harris
Board Administrator
Username: Admin
Post Number: 7473
Registered: 12-2004
Best of Black Box? 
Votes: 7 (A keeper?) | | Posted on Saturday, January 12, 2008 - 5:03 pm: | 
|
New Hampshire's 2008 primary election may prove to be the most fascinating presidential preference race in history.
- Both Democrat and Republican candidates have requested recounts
- More than half of New Hampshire's elections administrators hand count paper ballots in public at the polling place, with a public chain of custody. The rest of New Hampshire's towns and cities use Diebold voting machines to count votes in secret, with a secret chain of custody.
- Hand count and machine count locations, when calculated statewide, show an eerie statistic:
Clinton Optical scan 91,717 52.95%
Obama Optical scan 81,495 47.05%
Clinton Hand-counted 20,889 47.05%
Obama Hand-counted 23,509 52.95%
- Two hand count towns reported "zero" votes for candidate Ron Paul to the media, even though they did have votes for him. The town of Sutton reported zero, but had 31 votes; the town of Greenville reported zero, but had 25 votes. The two towns had misreported results affecting exactly the same candidate in exactly the same way.
- Results in many locations arrived up to four hours late on Election Night, surprisingly, from machine-counted locations -- not hand count locations;
- A single private entity had control over coding for every memory card in New Hampshire. According to the contract for LHS Associates, this firm requires a right of access to any voting machine at any time, services the machines, maintains the machines and handles repairs, replacements and troubleshooting on Election Day.
- Ken Hajjar, a key employee of this sole source private entity, LHS Associates, has a criminal record for narcotics trafficking. The state of New Hampshire knew of this conviction but approved the contractor anyway. According to a complaint filed with the New Hampshire Attorney General, Hajjar had called the Dan Pierce radio show in 1999 and threatened to rig an election.
- A high number of "other" votes appeared in Manchester, where over 570 people apparently decided to go to the polls and choose none of the first tier OR second tier candidates.
- The voting system in New Hampshire was updated, but to a version that had been proven to be vulnerable in studies in Florida and California. Instead of upgrading to newer versions which at least claim to address known security vulnerabilities, New Hampshire chose to implement none of the beefed up procedures or upgraded versions that other states are using.
Citizens from many different states are now examining New Hampshire's ballot chain of custody, because if that is as weak as their voting machine controls, the recounts will just produce new questions.
A newly aroused citizenry in New Hampshire and elsewhere is telling New Hampshire "trust us" is NOT the way to run elections. |
Brian London
Voting Rights Forum Participant
Username: Brian
Post Number: 7
Registered: 1-2008
Best of Black Box?
Votes: 2 (A keeper?) | | Posted on Saturday, January 12, 2008 - 6:21 pm: |
|
First let me say thank you to Bev for starting a new thread. I have spent most of the day harvesting data from the New Hampshire ELMIB community profiles. Who knew parsing pdfs would be such a pain. Anyway, I haven't done any analysis on it at all yet, but thought I should get the data out to other statistically minded people.
|
Mary L Burke
Voting Rights Forum Participant
Username: Marytnurse
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Saturday, January 12, 2008 - 7:02 pm: |
|
HELP!!!! I need to find a way to get info to Bev or Nancy Tobi to ask them if they will be able to help with the recount. David Bright is coordinating the effort for Kucinich, but I don't think they have any expertise on this issue, and everyone wants to support them in their recount. The chain of custody thing is worriesome and I don't want DK set up any further than he has been to look crazy (and NO I didn't caucus for him here in Iowa). same SN @ yahoo if anyone knows how to get either of them or any one else who may be an expert and available to be in NH. We have some people who might be able to help with monitoring the recount, but need to know where they are needed.
THANKS!!
Mary |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 1
Registered: 1-2008
Best of Black Box?
Votes: 2 (A keeper?) | | Posted on Saturday, January 12, 2008 - 8:24 pm: |
|
Brian - your file is missing data for the Dixville and Millsvie districts, but was very helpful, thanks!
Districts with diebold machines show disproportionately more votes for Clinton, but only in those districts where Obama polls higher than average. This could happen if diebold counts were altered only in those districts where Obama seemed strongest.
The stats show that less than 1:1000 odds for this difference occurring through chance alone, and that's controlling for fluctuations due to every one of the demographic variables in Brian's file (thanks Brian!)
In short, this is very fishy. I've sent this information to many sources and I'm afraid people are too stats-illiterate (or I'm too jargony) for the implications to be clear. If I have been unclear, please reply. |
Howard Randall Smith
Voting Rights Forum Participant
Username: Tidalcreek
Post Number: 43
Registered: 2-2007
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Saturday, January 12, 2008 - 8:30 pm: |
|
Chain of Custody Questions re Timing in the upcoming recount
I hope there is a way to shine a bright light on the chain of custody issue in New Hampshire. I don't know the most effective way, but one idea would be for the issue to arise from within the Kucinich camp, or the Howard camp, by way of showing, unmistakably, how a recount may, or may not be trusted, based on the chain of custody. Unfortunately, I have to say that "timing is everything". Such a development would have to be seen as part and parcel of the recount in order to be picked up and carried in the press.
Randy
|
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Saturday, January 12, 2008 - 8:31 pm: |
|
PS: please replicate my analysis by regressing Clinton's votes by district on Obama's votes, whether the district uses Diebold or hand-counts, their interaction, and any demographic covariates you like. You'll see the effect is highly significant. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 1
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 12:42 am: |
|
Being brand-new to this forum, and even to the website, I apologize if this has already been addressed. Is anyone rounding up volunteers to monitor the recount, to isolate all the "narrow points" in the chain-of-custody of the ballots? Speaking as a former precinct captain in northern Virginia, I sense there is a huge need of volunteers to monitor this recount and is anyone organizing them? I would think we need to make sure every nook and cranny is covered. If I can assist from my location in New Mexico, I'll be happy to. I can be emailed at: dondep (at) gmail. com.
Mr. Chatham, thanks for your analysis of the odds of the difference occurring through chance alone. Do you suspect a particular formula was then used in the case of jurisdictions where Obama ran stronger, but the Hillary Anomaly showed up? Also, does anyone have access to the actual exit polls from each location, particularly the vote-rich Manchester precincts? It looks like the 3,000+ Anomaly first reared its ugly head there early in the count and maintained the actual number until the last 15% or so. Last question; were the exit polls showing Obama running stronger in urban and more affluent areas, such as Manchester? |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7474
Registered: 12-2004
Best of Black Box?
Votes: 3 (A keeper?) | | Posted on Sunday, January 13, 2008 - 4:38 am: |
|
Hi all, and quick responses to some of these points:
Brian, thanks very much for the spreadsheet and Chris, thanks for the analysis.
A word about spreadsheets: One reason we got out of the gate so quickly in New Hampshire was that Black Box Voting did the tedious work of putting every location matched up with hand count or Diebold/LHS count and published it for everyone to have before the numbers even started rolling in. When Brian and others provide spreadsheets, it offers a great shortcut for others to crunch their own numbers. Just preparing the spreadsheet takes time.
Caveat: When you get a spreadsheet from someone else, you have to quickly check every input number to make sure whoever put in the raw data didn't have typos. Nothing would be more embarrassing than announcing a grand conclusion based on a typo. You also have to eyeball the formulas entered, for the same reason.
I will say that I am privy to some back-channel information and that someone else has been doing an entirely different study, using a slightly different approach to Chris, and was coming up with the same issue. I got the more detailed results of that study about 10 minutes before receiving Chris's information on Friday, and Chris, I hope you don't mind, but I forwarded your info to the EDA statisticians with a note. I know they appreciated it. They are right now in the process of further examination and confirmation of initial findings.
I encourage other statistically-minded individuals to crunch the numbers. The Republican numbers need to be crunched! And another study that needs to be done is a comparison of the initial results reported to the media with the official results.
One more caveat: As new statistical analyses roll in, especially if significant anomalies are found and get legs in the blogosphere, you will see false analyses show up, planted to weaken the message. Beware of anything from an unknown entity that contains what looks like a knockout punch. We see these things all the time. They're called "honey pots" and they're usually (a) EASY to understand for people not statistically inclined (b) DRAMATIC conclusions and (c) Just enough "Wow!" details to get legs along with just enough information that someone will come in and debunk -- after it has been widely reported.
We're in the realm of power politics, and they are well versed in use of the Internet for damage control. |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 1
Registered: 1-2008
Best of Black Box?
Votes: 3 (A keeper?) | | Posted on Sunday, January 13, 2008 - 4:42 am: |
|
The checkthevote numbers that show exact reversal of C vs O % do not match the numers on the NH SoS site. Furthermore, 11 towns listed as Hand-count by checkthevote are Machine-count towns by my list. I got a list of Hand-count ballots from the SoS office and correlated with SoS website data. Also, the checkthevote has three "unknowns" and a zero count for Windsor. Compare, for example, Campton where Checkthevote has 754 votes, and the SoS website shows 743. A large percentage of the towns show similar discrepancies.
There is no reason to think that if the tallies were switched, that H vs M would be perfectly swapped to 5 figures. H towns are demographically different than M towns, and would not be expected to decide the same to 5 figures. I think there is something wrong with the checkthevote numbers.
My analysis shows the following:
H-Clint 15,817/46,734 = 33.84%
H-Obama 18,744/46,734 = 40.10
M-Clint 96,789/238,800 = 40.53%
M-Obama 86,260/238,800 = 36.12%
I tried to correct for demographics, dropping small towns below 400 votes, and large towns over 1100 votes. There is significant overlap between H and M in the 400 - 1100 vote range. These results are:
H-Clint 34.19%
H-Obama 40.00%
M-Clint 37.30%
M-Obama 37.21%
I would have expected most of the difference H vs M to go away limiting the sample to medium towns. There still seems to be a 3% flip. Suggestive, but not conclusive.
|
Bev Harris
Board Administrator
Username: Admin
Post Number: 7475
Registered: 12-2004
Best of Black Box?
Votes: 3 (A keeper?) | | Posted on Sunday, January 13, 2008 - 5:44 am: |
|
To Mary (and welcome to Black Box Voting!):
quote: I need to find a way to get info to Bev or Nancy Tobi to ask them if they will be able to help with the recount.
Nancy Tobi, as we all know, lives in New Hampshire and is rather spectacular in her ability to analyze and communicate. However, she will not be available. Nancy is one of our great patriots who has contributed tremendously to the cause of election integrity. This work takes a brutal toll on the health of those on the front lines, as Tom Courbat can attest.
I can be reached using 206 number at this site. That is my cell phone. I made the difficult decision back in 2004 NOT to place a screener between me and my callers, because the best and most important information often comes in from people who might not get past a screener, because the screener might not recognize the relevance of what they have to say. Also because citizens benefit from mentoring and to the extent that I have time I do that as well.
This means I can be very difficult to reach, but I do process an amazing number of calls; I prefer e-mail because it's faster for me and most days I get to almost all of it, which averages about 800/day
Guidelines for calls
Keep them short.
If you are calling to get ideas for what you can do:
- Tell me quickly of your location
- Tell me quickly the kind of thing you are comfortable with, for example stats, legal stuff or public records, phone calls, letters, internet research. Give me a thumbnail, not a monologue.
I try very hard to provide tasks and tips that move the ball forward, in a compressed time frame. It's important to initiate actions that don't require my ongoing management time. Black Box Voting focuses on citizens who initiate action, not on organizing and leading a line of ants. We like the swarm approach and have faith in people's common sense and independence of action. (Isn't that what democracy IS?)
Providing tips, leads, front lines reports
Here's the tip delivery thing down to a science: "I have this for you; here is the backup documentation; good luck, bye!" I love that and some citizens have now got that down pat. That's how I got the Greenville tip ahead of everyone else, long before it was up on the sec. state site.
If you aren't sure how important the information is: Don't worry! Provide it anyway, though I prefer it by email. If it seems important, call it in.
It is common for people who have some of the best information to not be aware of how important it is, and I find they may be tentative and kind of slow about spitting out what they know because they are afraid to take my time with something irrelevant. DON'T BE! If you have something you think might be important, it really might be, and those are exactly the kinds of tips that unravel things. I won't mind if it doesn't pan out; as an investigative writer, I have a lot of my own tips and theories that don't pan out either, but you never know till you check it out!
Don't call just to chat
If you call me during a live election it should be to provide specific information, not to have a chat about the state of the world and your general opinions and concerns. Here is my least favorite conversation:
"Hi...my name is [name]...pause ... and I'm just wondering, are you guys looking into what's going on in [current election]? So what's going to happen? What do you think about [name of candidate]? I'm really worried about this because [name of candidate] is going to help America on [name of issue]. What do you think about that?..."
As you can see, this kind of call (a) takes a lot of time (b) contributes no useful information and (c) removes time from looking at the real stuff coming in.
Chatty calls with no end goal are the down side of not using a screener. I get a lot of them. Email your comments, I will read them, but don't take up phone time just to socialize.
Calls regarding helping with recounts
Black Box Voting is a 501c(3) nonprofit and we cannot assist a candidate or a campaign. We can assist VOTERS and we can provide basic procedural info to any candidate, including what to look for as red flags and what kinds of questions they should be getting answers to.
As far as citizens monitoring the New Hampshire recount, I hope to have more of that information up Monday or Tuesday.
We don't know which areas the Republican candidate will recount, and which the Democrat will recount, and we don't know if chain of custody is intact.
If chain of custody is not intact, things get a lot more difficult for monitors, though perhaps not impossible.
quote:David Bright is coordinating the effort for Kucinich, but I don't think they have any expertise on this issue, and everyone wants to support them in their recount.
"Helpful" people will put information in his ear that will steer him in the wrong direction. Same with Albert Howard, the Republican candidate. I don't know what we can do about this. I'm not going to call the Kucinich campaign and I already know that he's going to take more influence from insiders than from me. My focus is on the citizenry, though I do take calls from the candidates if they have questions. I don't contact them, they have to contact me, and when they do I provide the same information to any who request it.
quote: The chain of custody thing is worriesome and I don't want DK set up any further than he has
Chain of custody is all the marbles.
quote:We have some people who might be able to help with monitoring the recount, but need to know where they are needed.
The candidates won't choose whether they'll recount just some areas or all areas until, I believe it is Tuesday. The counting is done in a central location. One thing that's important right now is to pin down the rights of citizen observers. I would appreciate that information, along with information about what citizens could and could not see with the Nader recount. |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7476
Registered: 12-2004
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Sunday, January 13, 2008 - 5:55 am: |
|
To Don, welcome to Black Box Voting!
BBV doesn't organize volunteers, we help individual citizens initiate actions which can include pointing them and the citizens they organize in productive directions. We focus on the leaders, not the followers, and on the individuals, who often get more done than some of the groups anyway.
1. Need to know the rights of the citizenry to observe the recount
2. Need to know past experience of citizens who tried to observe.
If you were watching the Nader recount in NH in 2004, would love it if you can check in with a front lines report on how it was set up, what you could see, what you couldn't see, what you focused on, what you would do differently this time.
3. Monitoring chain of custody: One of the participants at BBV provided the New Hampshire procedures, which is to have the state police bring the ballots in to a central location. Apparently in 2004 with the Nader recount there was media coverage of a couple guys who always are in charge of this, funny names like Fritz and Bubba or something. If someone can run that down and post it here that would be great.
4. There's always "the procedures" and then "what they did" -- front lines reports shed a lot more light on procedures. At this stage, the most important thing is to learn more about "what they did" in previous situations.
5. Stats are well along to identify focus locations for the Democrats, but are badly needed for the Republicans. A focus that might be valuable would be the fourth-fifth place positionings.
6. I'd also like to see an analysis of the early reported media results vs. the final results. It came to my attention in Iowa that reporting false results to the media may not be a crime. Since election night results end up on page 1 and corrections end up on page 12 or nowhere at all, this is a relevant way to cheat when the focus is public perception of candidate positioning. |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7477
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 6:02 am: |
|
Mark, thanks for your analysis and for checking the checkthevote numbers. I'm curious as to whether they used my spreadsheet on the handcount vs machine count locations, which I pulled from the sec state web site shortly before the election, or their own.
Obviously if my own is wrong I'll be very concerned. That would be due to a late-stage change in sec state info, since that's the source of my info. I pulled the information a month or so before the election.
Can you identify which 11 counties differ, so I can look at the info distributed by BBV to make sure we got it right? |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:23 am: |
|
Bev,
I visited the SoS office on Wed, and asked about hand vs Machine count. They gave me a list of towns that recieved ballots for hand-count. I put that information into the spreadsheet I sent you.
There could have been transcription errors on my part. Unfortunately I left the list at work, not realizing at the time that there was an issue. I will double check myself on Monday.
Please note that Checkthevote has issued corrections on it's website about H vs M towns. However, I'm still not sure they got it all right. They switched Plymouth to Machine count, but I'm almost certain it's handcount. As I recall it was the first town on the list because it got by far the most hand-count ballots, and the list was in order of number of ballots requested. |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7481
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:23 am: |
|
Off topic for NH - cross posted from Michigan section of these forums:
Note to researchers - unlike New Hampshire, which has a municipality-based election structure with different voting systems within the same counties, Michigan has a municipality-based system but UNIFORM systems within each county.
A couple months ago I prepared this spreadsheet containing all 1900+ townships with their counties and voting system attached; it also has contact information for the clerks to help you check out discrepancies quickly (like the Greenville and Sutton incidents we saw in NH).
http://www.bbvdocs.org/MI/state/Michigan-townships-by-county.xls
(512 KB)
(from admin): Note the date, prepared as of the sec. state data a couple months ago. All formal analyses should use most current sec state data. As with everything else, make sure to use official sources, but you can update quickly by eyeballing those vs this spreadsheet, if any updates are needed) |
John Belmonte
Voting Rights Forum Participant
Username: Jbelmonte
Post Number: 6
Registered: 6-2006
Best of Black Box? 
Votes: 1 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:51 am: |
|
Mark Schauer wrote:
quote:
Please note that Checkthevote has issued corrections on it's website about H vs M towns. However, I'm still not sure they got it all right. They switched Plymouth to Machine count, but I'm almost certain it's handcount. As I recall it was the first town on the list because it got by far the most hand-count ballots, and the list was in order of number of ballots requested.
Yes, checkthevote and others have been using incorrect/incomplete town hand/machine-count assignment data which does not match nh.gov site (http://www.sos.nh.gov/voting%20machines2006.htm). I've been trying to get these people corrected for several days. The nh.gov site clearly shows Plymouth as machine count, why do you think it's otherwise? |
Drew J. Jones
Voting Rights Forum Participant
Username: Drewjjones
Post Number: 6
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:25 am: |
|
Most of the fraud, assuming we are dealing with fraud, seems to have occurred in Hillsborough and Rockingham counties. I received this spreadsheet earlier with graphs noting discrepancies. |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 3
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:26 am: |
|
I got a list from the NH SoS office on Wed of the towns that requested Hand-Count ballots (I work in Concord, NH). The Assistant SoS told me that the towns on the list Hand count, and all others Machine count. Unfortunately I left the list at work, I will double check Monday, but as I recall Plymouth stood out at the head of the list because it requested the most ballots (the list was ordered in terms of quantity of ballots requested, not alphabetically). Apparently it is the biggest town that hand-counts.
On Monday I will double-check my list, and if Plymouth is there I will call the town clerk and find out how they counted. |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 4
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 9:11 am: |
|
Drew,
I don't thing Rockingham is a good place to start. Remember that, in general, Hand-count vs Machine count are dealing with very different demographic regions.
The last sheet on my spreadsheet (posted above) compares towns of similar size within each county. I excluded all towns except those with vote totals between 400 and 1200. Comparing the Clinton "bounce" ([%MClint-%HClint]-[%MObama-%HObama]) For Rockingham county it is -1.96% (Clinton did worse in the machine counted towns of intermediate size). I would focus on Cheshire county which shows a +19.74% bounce.
I would suggest initially focussing on Towns with the best chain of custody in counties with the biggest bounce in the 400-1200 range. I will try to learn more about chain of custody in various towns on Monday. I will be focussing on the smaller towns in Cheshire and Hillsborough counties. |
Bill Bowen
Voting Rights Forum Participant
Username: Bbowen8
Post Number: 10
Registered: 11-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 11:32 am: |
|
I have a general question about the chain of custody issue... I guess, given enough time it could be possible, but even if LHS were able to swap ballots from one town to another to match up say the Clinton/Obama race, wouldn't they have a very hard time making all the ballots for the other candidate's (Edwards, Richardson, write-ins) numbers match up? That would seem to me to be a logistical nightmare for someone to match ballots so all four Democratic candidates equal the final tallies, and even more so on the Rebuplican side where more candidates are on the ballots.
Obviously we need to trust the chain of custody, but it does seem to me that even though it's possible to swap ballots so the totals match the results, I think it would take quite some time to make sure all the totals for all the candidates match when you move ballots from one location to another for a recount. Again, I'm not saying it's impossible, but how improbable or difficult is this scenario? Maybe Diebold has machines built specifically for this purpose!
Thanks,
Bill |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4356
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 11:53 am: |
|
That depends.
If plenty of extra ballots were printed and are not accounted for in the chain of custody, then it's not really a "problem".
Bev recently described a situation in which there were lots of surplus ballots printed but no one kept track of them and they just disappeared. And this was someplace that got a lot of publicity for questionable elections. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 1:25 pm: |
|
Thanks for the welcome, Bev. Love the 'swarm' approach too!
Another concern would be the out~of~towners who are allowed to vote in NH so long as they signed a form promising they would move into the state within a certain time frame.
(BTW, I have to use another PC to view the spreadsheets as I can't afford the Excel program at the moment; now I can read the attachments with the stats, though it's only Excel Viewer.)
To Mary L Burke; May I suggest having your helpers go to where the ballots are and put a 24/7 watch on? Including entrances. My first job in politics was counting the vote, and not even a pizza delivery is allowed in the machine area. If someone can collect the lists of out~of~towners, I'll do the work to input the names and follow up on how legit they are. I'm not as qualified a statistician as Chris, Mark, Drew or Bev, but I can do some of the leg work and I'm a quick study.
p.s. I think we should be careful about throwing out numbers other than the "official" NH ones, for obvious reasons. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 3
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 1:32 pm: |
|
p.s. I forgot to include Brian above. Brian, my Excel Viewer is 2007; is your file above relying on a newer feature such that I can't open it? |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4359
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 1:43 pm: |
|
quote:who are allowed to vote in NH so long as they signed a form promising they would move into the state within a certain time frame
Don,
Thanks, that helps me understand why some people commented on the huge number of MA license plates seen at some polling places. I didn't realize they could vote in NH if they signed a form. Good grief. That is a huge loophole.
Have you got Open Office? It's free and is compatible with most MS files (.xls, .doc, .ppt etc.). You can download it at www.openoffice.org |
Gentry Lange
Voting Rights Forum Participant
Username: Gentrylange
Post Number: 42
Registered: 9-2005
Best of Black Box?
Votes: 2 (A keeper?) | | Posted on Sunday, January 13, 2008 - 2:52 pm: |
|
http://www.sos.nh.gov/recount%20press%20release.pdf
The SOS of NH is saying the recount is starting Wednesday the 16th for both the D and R count. With a place TBD.
As to the extra ballot concern, if NH is typical of poll based polling systems then there will be a projected turnout + buffer amount printed. So if there's a 50% predicted turnout they'll print 50% + n (n being the buffer).
Now in heavy Vote-By Mail states, like Oregon or Washington, this problem will be HUGE. Because they print 100% of the Ballots all the time. So with a 50% turnout they'll be 50% extra ballots just floating around. Plus it has been revealed that PSI Group out here (WA) has previously printed a lot of extra ballots anyway.
But I really don't know the buffer in NH, nor the amount of absentee voting that can or does go on. Also who is printing the ballots? Do they print a bunch of extra votes? All worthy questions if you want to start thinking about Vote Stuffing vectors of attack. |
Brian London
Voting Rights Forum Participant
Username: Brian
Post Number: 8
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Sunday, January 13, 2008 - 3:05 pm: |
|
Don, I actually had the same problem on my work computer. My file is a CSV, not an Excel file. I don't know why the forum changed the extension. Somewhere between the forum software, the browser, and the viewer, the Content-Type got messed up. Just download the file to your hard drive and change the extension to .csv. |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 83
Registered: 12-2005
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Sunday, January 13, 2008 - 4:44 pm: |
|
It isn't just the extension of NHData.csv (23.9 kB) that is wrong; the link is to a completely different file NH_Data-71290.xls (157kB).
Oops! Never mind, I just looked at the wrong file. I have found a few files here with incorrect extensions.
(Message edited by Mike_LaBonte on January 13, 2008) |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7483
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 5:47 pm: |
|
Hey Mike -
When does LHS pick up the memory cards after an election? Is it unusual that they went to Nashua and picked up its memory cards on election night? According to the Nashua elections official, he figured they were already reprogramming them.
That is destruction of an election record and is exactly the kind of situation that caused an Alameda County judge to order a new election recently. |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7485
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 6:06 pm: |
|
Mark, you haven't provided any real evidence for the demographic differences. We're not dealing with Dem vs. Rep we're dealing with Dem vs. Dem and Rep. vs Rep, so you can't extrapolate from anything at all but previous primaries. |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7486
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 6:13 pm: |
|
Bill,
Why would you swap ballots for the other candidates if all you need to do is flip the top two? Sort, stack, swap, shuffle. |
John Howard
Frequent Voting Rights Forum Participant
Username: Harmonyguy
Post Number: 548
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 6:14 pm: |
|
Did those precincts in NH that used Optical Scan ballots, use the scanners to electronically communicate the results to a centralized GEMS system somewhere, or did they merely print the local reports and then phone in the results?
HG;) |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7487
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 6:20 pm: |
|
I think they print the results and then they are physically taken to the secretary of state. At the secretary of state, an employee enters them into an Excel spreadsheet. |
John Howard
Frequent Voting Rights Forum Participant
Username: Harmonyguy
Post Number: 549
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 6:26 pm: |
|
Thanks Bev - that's what I thought I'd read.
In think-and-analyze mode....more to come.
HG;) |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 84
Registered: 12-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 6:29 pm: |
|
I will try to find out tomorrow how long memory cards usually stay at City Hall. As a volunteer election official I don't work there, so I don't know every detail. I do know is that on election nights the machines come back and are stored in a locked room behind a locked room with an alarm, with the cards still sealed in.
When they go to LHS is probably arbitrary, although you would think they would not want the cards until it was time to program them. At the same time, one would think the cards would be kept as long as the ballots, up to 23 months. But I'm pretty sure we have only 2 sets, and both are programmed for each election.
LHS is just 20 minutes away for us, by the way. I go by the building where LHS is located a few times a week. You can almost see it from I-93. |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7488
Registered: 12-2004
Best of Black Box?
Votes: 2 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:00 pm: |
|
Breaking ... back channel, from a credible source: preliminary analysis looks like the Obama/Clinton flip is accompanied by a Paul/Giuliani flip in the same locations with the same pattern. |
Shawn MacFarland
Voting Rights Forum Participant
Username: Smacfarl
Post Number: 1
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:07 pm: |
|
Bev,
The black box vote list of Diebold vs hand counts
http://www.bbvdocs.org/NH/state/Jan-08-votingsystems-NH.txt
From admin: Checked latest update from all locations with most current sec state info last night, Jan. 13, 2008. The locations identified by Shawn, below, have changed to Diebold voting machines and the above spreadsheet should be updated accordingly. For those researchers new to this issue, always check numbers with original source prior to releasing conclusions.
is significantly different than this list from the NH Secretary of State from Nov 26,2007.
http://www.sos.nh.gov/voting%20machines2006.htm
Notably added by the SOS in November are
Campton
Claremont - 3 Wards
East Kingston
Fitzwilliams
Franklin - 3 Wards
Hillsborough
Moultonborough
New Boston
New Durham
Nottingham
Ossipee
Plymouth
Rindge
Tamworth
Have all these locations gone back to hand counting since November? Is the sos list wrong?
What's the deal? |
Richard H Beeson
Voting Rights Forum Participant
Username: Rhb
Post Number: 5
Registered: 3-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:29 pm: |
|
Why are the candidates names not written in the same fashion in these two Post Number: 7488 messages? One web page / One email ?????
***********************************************
From the web page:
http://www.bbvforums.org/cgi-bin/forums/show.cgi?1954/71287
Bev Harris
Board Administrator
Username: Admin
Post Number: 7488
Registered: 12-2004
Posted on Sunday, January 13, 2008 - 7:00 pm:
Breaking ... back channel, from a credible source: preliminary analysis looks like the Obama/Clinton flip is accompanied by a Paul/Giuliani flip in the same locations with the same pattern.
*********************************************
From the email: Sun, 13 Jan 2008 19:00:50 -0800 (PST)
Bev Harris
Board Administrator
Username: Admin
Post Number: 7488
Registered: 12-2004
Posted on Sunday, January 13, 2008 - 7:00 pm:
Breaking ... preliminary analysis looks like the Clinton/Obama flip is accompanied by a Giuliani/Paul flip in the same locations with the same pattern.
*********************************************
If one does not look at both they will not know about this situation. Curious... |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7489
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:34 pm: |
|
I edited the post to indicate which direction the flip went. The email went out automatically and the edit is what you see now. Of course, which direction the flip went depends on if you were standing in hand count territory or LHS territory, so I guess my edit was pointless. Sorry about the confusion. |
Eric Adler
Voting Rights Forum Participant
Username: Scientist
Post Number: 1
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:40 pm: |
|
It turns out that the raw exit polls taken during the day seemed to point to a close race.
http://machinist.salon.com/feature/2008/01/11/new_hampshire_vote/
" Daniel Merkle, who heads ABC News' "decision desk" -- which was getting the exact same exit polling data that folks at NBC were getting -- told me that the numbers he was receiving during Election Day did not show a certain Obama win. Merkle said the data indicated "a very close race on the Democratic side," and "that's what it ended up being."
"It was within a couple points," Merkle said. "When we're seeing an exit poll within a couple points, that's a close race." The exit poll numbers, he added, were a "surprise" compared to pre-election polls. "The exit poll was not showing an 8- to 10-point Obama lead. It was showing a close race."
I do not believe that there was fraud in the NH primary. People do make mistakes in tallying at end of a full day at the polls.
I am happy that Kucinich has asked for a recount. It is always good to validate the elections when there is the slightest suspicion. |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 85
Registered: 12-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 7:51 pm: |
|
I was in the Ossipee town hall just ahead of Nov 2006, and noticed an Ethernet wire in their big hall. I asked about it and was told it would be used to transmit vote results to Concord (by an RF link, if I recall correctly).
To go from that to hand counting is a surprise. |
David Luebbers
Voting Rights Forum Participant
Username: David_luebbers
Post Number: 33
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:16 pm: |
|
"I was in the Ossipee town hall just ahead of Nov 2006, and noticed an Ethernet wire in their big hall. I asked about it and was told it would be used to transmit vote results to Concord (by an RF link, if I recall correctly).
To go from that to hand counting is a surprise."
http://en.wikipedia.org/wiki/Ethernet
RF link is Wireless.......... As i have stated before .. Wireless can be hacked in 3 min .. (Thats with the target system using the best encryption availiable)
The majority of systems that have wireless though have no WEP (Wireless Encryption Protection) Enabled.. They come Standered with it but by default the option is turned off and the user has to turn the option on.
I know about Wireless... It is totally completely insecure.
http://www.networkworld.com/podcasts/panorama/2007/022807pan-airdefense.html
"It doesn't matter who votes. It only matters who counts the ballots." Stalin
|
Shawn MacFarland
Voting Rights Forum Participant
Username: Smacfarl
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:19 pm: |
|
Attention Bev Harris,
checkthevotes has indicated that they believe the New Hampshire Secretary of state list of Diebold districts is more accurate than the black box vote one. (See my post above for questions to you about this.)
From the Update at Checkthevotes.com
"Also, I was alerted by another person who is analyzing the stats that because of some bad lists that have been circulating, I was attributing some machine count precincts as being hand counted. I have corrected this. The precincts that have changed are: Campton, Claremont, East Kingston, Fitzwilliam, Franklin, Hillsborough, Manchester, Moultonborough, New Boston, New Durham, Nottingham, Ossipee, Plymouth, Rindge, and Tamworth"
He is referring to
http://neggie.net/vote2008/nh_primary.cgi
who is using the Secretary of State list.
This would seem to coorespond with Mike LaBonte account above about Ossipee.
(Message edited by smacfarl on January 13, 2008) |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 86
Registered: 12-2005
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:28 pm: |
|
I found an interesting post at Slashdot by someone who campaigned for Obama:
http://politics.slashdot.org/comments.pl?sid=413698&cid=21986704
"I campaigned for Obama for several days out of the North Conway NH office. While the media reported a 10-12% lead, none of us inside the Obama campaign believed them. At best, our own internal polling put us at 1-2% behind Clinton in rural areas and slightly ahead in the urban counties.
In Ossipee, where I spent the majority of my time, Clinton won 281 to 261 over Obama (hand counted). There was record-shattering voted turnout in the area for both parties. Previously, the record was ~1000 voters. On Tuesday over 1500 voters showed up. Several nearby towns even reported running out of paper ballots.
I think the real problem was how the media handled their polls. Many Obama supporters I talked to on primary day mentioned that they were planning to support Ron Paul or vote against a candidate in the Republican party because they didn't believe Obama needed their support. Mind you, these are people with Obama signs in their yards who had actively been helping in his campaign. I wonder how much credit we can attribute to voter complacency rather than some Diebold conspiracy theory.
In any case, I don't understand all the fuss. Obama and Clinton were awarded the same number of delegates. This whole mess only matters to the media and spin people."
I'm ignoring the last 4 sentences. I had a theory that some Republicans actually voted for Clinton, considering her to be the most defeatable in the election. But this theory may be even better. I am convince that cross-over voters exist in significant numbers in NH, but they would report who they really like when polled.
I'm glad about the recounts. They almost certainly will not match the election day counts vote by vote, but if they show that Clinton did win, I'll stick with this theory over the foolish ideas put forth by pundits for why the polls were wrong. |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 3
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:37 pm: |
|
I wonder if I can predict the exit poll results within their sampling error by statistically "correcting" for the apparent fraud (i.e., the obama/hillary flip on diebold machines in districts where obama was strongest).
To do this I need district by district exit polls for the democratic NH primary, with the estimated sampling error of each (or I can calculate the latter if the data includes the # of individuals polled in each distrct). Can anyone help or point me in the right direction?
@Don - I can't answer that question - the problem is that my analysis method would necessarily make it appear that there's a formula even when there's not (for technical reasons). |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 87
Registered: 12-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:40 pm: |
|
I should emphasize that I was told the Ossipee link would be used to transmit results. Also it was described as experimental. I find no indication anywhere that it ever happened. |
David Luebbers
Voting Rights Forum Participant
Username: David_luebbers
Post Number: 34
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 8:45 pm: |
|
"I should emphasize that I was told the Ossipee link would be used to transmit results. Also it was described as experimental. I find no indication anywhere that it ever happened."
Who told you this ?
Also.. Wireless is not experimental at all..
"It doesn't matter who votes. It only matters who counts the ballots." Stalin
|
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 88
Registered: 12-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 9:35 pm: |
|
I can't remember who told me. I was in Ossipee for the NH boating class so I met everyone there only that one time. Wireless is not experimental, but using it to transmit votes was. Actually I think it may have been a line of sight system, implying microwave. Can't remember.
Again, I don't know if it ever happened. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 4
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 9:44 pm: |
|
Chris: If you could let us know if and when you get exit-poll data, the vote-rich Manchester wards would be the best place to focus. From what little we could glean from the talking heads, who are now extremely silent on this issue, the exit polls all the majors used found that the urban demographic was more inclined towards Obama, which calls into question the Manchester totals which show a 45+% for Hillary to 30+% for Obama. After crunching the numbers we do have, I think the most important theft occurred in those 12 wards. If the numbers were 'flipped', then the 3,000+ margin of difference that showed up for Hillary began there (that was also one of the first areas to come in).
Mary: If your people are still looking for where to position themselves, I would strongly recommend and encourage they monitor the following locations and locating the party monitors for the following wards (this info is available at: http://www.sos.nh.gov/polling.htm
We need to find out the particulars of the "chain-of-custody" to see WHO transported the ballots from those locations, and to where they were taken. Otherwise I see this turning into a worse nightmare if LHS held onto the machines AND the paper ballots from time of vote to time of lock-up in.....where? Also, just to make sure all avenues are examined, does this Leo Bernier have a 'record' of any kind? If anyone on the 'inside' was compromised, not that I'm accusing anyone, he would be the most likely suspect.
Manchester - Leo Bernier
Ward 1 Brookside Congregational Church, 2013 Elm St. 6:00am - 7:00pm
Ward 2 Hillside Middle School, 112 Reservoir Avenue
Ward 3 Carol M. Rines Center, 1528 Elm St. (Use Rear Entrance-not Elm Street)
Ward 4 McDonough School, 550 Lowell Street
Ward 5 Beech Street School, 333 Beech Street
Ward 6 St. Pius CCD Center, Candia Road and Sarto Street
Ward 7 St. Anthony Community Center, 148 Belmont Street
Ward 8 Jewett Street School, 130 South Jewett Street
Ward 9 Bishop Leo E. O'Neil Youth Center, 30 S. Elm Street
Ward 10 Parker Varney School, 223 James Pollock Drive
Ward 11 Gossler School, 99 Sullivan Street
Ward 12 Northwest Elementary School, 300 Youville Street |
Bill Bowen
Voting Rights Forum Participant
Username: Bbowen8
Post Number: 12
Registered: 11-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 10:32 pm: |
|
I found the following online article requesting the United Nations to monitor U.S. elections for fraud.
http://blackstarnews.com/?c=119&a=4109
What is the precedent for this? Has it ever been done in the U.S? Does anyone know of any countries where the U.N. has oversaw elections, and what is typically required in order for them to oversee an election? As long as our politicians aren't involved, it could highly raise the election fraud issue(s).
Bill |
David Luebbers
Voting Rights Forum Participant
Username: David_luebbers
Post Number: 37
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Sunday, January 13, 2008 - 10:58 pm: |
|
Mike..
Microwave systems might be it but Microwave is dangerous and requires somewhat large equipment especially for long distances. (I am not as familure with Microwave as i am with Wireless)
Wireless also has limited range depending on the hardware in question. I am not up on the current tech for wireless but me and a friend was able to send and recieve a Wireless signal from 30 or so miles.. Granted we used a dish (Small one that most Direct Tv dishes use) Very man portable.
Both Microwave and Wireless Signals can be intercepted . (Note.. Microwave signals are Much harder to intercept then Wireless)
It is a shame you don't know exactly what they were using.
"It doesn't matter who votes. It only matters who counts the ballots." Stalin
|
Bob Fleischer
Frequent Voting Rights Forum Participant
Username: Rjf7r
Post Number: 151
Registered: 9-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 7:09 am: |
|
I think there is some serious confusion here about "wireless" and "microwave". "Wireless" is the more generic term for any radio communication; "microwave" refers to a specific radio signal wavelength range. I think that in at least one case above "wireless" was used to refer to the WiFi networking technology, which is indeed wireless, and is indeed easily hacked. However, WiFi is only one of many wireless technologies, and some are quite secure. Ironically, WiFi uses the same wavelength signals as do microwave ovens; in that small sense WiFi IS microwave. |
Bob Fleischer
Frequent Voting Rights Forum Participant
Username: Rjf7r
Post Number: 152
Registered: 9-2005
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Monday, January 14, 2008 - 7:18 am: |
|
re "It is always good to validate the elections when there is the slightest suspicion":
I hope that, regardless of the outcome, we (the election integrity community) can spin the "take away" message to be something like "if we had immediate automatic auditing, we wouldn't have this uncertainty after so many elections".
I think we really need to work on our message, and work on getting it out; if it turns out that no problem can be proven, you can be sure that the opponents of reform will spin this to "the machines are proven to be correct". |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4368
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 7:23 am: |
|
Yes. Immediate automatic auditing--of all results, regardless of whether or not they "seem" okay or match exit polls.
Immediate double-checking has to become a non-negotiable aspect of our election routine. And if there are any discrepancies then we need to triple-check until everyone agrees. |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 5
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Monday, January 14, 2008 - 8:08 am: |
|
Just called Ossipee town clerk. Her name is Kelly. Ossipee is using machine counting. The ballots and memory cards are in her safe. Chain of custody is good. No audits on day of election, only before.
I had a very nice conversation with the Plymouth TC, Karen. They used a machine for the first time this election. She is very concientious, asked me questions about concerns. was a skeptic, but had a very positive experience with the machine, and no problems. They did a mock election before they bought the machine, experimented extensively with marking. 1 week before the election they hand tallied 60 ballots, some for all candidates. All came out right. I expressed concern about the possibility of the memory card being programmed to give a different result on the day of election. She admitted that she could not assure that didn't happen. Voters tabulated their own ballots. All ballots and memory card are sealed inside the unit, and it is in their vault.
Concludions:
1) Chain of custody is secure in Ossipee and Plymouth.
2) The only way the total is off in Plymouth is if the card was intentionally programmed to read differently on election day. There was no accidental mix-up in the memory card.
Recommendations: A recount of Plymouth should prove if someone flipped all the votes in the state, intentionally throwing the election.
The only other possibility I can think of is that some counts were tampered with. A count of Manchester should answer that, if chain of custody issues can be answered there. |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 6
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 8:11 am: |
|
Does anyone have any questions about Hand vs Machine count towns at this point? |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4369
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 8:20 am: |
|
quote:Does anyone have any questions about Hand vs Machine count towns at this point?
Yes. Is there a final, definitive list of "approved", agreed official results?
And is there a final, definitive list of ballot-counting methods used in each location?
Have the necessary statistical regressions been done using the final, corrected numbers and locations?
Right now I'm not sure whether any of the statistical analyses should be passed on, since it's not yet clear to me whether they were carried out using the official count results and the correct breakdown of counting methods. |
Shawn MacFarland
Voting Rights Forum Participant
Username: Smacfarl
Post Number: 3
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 8:24 am: |
|
Mark Schauer and Bev Harris,
Given that the numbers at the top of the page do not reflect all the cities that use machine counts, being that they are based on
http://www.bbvdocs.org/NH/state/Jan-08-votingsystems-NH.txt
Correct. I pulled my list over Thanksgiving; apparently they updated Nov. 26. Always confirm with original source data before releasing conclusions on any statistical study.
And that as of this morning I have received an email from the office of the Secretary of State confirming that this is the official list of machine counted cities and towns in NH.
http://www.sos.nh.gov/voting%20machines2006.htm
And now that mark has confirmed that Ossippee is a machine count town, contrary to the slashdot comment. Can we expect the numbers at the top of the page to be revised? And further can we expect Black Box Voting to lead the charge in getting the accurate information out about this to the rest of the concerned voting sites on the net?
Further can we all agree that the analysis of these false numbers has NOTHING to do with anything at this point and should serve as a mea culpa for voting reform sites collective credibilities?
The press has already acknowledged it's over zealousness around NH.
Also this is the 3rd time I have posted on this thread and the other two having gotten no response. I understand that I am a new to this community but that should not negate the information I am offering here. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4370
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 8:48 am: |
|
Hi Shawn,
Just because you don't get a personal response don't assume that your information is being ignored.
I'm eager for Bev to respond to these various questions. |
Brian London
Voting Rights Forum Participant
Username: Brian
Post Number: 9
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 9:14 am: |
|
Shawn, of course any statistical analysis is going to be done with the best data we have at the time. Note that the statistical results prove nothing, and I have done my best during my posts to say that. They only are one piece of a large picture. I will be happy to rerun statistics with the official voting method list, but they will still have all of the same caveats. Also, does anyone know where the most up-to-date, "official" results are?
I think more important than getting better data out is to get better talking points out. This is not about single case of anomalous results. Rather, it's about advocating for a voting system we can be confident in. Results like the ones published here demonstrate the problem of lack of confidence, not a single case of fraud per se.
Converting the entire country to hand counts will not solve the problem. Both machine counts and hand counts can be manipulated. However, since they need to be manipulated in different ways, doing both make skewing an election much harder. The chain-of-custody issue with the physical ballots is a problem too. If voting is so essential to our democracy, why don't we protect the votes with the same level of security that we use on other documents essential to our national security (think satellite assignments or nuclear launch codes)? Failing that the level of security (to prevent tampering) given to DNA samples in a murder case does not seem too much to ask. |
Bev Harris
Board Administrator
Username: Admin
Post Number: 7490
Registered: 12-2004
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Monday, January 14, 2008 - 9:15 am: |
|
The analysis using the latest figures is actually worse, and further, debunks the Farhad Manjoo article.
It should be obvious to any researcher that as a final step, all source data should be updated with the latest version from the most official site. My data was apparently pulled from the sec state office on the day or the day before they updated the list.
If other researchers used unofficial data from Black Box Voting, certainly they would want to verify it with the original source.
We have confirmed that Black Box Voting had 11 misidentified locations as far as hand counts and machine counts.
Do the numbers, folks. The statisticians I have been in touch with say the updated numbers make the anomolies more pronounced. |
Drew J. Jones
Voting Rights Forum Participant
Username: Drewjjones
Post Number: 7
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 10:12 am: |
|
They do, indeed, Bev. A friend and I created spreadsheets containing the entire vote tally for all candidates of all parties from the SoS's website. I'm at work, so I don't have them with me, but, as I recall, we found (rounding)
OptiScan Hand
Clinton 52% 46%
Obama 48% 54%
That's using the official tallies and the official list of towns/cities using AccuVote.
I'll post the spreadsheets when I get home. |
Drew J. Jones
Voting Rights Forum Participant
Username: Drewjjones
Post Number: 8
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 10:14 am: |
|
I'll note that I may have the percentages reversed. It may be 54% for Clinton in OptiScan and 52% for Obama in hand, but I think I had it right above. |
Marjorie Miller
Voting Rights Forum Participant
Username: Marjiemiller
Post Number: 1
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Monday, January 14, 2008 - 10:22 am: |
|
Hi Bev and crew,
Here's a simple idea to do a quick check on the vote. Create an email account for each precinct in the state. Then ask voters to email their vote and their contact infomation to the address, such as, suttonNHvotecheck@gmail.com. This would be a good guide. Canvassers could go around with a laptop to people who voted who have not responded. For instance, if there were 294 voters in Sutton, when you have 294 emails the vote will be complete. You could further automate the recount by perhaps even having email addresses like SuttonNHvotePaul@gmail.com, SuttonNHvoteKucinich@gmail.com.
This final output corresponds with the open-code web-based vote account system I am currently designing and programming. It will record your vote in a secure database, and at the same time, send you an email, and email the county vote email address at the same time. Anytime a "vote account" is changed, the account owner and county, will recieve an email. Perhaps, even a snail mail as well.
Whenever a "recount" is desired, you will then have multiple avenues to pursue, reducing the "chain of custody" corruption problem somewhat, with secondary flags. All anyone has to do is then re-forward the email for a machine recount, which could be automated, or print out all the paper "email-votes" and count them by hand if they wish. This does not preclude recounting the original optical ballots as well.
Indeed, the fact that they do not know how many email votes we have counted for each candidate, and can follow up to get affidavits for votes cast, means that they will not know how many checks we have against their number tampering.
When one balances their checkbook, it is the fact that 2 totals do not match that alerts you to a possible addition or procedural error occurred, with no evil intent whatsoever. All one is looking for is accuracy. Why current voting systems do not do an innocent accuracy check on the 2 vote counts, just to make sure there was no accidental miscount, is unexplainable, and unconscionable, in terms of accountable systems design. Indeed, the more ways to duplicate the counts total, the more reliable the count.
Please let me know what you think of the idea and whatever help I can be in helping this happen. I will increase the priority of my www.votebankaccount.com project, where people can "spend their political capital", just like a bank account. Indeed, ideally, specifying an email account to have a copy of your vote sent to should be part of your voter registration info. If that had been done already, we'd be half-way there now, able to contact all the people who voted and asking them to re-forward their votes for a quick email vote re-count. It doesn't matter if everyone has a computer, just an email account, created if necessary, and access to a computer, somewhere.
Email accounts can be checked against voter registration, and notify the voter if their vote is not counted for some reason, immediately, unlike the 10 days it took Texas to tell me by mail last year, that my "provisional ballot" would not be counted, with no reason listed. I figured it was because I didn't vote for Gov Rick Perry...;)... Snail mail could follow, but at least it won't be long after anyone cares about the election results.
Marjorie Miller |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4373
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 10:25 am: |
|
Drew,
I don't know whether to say thank you or to punch you for putting up figures that might be right and might not!
Aaarrrgh!
Will the real statistics please stand up?!
What is certain is that you and others have independently found and confirmed is that statistics show a significant advantage to Clinton and a significant disadvantage to Obama in relation to counting done on the optical scanners.
Does any significant impact show up in relation to any other Democratic candidates, or in relation to any of the Republican candidates? Who else may have been affected? |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4374
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 10:27 am: |
|
And can someone answer the inevitable question about whether this is just an easily-explained rural/urban difference?
What variables can be ruled out as explanations for this difference? |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 4
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 10:30 am: |
|
I've redone my regression with the correct diebold vs hand count values.
I'm am not a professional statistician, BUT:
1) Diebold machines give Hillary 5.6% more on average (with less than 1:1000 odds of this occuring due to chance alone) after controlling for the influence of total population, median age, and education [% holding bachelor's degrees].
2) Diebold machines give Obama 4.2% less on average (same as #1: less than 1:1000 odds of this occuring due to chance alone) after controlling for the influence of total population, median age, and education [% holding bachelor's degrees].
3) The simple interpretation I gave in previous comments is no longer supported: if anything, hillary gets more votes from diebold machines where obama is weaker rather than stronger (less than 1:1000 odds for this occuring through chance alone) after controlling for fluctuations due to obama's votes, population, median age, education [% with bachelor's] and the simple effect of diebold machines [reported in #1 above]).
Caveat to #3: when I include quadratic effects in the model (meaning that I allow for different effects due to obama's votes on average as I do for districts with extreme obama votes) they are highly significant and go in the more obviously nefarious direction I'd described earlier... BUT that level of complexity is pretty much beyond my comfort-zone for interpreting things. So I'll leave it to the experts from here out, with a STRONG recommendation for them to look at quadratic effects and interactions with the diebold variable.
still looking for exit poll results by district....
(Message edited by neuronomy on January 14, 2008)
(Message edited by neuronomy on January 14, 2008) |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 5
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 10:32 am: |
|
Catherine, my # 1 and # 2 above control for total population, so I don't think this is urban vs. rural. Those effects should have been eliminated statistically if they were to blame.
That was my first question too, and was in fact the only reason I got involved in this in the first place (since it would have been so easy to debunk). |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 7
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Monday, January 14, 2008 - 11:11 am: |
|
I think my numbers are correct and complete.
H-Clinton 15423/46734 = 33.00%
H-Obama 18096/46734 = 38.72%
M-Clinton 97185/238800 = 40.70%
M-Obama 86908/238800 = 36.39%
For similar sized towns (400 1200 Votes)
H-Clinton 34.47%
H-Obama 39.67%
M-Clinton 36.89%
M-Obama 37.65%
Everything is in the spreadsheet. |
Michael D Aasen
Voting Rights Forum Participant
Username: Michaelaasen
Post Number: 1
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 11:30 am: |
|
I see that velvetrevolution.us (http://www.velvetrevolution.us) has posted a $100,000 reward: "we are offering a $100,000 reward for information leading to the arrest and conviction of any person or persons who caused a manipulation of the Diebold machines used in the New Hampshire primary election sufficient to flip the Obama/Clinton race, or information that results in an official change of the first place winner of that Democrat race."
What is blackboxvoting.org position/thinking on this type of reward? I'm convinced that if blackboxvoting.org set up and managed a reward program it would be well funded. |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 89
Registered: 12-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 11:45 am: |
|
Bev, the scoop on memory cards in Massachusetts is that we are required to hold them until it is no longer possible to request a recount. The cards are kept sealed in the machines. The backup cards are kept in a sealed envelope in a vault.
In practice, we send cards out only when they need to be reprogrammed. Ours are going out this week to be programmed for our Feb. 5 primary. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 5
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 12:02 pm: |
|
"still looking for exit poll results by district...."
Chris, I just asked Gary Langer of ABC News (their 'poohbah of polling') for the exit polling data, particularly for the 12 Manchester wards (though to have it for the entire state would be best). As noted above, it is easier to throw an election in larger, more impersonal urban areas, and as others have also noted, the Manchester precincts were among the first to come in and reflected the static 3,500 or so vote margin of difference that held uniformly through election night.
I still think that unless there are people 'on the ground' to physically monitor the actual paper ballots, there's too much of an opportunity to tamper with the evidence. If the fox guarding the coop is the only one to count the remaining chickens, the recount will end up simply validating the initial count.
[edit: didn't see that the video of a Diebold machine being hacked in court was available on the front page of this site.]
(Message edited by Dondep on January 14, 2008) |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 6
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 1:55 pm: |
|
Mark - Your file is missing Westmore, and your file's record for Dixville is 0. If those aren't the correct values, could you update?
FWIW rerunning with your data provides roughly the same results |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 8
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 2:37 pm: |
|
Chris,
There is no Westmore, it is Westmoreland which is in the spreadsheet.
Somehow I got the 7 Dixville votes for Obama in the Dix' Grant row. This does not change any of the % numbers listed above. |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 7
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 3:09 pm: |
|
Does someone have the square milage for each district in NH? It occurs to me that total population may not be a good proxy for urban vs. rural if the districts vary in size. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4376
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 3:24 pm: |
|
Chris, this is exactly the point that Kurt has made on numerous occasions. The relationship between a district's geographic size and its population size may be important.
Could one reduce that relationship to a fraction, and then compare that fraction to the results that came from the two counting methods, or include it in the equation in some appropriate way? |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 8
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 3:51 pm: |
|
Catherine, right, if I had geographic size I would just use a fraction in the analysis - it would be easy. (I'm new here so if there are other shortcomings to my analysis let me know, and we'll see if I can address them).
Thanks to Brian London's file, I *can* use whether each district has municipal water or sewage, which seems like a good proxy for rural/urban.
The aforementioned results are all still significant (at the same < or = 1:1000 level ) after controlling for either or both of them. The more important (non-diebold-related) variables for predicting clinton's votes seem to be household income and education, so other people running stats should be sure to include those as covariates. |
Mary L Burke
Voting Rights Forum Participant
Username: Marytnurse
Post Number: 3
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 4:22 pm: |
|
Mike Labonte, having been on the ground in Iowa, I can tell you for certain that ABSOLUTELY NO ONE was told about internal poll numbers in the offices, even tho we were polling ahead. The internal poll information was kept strictly confidential from us. I suspect this poster was either misinformed or a plant. From whose campaign, I still wonder.
FYI, I've heard from Nancy Tobi, who indicates that she suspects that either no one tampered, or that anyone who may have tampered would have already replaced the ballots as the chain of custody is not under public purvey. } |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 24
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 4:39 pm: |
|
Do we know yet if the recount will include selection of Vice President? |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 25
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 5:05 pm: |
|
Just found this...another statistical analysis of the NH primary done by 3 people with some pretty good credentials.
http://electionarchive.org/ucvData/NH/DemPrimary2008-PairedPrecinctStudy.pdf |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 26
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 5:13 pm: |
|
quote:Does someone have the square milage for each district in NH? It occurs to me that total population may not be a good proxy for urban vs. rural if the districts vary in size.
Here are two maps that may help with your analysis.
http://dyn.politico.com/nhprimaries/nhmap-popup.html
http://www.boston.com/news/politics/2008/nh/nh_primary_gop_results_by_town/
I would guess that you could use some kind of computer program to calculate square mileage for precincts if you know the square mileage for the entire state. |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4379
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 5:46 pm: |
|
Bill, thanks for the link to the ElectionArchive.org analysis. The stats comparing similar pairs of precincts from the same counties are very interesting. |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 9
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 5:47 pm: |
|
My simple-minded analysis of similar size towns is reinforced by the much more in-depth analysis Bill linked to. However, the suggestion that it could have been accidental is, in my oppinion, not possible, at least on a state-wide level. My conversation with the Plymouth Town Clerk convinced me that she thoroughly tested the Diebold machine 1 week before the election, and it was working perfectly. It was then secured until the election. I conclude that there are three possibilities:
1) The count was accurate, and the differences seen are due to demographic differences not corrected for by population density (towns more likely to buy voting machines are more likely to go for Clinton for some unknown reason). Demographic analysis is complicated by the small number of comparable size towns. Several counties are very large encoumpassing very different areas.
2) The memory cards were programmed to give different results on election day. If this was true state-wide, a hand count of Plymouth and/or Ossipee should reveal it, since chain of custody seems to be secure in these towns.
3) Only some (large) cities had errors or were tampered with. I have not yet found out the machine-checking proceedures of major wards. Places to check would be Manchester, Nashua, Concord. Most of the population is in the Manchester/Nashua area.
In my opinion there is no way there was an error or mistake in programming of the memory cards state-wide. |
Mac Hathaway
Voting Rights Forum Participant
Username: Mac_hathaway
Post Number: 82
Registered: 8-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 6:01 pm: |
|
Couple of things:
What you are looking for is population density, which could be a good stand-in for how "rural" an area is. MA has this data on-line. I don't know if NH does or not.
Second, does anyone know for certain whether the recount will be done entirely by hand-counting, or whether the machine-counted ballots will just be machine-counted again?
Bev, and Mike Labonte: I recently got a new job, so my schedule is much tighter than in the past, but let me know what I can do to help with regard to the emailed "call to action" for MA that I saw recently from BBV.
Mac |
Brian London
Voting Rights Forum Participant
Username: Brian
Post Number: 10
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 6:56 pm: |
|
Mac, NH law says that recounts must be done by hand.
Also, my understanding is that MI uses exclusively optical scanners. Is this true? If so, is there anything we should be thinking about or watching for tomorrow during the Republican primary? |
Shawn MacFarland
Voting Rights Forum Participant
Username: Smacfarl
Post Number: 4
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 7:39 pm: |
|
Request to the thread.
Could you all go through and correct the things that are obviously wrong on this thread, at this point, so that misinformation does not get propagated.
* Bev's first post with the Vote totals are wronf because it's based on the bbv list and not the secretary of state.
* The first couple of attachments are also wrong because they use this same wrong data.
* the link to the slashdot posting about campaigning in Ossippee is dead wrong because the votes were machine counted this year.
* Bill Glenn's link to a statistical analysis looks to also be innacurate, because it appears to be based on the bbv list of machine counted districts. I don't know this for sure but the file
http://electionarchive.org/ucvData/NH/2008NHDemPrimComplete-EDA.xls
Found in the same web server directory that looks to have supplied the base data for the pdf is also wrong in that it ascribes the 18 wards that were counted by machine in the last election as counted by hand. Bill Glenn may want to notify the authors of this paper.
I am growing very very concerned that in the enthusiasm to find something there has not been enough time spent to ensure accuracy. Case in point everybody is using the bbv list of machine districts to do their analysis, and that list has been confirmed by the Secretary of State to be wrong. |
Timothy A. Balcer
Voting Rights Forum Participant
Username: Knome
Post Number: 5
Registered: 6-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 7:44 pm: |
|
Folks,
Unless we know exactly how the machines are tested, we can't come to any conclusions about how tampering might have been accomplished such that tests might be foiled. However:
One simple way would be to rig the software to flip votes only after more than a few hundred (or some percentage of the local registered voters) was passed. That way almost any test could be foiled unless you ran as many ballots through as could be expected to be run through during the election. You could compensate for the votes you had NOT flipped during the early period by poaching across the field for the extra to make up the difference.
Also, if such an alteration were timed so that it was only active during the hours the polls were open, that would also frustrate any tests that would be made. There is some disagreement here about how that might be accomplished, however I can assure you that looking at the time is not a CPU intensive procedure.
The beauty of this is that even if we DO find out that the votes were flipped, the damage is already done.. the robbed candidates will get their votes back, however it is easily blamed on "technical glitches" and also, you have already robbed the candidates affected of their momentum. You have hedged into other states, potentially giving your candidate(s) those delegates. NH is small, so it could be seen as a sacrifice ploy. Heck, I would expect the makers to come out with "See? That old hardware is obsolete and breaks alot. You need the new stuff" Mind you, it would give a lot of fuel to the fire for eliminating machines.
And on and on. We can conjecture about this, but it comes down to this: Not only is it possible, it is trivial to alter a vote when any electronic device is involved in either counting or actual voting. I talk about this to other folks and they try to draw me into specifics, or into admitting that I think there is some world spanning conspiracy.. but I always calmly say that it's not about conspiracy.. it's about transparency.. and as long as you have electronic votes, you don't have transparency.
I was asked recently how -I- would hack the vote, if I wanted to. I won't answer here but when I was done explaining how I would do it the person I was talking to (a fellow security expert) said "I'm sure glad you don't work for the bad guys"
I have news for him.. there are other people smarter than I am who would love to take down a few million for some hacking jobs.
Sorry.. I'm just frustrated  I know I am preaching to the choir here but honestly, debating what we -think- may have happened isn't terribly profitable. I believe we should focus on data, and put forward theories as to the comportment of that data, and not get mired in discussions of "was the vote hacked". In the security world you always assume you HAVE been hacked when you look at a system. That's the approach we should take here. We should be doing full on OPFOR and not debating whether or not we need to do the exercise.
(Message edited by knome on January 14, 2008) |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 27
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Monday, January 14, 2008 - 8:54 pm: |
|
(Message edited by Bill_G. on January 15, 2008) |
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4383
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 12:20 am: |
|
Bill,
It is not helpful to focus on speculation.
The first priority is to get the data and all the facts. That in itself will clarify where to look further. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 6
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 2:15 am: |
|
I agree, Catherine (although I agree with your choice of #3, Bill, we could speculate it is much vaster than any of us can fully imagine) and the definition of 'data' in that sense is the actual paper ballots. If I understand correctly, the recount will done by hand, meaning that the minimum to ensure is that the Diebold machines showing the "Hillary Anomaly" are all thoroughly recounted by hand. Those ballots are still in the safes of the town clerks I imagine, but I will take the time to confirm that tomorrow. From that point on, a thorough review of the paper recount by monitors should be the most effective way to verify. However, Hillary's people may be working to have the other precincts challenged as well, despite them matching the exit polling. Even so, a count (first-time count via paper) in full view of the monitor(s) should be sufficient to verify that raw data.
That would mean a minimum of 12 volunteers to witness the hand count of the paper ballots - Manchester. For Nashua, 9. (Concord, interestingly, reflects the exit polls.) For anyone with volunteers on the ground, I compiled this sliver from the list to help prioritize where the most population reside, where the recount would have to show a difference, and where the anomalies occurred:
Atkinson - 1,469 votes
Auburn - 1,098 votes
Barrington - 1,941 votes
Bedford - 4,021 votes
Belmont - 1,305 votes
Berlin - 2,739 votes
Derry - 5,230 votes
Dover - 7,405 votes
Epping - 1,269 votes
Farmington - 1,181 votes
Goffstown - 3,335 votes
Hampstead - 1,760 votes
Hampton - 3,974 votes
Hooksett - 2,600 votes
Hudson - 4,376 votes
Jaffrey - 1,228 votes
Kingston - 1,113 votes
Laconia - 3,131 votes
Lee - 1,336 votes
Litchfield - 1,565 votes
Londonderry - 5,369 votes
Manchester - 20,935 votes
Merrimack - 5,478 votes
Milford - 2,833 votes
Nashua - 17,160 votes
Pelham - 2,484 votes
Pembroke - 1,550 votes
Plaistow - 1,374 votes
Raymond - 1,733 votes
Salem - 5,599 votes
Sandown - 1,002 votes
Seabrook - 1,467 votes
Somersworth - 2,660 votes
Swanzey - 1,582 votes
Weare - 1,591 votes
Windham - 2,439 votes
I'll confirm as soon as possible tomorrow. |
Tim Chng
Voting Rights Forum Participant
Username: Chngthengteng
Post Number: 1
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 4:44 am: |
|
Let us help Kucinich pay for the recount. If we can find 3000 people to donate $25 each, than he will have the funds to pay for the recount:
http://www.dennis4president.com/go/homepage-items/help-defend-the-integrity-of-o ur-voting-system/ |
Mark Schauer
Voting Rights Forum Participant
Username: Marks
Post Number: 10
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 7:12 am: |
|
How do I volunteer to help with the recount. I can spare one day in the next week or so. |
John Belmonte
Voting Rights Forum Participant
Username: Jbelmonte
Post Number: 7
Registered: 6-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 8:23 am: |
|
Regarding the concerns of incomplete/inaccurate data being used in the various analyses, I'll state what I know.
First, many of these problems would have been avoided if nh.gov had provided data in a suitable format such as .csv, .xls, or Google spreadsheet rather than sloppy HTML scattered across multiple pages. Due to the lack of suitable format, several people created spreadsheets at various times, perhaps before all votes were reported, and from various sources (nh.gov, politico.com, concord monitor.com). Since the analysis folks just wanted to focus on analysis, they may have used these 2nd hand spreadsheets instead of grabbing the latest data from a trusted source. I noticed bad data propagating on the 11th and have since spent much effort contacting spreadsheet and analysis authors to stomp it out.
The main problem areas I've seen are old Diebold town data, incomplete vote totals, and a town omitted from the vote data. Unfortunately the data on ronrox/checkthevotes.com page had all of these problems (some have been corrected) and was widely copied. Here are the telltale signs of bad data that I look for:
* Campton not listed as using Diebold (means old Diebold town list used)
* Acworth vote data missing (I think this was due to parse mistake by ronrox site author)
* Windsor has 0 votes (means incomplete tally, I think propagated by ronrox)
Since I haven't created a spreadsheet myself from the official data, nor verified spreadsheets published by others, I don't have a canonical spreadsheet to recommend. If someone would step forward with a spreadsheet that was automatically created from latest nh.gov data, and a couple other people would independently verify the spreadsheet, that would yield something we can trust a little. Otherwise the only responsible thing analysis authors can do is grab the data from nh.gov themselves. |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 9
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 8:51 am: |
|
Not to be discouraging, BUT it's not meaningful to look at vote percentages and "flips" due to machines in "similar towns" as people are doing above unless the precincts are well matched on any demographic that might matter. Therefore I don't think your lists of precincts to prioritize are meaningful. However, these influences can be statistically adjusted for, yielding a much more sensitive and meaningful measure of which towns diverge from the expected more than others.
The following precincts using accuvote have unusually large numbers of Clinton AND unusually small votes for Obama AFTER ADJUSTING for any skew due to variance in education, income, total population, and municipal water [urban/rural]:
Priority-Precinct-"Summed" Skew
1 Seabrook 0.55162
2 Winchest 0.23201
3 Berlin 0.22736
4 Salem 0.22171
5 Sandown 0.22065
6 Wakefiel 0.2038
7 Atkinson 0.20201
8 Raymond 0.20138
9 Plaistow 0.18041
10 Rocheste 0.17726
11 Pelham 0.16831
12 Franklin 0.16694
13 Hampstea 0.16582
14 Danville 0.16514
15 Somerswo 0.16374
16 Hampton 0.15637
17 Allensto 0.149
18 Hampton 0.14576
19 Farmingt 0.14495
20 Hampton 0.14397
21 Newton 0.13764
22 Fitzwill 0.13212
23 Lee 0.12032
24 Gorham 0.11991
25 Epsom 0.11909
26 Milton 0.11385
27 Belmont 0.1107
"Summed Skew" is the extra percentages Clinton seems to have after controlling for demographics PLUS the absolute value of Obama's apparent "undershoot" after controlling for demographics.
That's right, Seabrook is diverging from what we'd expect based on demographics by 50% !. I assume there is something going on there - whether a mistake in our data (this is all with Mark's figures and Brian's demographics, thanks to both) or something more nefarious.
There are more than 27 but it becomes increasingly difficult to tell what's measurement error and what's not below that.
Here is the same list but only for Clinton's votes.
1 Seabrook 0.30301
2 Winchest 0.13898
3 Salem 0.13553
4 Plaistow 0.12304
5 Atkinson 0.116
6 Sandown 0.11495
7 Pelham 0.11285
8 Berlin 0.11071
9 Wakefiel 0.10493
10 Raymond 0.10322
11 Danville 0.10035
12 Rocheste 0.10011
For statisticians: to arrive at these numbers I ranked the sum of the unstandardized residuals of both CLINTON and OBAMA's vote percentages after regression on % holding bachelor's degrees, median household income, total population, and whether the precinct has municipal water supply. I listed only precincts with accuvote machines. The second list is just the rank ordering of CLINTON's unstandardized residuals for precincts with diebold machines. |
Jason Reed
Voting Rights Forum Participant
Username: Jasonr54
Post Number: 15
Registered: 10-2007
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 8:56 am: |
|
Brian:
"Also, my understanding is that MI uses exclusively optical scanners. Is this true? If so, is there anything we should be thinking about or watching for tomorrow during the Republican primary?"
Look for a Romney McCain flip. McCain's their man. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 7
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 9:20 am: |
|
Thanks for the more precise listing, Chris. Almost all of the towns overlap with the list I compiled, which were those Diebold-counted jurisdictions with the percentages aberrant to the exit-polling. Interestingly, Concord was the only one of the 3 "cities" that reflected exit-polling data; both Manchester and Nashua showed the "flip". However, you seem to account for that in your adjustments, but I'm curious: are you saying that Hillary ran 15% ahead of Obama in the more urban and affluent jurisdictions? From the generalized exit-poll data provided by CNN, the opposite I understood to be true. |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 8
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 9:35 am: |
|
Okay, I just got off the phone with the Sec'y of State's office; ALL ballots, from EVERY jurisdiction in New Hampshire, are being trucked to Concord for the recount. That means EVERY SINGLE BALLOT CAST will be recounted, and the PAPER ballots from the Diebold districts will be counted (as opposed to simply recounting the machine totals). The recount will commence at approx. 900am tomorrow morning, at the Archives Building in Concord, located at 71 South Fruit St. Anyone in the public is welcome to come, so I suggest anyone checking in here please go down and monitor the process closely. The campaigns are allowed to name particular monitors (I imagine they get a position close enough to read the tick marks on the ballots, whereas the public just watches the process from a gallery). I'm going to notify the campaigns as best I can, but if any readers know people from inside the campaigns, please notify them and have them contact the SoS's office to insure access.
Good luck, people.
(Message edited by Dondep on January 15, 2008) |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 10
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 9:41 am: |
|
Don: fascinating that we get such strong overlap with such different methods!
It would be very hard to compare those adjusted numbers to exit polling data, because the numbers I report above have been transformed with respect to demographics. So those numbers speak to nothing EXCEPT the deviation from vote percentages that would otherwise be expected due to wealth, urban/rural, education, etc. They certainly don't reflect anything about urban & affluent jurisdictions, since those factors have been controlled for (i.e., equated or "zeroed out" from the analysis).
(edited for clarity)
(Message edited by neuronomy on January 15, 2008) |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 92
Registered: 12-2005
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 9:50 am: |
|
If anyone is still looking for square miles data I posted some files in another thread:
http://www.bbvforums.org/cgi-bin/forums/show.cgi?tpc=1954&post=40172#POST40172 |
John Belmonte
Voting Rights Forum Participant
Username: Jbelmonte
Post Number: 8
Registered: 6-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 9:51 am: |
|
quote:ALL ballots, from EVERY jurisdiction in New Hampshire, are being trucked to Concord for the recount.
This is bad-- I'd much rather have the ballots sit in individual safes within mom & pop town offices and be recounted on location. If all ballots are moved/aggregated it becomes much easier to tamper with ballot sets.
This crucial decision was made by whom? I'm sure we could gather up volunteer citizens to perform or oversee counts in the individual towns.
My naive suggestion: start contacting town clerks in Chris Chatham's ranked list of suspicious tallies and try to convince them not to allow their ballots to be moved-- in the interest of a fair recount.
(Message edited by jbelmonte on January 15, 2008) |
Bill Bowen
Voting Rights Forum Participant
Username: Bbowen8
Post Number: 16
Registered: 11-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 10:01 am: |
|
This is incredible news about the recount in New Hampshire. I've been trying to keep up with as many new threads as possible here, but has a chain of custody been identified for all of the ballots in every precinct? I thought I read somewhere that the state police were responsible for transporting ballots (and I believe Bev expressed concern for organized crime near the Boston areas). Is this true for all of New Hampshire?
Regardless, if this is the way New Hampshire is proceeding, it seems to me that it would be extremely difficult for LHS or Diebold to cover up fraud in every town and precinct if indeed it was incurred by them. Am I being a bit too optimistic, or is this type of recount the dream it sounds like?
Bill |
Bill Bowen
Voting Rights Forum Participant
Username: Bbowen8
Post Number: 17
Registered: 11-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 10:06 am: |
|
On another note, what would have to happen at the recount in order for New Hampshire to overturn the election results, and call for a re-vote? Is there a certain percentage of towns that have to be off by a certain percentage? Is it completely up to the Secretary of State's discretion?
And does anyone know much about Secretary of State Gardner's impartiality? Is he a trusted official, or someone running the state's campaign for a candidate in one of the races!?
Bill |
Bill Bowen
Voting Rights Forum Participant
Username: Bbowen8
Post Number: 18
Registered: 11-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 10:19 am: |
|
Sorry, one more...
Are the optical scan ballots being counted by hand also, or are they being run through Diebold machines again? I'm not sure if that was clear in Don's post.
Bill |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 9
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 10:26 am: |
|
Edit to address the previous question: Yes, ALL paper ballots will be counted. The Diebold machines will be left out and the paper trail ONLY counted (although if there's a discrepancy reason tells us they'll try to go back and recount the machine to account for it).
Good points, Bill; I've no idea off-hand about the Sec'y of State's candidate affiliation, but I'm aware that most of the Dem party machinery is/was geared toward Hillary's camp. Still, for him to grant a recount so quickly augers well for us. We should have our antennae out for possible shenanigans by ANY party, but John brings up the bigger worry: the "chain-of-custody" of the paper ballots. Alas, I doubt we can change the process at this stage, not that we want to, but as I indicated earlier on this thread, any volunteers that aren't sure where to go should go to their nearest town clerk and monitor the comings and goings of any and all, and even more importantly the ballots as they are loaded and unloaded. It's true that there will be more opportunity for the ballots to be "switched out" at the weak links, even by blackmailing one or 3 drivers/attendants that will have custody of the ballots on their trip from the town clerk's safes to Concord, which is taking place this very day as well as yesterday. The 'fix' may have already been made permanent, if indeed there is one. Also, not to forget the ballots cast by those "out-of-staters" who signed forms indicating moving into New Hampshire within a specified timeframe need to be followed up on. I tried calling the Sec'y of State back to inquire about that, but the line has been busy ever since.
(Message edited by Dondep on January 15, 2008) |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 10
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 11:04 am: |
|
I've just gotten off the phone with a handful of television stations in New Hampshire, alerting them to the recount tomorrow - the when and where. What's absolutely shocking is that virtually NO-ONE knew of it; they were vaguely aware that a recount would take place, but that's all. The receptionist at New Hampshire Public Television was aware but was ALSO amazed that there is no coverage by the media at this point. I called Manchester Community Access and the programming director was unaware of it, but after talking to him he said he would call Sec'y Gardner's office and may send a camera crew down. If anyone else reading and posting here can help "alert the media", it would go a long way to having the process monitored publicly.
Here's a link to New Hampshire's television stations:
http://www.directorynh.com/NHAdvertising-Media/NHTVStations.html
"It's Up To Us"; it's only now dawning on just how real that statement is. |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 11
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 12:11 pm: |
|
Wow Don, that's great!
Mike: thank you for sq mileage, but I cannot cleanly align those mileages with the precincts (there are some towns whose names don't seem to match up with the precinct list I've got). ALso, given that there is not a one to one relationship, I am not sure what to do with that data.
IMO population density correctly-aligned with each district is still the "most needed" variable for statistical analysis. |
Tom Courbat
Voting Rights Forum Participant
Username: Leftisbest
Post Number: 79
Registered: 6-2006
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 12:54 pm: |
|
UPDATE:
Kucinich has provided the necessary funds by the 3 p.m. EST deadline today to begin the hand-recount of all democratic ballots tomorrow. There had been some question as to whether the fundraising effort had come up short at the last minute. Fortunately, Nancy Swett of the NH SOS office confirmed at 3:03 EST today that the funds had been received. She was not able to confirm or deny if the funds to conduct the hand-recount on the Republican side had come in yet.
So, the Democratic Party recount will begin tomorrow as scheduled!!! |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 94
Registered: 12-2005
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 1:40 pm: |
|
Chris, I can fix up those files so the town/city names line up, with 1 area value for each town/city. I would just need to know the names you want. |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 12
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 2:00 pm: |
|
Mike - I've attached my list. You're awesome! |
Mike LaBonte
Voting Rights Forum Participant
Username: Mike_labonte
Post Number: 95
Registered: 12-2005
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 3:50 pm: |
|
Chris, here are the NH municipal areas, fixed up. Man, that took a lot of PDF lookups! One glitch: your list has Wentwor1 and Wentwor2. But only Wentworth matches, so I left Wentwor2 blank.
|
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 13
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 4:51 pm: |
|
Mike, I've rerun the analysis with population density as a covariate (THANK YOU!!!) and the results are unchanged.
Ironically both population and population density seem to be relatively poor predictors of Hillary's votes when controlling for other factors, contrary to media reports.
Newcomers to this discussion may find the following description and consolidation of data interesting:
http://scienceblogs.com/developingintelligence/2008/01/the_diebold_effect_hillar ys_vo.php |
Tom Courbat
Voting Rights Forum Participant
Username: Leftisbest
Post Number: 81
Registered: 6-2006
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 5:41 pm: |
|
Discussion going on right now on www.toginet.com. Will continue until 6 p.m. PACIFIC time. |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 28
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 7:40 pm: |
|
Perhaps another anomaly...
As we know Obama won 52.95% of the hand-counted vote (in precincts with a smaller average population) and Clinton won 52.95% of the computer counted votes (in precincts with a higher average population)...yet..
Obama won 64.84% of the hand counted vote precincts
and
Hillary won 60.82% of the computer counted vote precincts
This seems very very odd to me...i would expect that Hillary would have won a much higher percentage than Obama in the computer counted vs. hand counted precincts.
Do you follow me? The margin of error should decrease as population size increases.
(Message edited by Bill_G. on January 15, 2008) |
michael d. betts
Voting Rights Forum Participant
Username: Nonnod
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 8:28 pm: |
|
I've been looking over all the great data analysis. You folks are great. On primary night I could see something wasn't right and I started doing my own analysis. I'm no statistician (I can't even spell it)(is this correct?). Anyway I just looked at my notes from primary night and I have a theory that if someone wanted to "fix" the results (via Diebold) they would avoid the smaller towns because a group of 100 might get together after the election and say "we all voted for Obama and the machine only counted 99 votes" However, I don't think this is likely to occur in larger towns. My rough figure (I rounded off) say that
Clinton won by 7000 vote and picked up large percentages of of votes in the larger towns:
Berlin 666
Derry 700
Hampton 400
Hudson 650
Manchester 2500
Nashua 1800
Salem 1300
I'm stopping that 8016 votes right there.
Also, these large Clinton percentages are out of line with the rest of the state
Ex. Salem Clinton-2867 Obama-1508 (that's nearly 2to1
Nashua Clinton-5814 Obama- 4068
nearly 3 to 2
Most of the medium size town don't have such large differences either way (clinton over obama or obama over clinton)
My theory is that if votes were flipped they were flipped in several large town (to avoid scrutiny) has anyone looked at this?
Mike |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 14
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 9:53 pm: |
|
Someone brought up an interesting point: what is the ordering of candidates on the diebold machines vs. the paper ballots? In psychology we always counterbalance the order of options because there are strong order effects. They could conceivably explain these results, unless candidate order is randomized or counterbalanced. Anyone know? |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 11
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 10:54 pm: |
|
Chris, I've heard about the "psychological effect" about ballot placement, but despite the fact that there is something to be said about the order in which names appear, in reality it has minimal if any impact at all on the vote. When this aspect would come up in discussions of the vote count in my precinct, it would only become a recognizable factor when the IQ of the voters were in question. Seriously. Even in doing anecdotal 'exit polling' at my precinct, it was an insult to virtually anyone who was questioned about it - as you might imagine. The last time a pundit addressed it was when it was given a .3 (that's point 3) advantage by someone on MSNBC, Chuck Todd I believe. (I may be wrong on the individual.)
michael, your analysis closely reflect my own. I call the skewed areas you listed the "Hillary Anomaly". While I want to shy away from speculating too much (see our discussion above), often it does help if one looks at the same analysis from a theoretical POV and see if the anomalies are more or less logical. In this instance; if there was a serious attempt to hack the entire state on behalf of one candidate in as believable a result as possible, as opposed to say a technical flaw in certain machines, these numbers are more logical. One would hide the flipped margin in the larger more vote-rich towns, as I indicated before. Still, the hackers would have to be very careful; for example, of the 3 largest urban areas, Concord alone reflects the exit polling data. The other two, Manchester with 12 wards and Nashua with 9 wards, absurdly reflect the "Hillary Anomaly" that bucks both the tracking and exit polling data. That's why Chris is finding "Ironically both population and population density seem to be relatively poor predictors of Hillary's votes when controlling for other factors..." as he says above -- when looking for factors that 'prove' the statistical reliability of Hillary's margin, as opposed to factors that prove out the tracking and exit polls favoring Obama.
Knowing how deadly, deadly serious this issue is, if true, then it shows how extremely important it is to monitor the "chain of custody" so that true old-fashioned "ballot stuffing" can't take place. My guess is that efforts are underway even now to "switch out" a paper trail that would correspond with the machine vote and thereby cover the "Hillary Anomalies". |
David Luebbers
Voting Rights Forum Participant
Username: David_luebbers
Post Number: 47
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Tuesday, January 15, 2008 - 11:27 pm: |
|
Yea that someone was me a couple days ago.
Lol.. I made that comment based on something i read about how the top people on the list usualy get 2% average more votes soly because they were near the top of the list
(People just picking the first name they recognize)
"It doesn't matter who votes. It only matters who counts the ballots." Stalin
|
Catherine Ansbro
Frequent Voting Rights Forum Participant
Username: Catherine_a
Post Number: 4389
Registered: 12-2004
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 1:08 am: |
|
quote:I made that comment based on something i read about how the top people on the list usualy get 2% average more votes soly because they were near the top of the list
I believe this is a well-documented, reproducible statistical effect. Some places require that ballots (or listings on DREs) be divided into segments which each have the candidates in a different order.
Perhaps the statistical analyses could incorporate this factor as well. |
Jim Van Zandt
Voting Rights Forum Participant
Username: Vanzandtj
Post Number: 1
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 4:10 am: |
|
I voted in Nashua, using a mark-sense ballot. The candidates were listed in alphabetical order. I expect the order makes a difference in local races for obscure posts (e.g. registrar of deeds) where many of the voters haven't heard of any of the candidates. In this case there were a lot of names to look through, but I doubt anyone had trouble finding the right one. Also, any attempt to level the playing field (e.g. listing Clinton third in Manchester but 13th in Nashua) would present lots of opportunities for error (the machines would have to be programmed differently). |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 29
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 6:12 am: |
|
Don Depeller wrote:
quote:Knowing how deadly, deadly serious this issue is, if true, then it shows how extremely important it is to monitor the "chain of custody" so that true old-fashioned "ballot stuffing" can't take place. My guess is that efforts are underway even now to "switch out" a paper trail that would correspond with the machine vote and thereby cover the "Hillary Anomalies".
That would be next to impossible if the recount includes the choice of Vice President. It appears to me that this isn't a case of simply flipping the voter profile in select precincts... so that makes substituting paper ballots extra challenging.
My biggest fear is that the paper ballots would be stolen. |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 30
Registered: 1-2008
Best of Black Box?
Votes: 1 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 7:31 am: |
|
Can somebody with some computer skills cache all the exit polling data from the National Election Pool (which is a consortium formed by NBC, ABC, CBS, CNN, Fox and the AP)
http://www.edisonresearch.com/aboutus/index.php
Here is the exit polling data from the National Election Pool for the 2008 Democratic Primary.
http://www.msnbc.msn.com/id/21225995/
If you want a little backround on the NEP I suggest reading this article:
Footprints of Electoral Fraud: The November 2 Exit Poll Scam
by Michael Keefer
http://globalresearch.ca/articles/KEE411A.html
If the 2008 NH primary was rigged I have no doubt that the National Election Pool worked in collusion with whatever organization rigged the election. Take a look at the results of the exit polling data and see for yourself.
The data seems very weird across the board. For example, look at the question "What should the U.S. do in Iraq?" The exit polling data shows that Hillary Clinton is heavily favored by voters who want to pull the troops out as soon as possible (41% to 34%) while Obama is strongly favored by voters who want to keep troops in as long as needed (51% to 24%). But then when asked about what issue is the most important to you...voters who said the war in Iraq was the most important issue supported Obama (44% to 35%).
Also, when minority women were asked who they voted for...50% chose Clinton over 38% for Obama. This makes absolutely no sense considering how badly Clinton did in Michigan with minority voters. |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 31
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 7:47 am: |
|
quote:Kucinich hands over $25,000 to start NH recount
http://www.unionleader.com/article.aspx?headline=Kucinich+hands+over+%2425%2C000 +to+start+NH+recount&articleId=c5b584b0-6965-4ded-bdd4-b554a0a11399 |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 15
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 8:11 am: |
|
@Catherine, I agree that we should incorporate order effects - <strikeout>what if they are exaggerated on voting machines since people are unfamiliar with them?</strikeout>
EDIT: I forgot all ballots are paper and then scanned; so the "less familiar" explanation need not be considered. Grabbing at straws here for an innocuous explanation of the effects..
(miscellaneous updates: controlling for county membership and/or whether each precinct had a relatively close or distant race between Obama & Hillary leaves the diebold effect significant at p<.001).
(Message edited by neuronomy on January 16, 2008) |
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 16
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 9:54 am: |
|
One more: even if there aren't large differences between the recount & the original count, we may be able to statistically mine this for differences after controlling for demographics. Anyone know if we will get the recount data at the precinct-by-precinct level? |
Arkadiy Belousov
Voting Rights Forum Participant
Username: Arkadiy
Post Number: 1
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 10:25 am: |
|
Here is the exit poll data, cut and paste from Firefox to Excel using Table2Clipboard extension.
|
Chris Chatham
Voting Rights Forum Participant
Username: Neuronomy
Post Number: 17
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 10:57 am: |
|
Arkadiy, thanks, but for stats I need the precinct-by-precinct breakdown. If there is such a thing, it has not been released to the public; we'll need an inside source in all likelihood. |
Arkadiy Belousov
Voting Rights Forum Participant
Username: Arkadiy
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 11:23 am: |
|
http://www.exit-poll.net/
>>>>>>>>>
Any news organization in the United States and around the world can subscribe to the election exit polls for live Election Night data feeds. Academic institutions and other interested parties can purchase data for analysis after the election.
<<<<<<<<<
I wonder if per-precinct data is available for sale. |
Bill Glenn
Voting Rights Forum Participant
Username: Bill_g
Post Number: 32
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 5:34 pm: |
|
Arkadiy Belousov wrote:
quote:Here is the exit poll data, cut and paste from Firefox to Excel using Table2Clipboard extension.
Thanks...I'm wondering if it is possible to do a PDF snapshot of the exit poll just so we have a record in case they decide to pull it off the web. |
Kevin Davis
Voting Rights Forum Participant
Username: Millman
Post Number: 2
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 7:30 pm: |
|
Why isn't this stuff all over the news??? This is BIG STUFF! And why are these machines used at all? This election is being stolen and there's nothing we can do about it!!! |
Jody Hoelle
Voting Rights Forum Participant
Username: Chandidevi
Post Number: 3
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Wednesday, January 16, 2008 - 9:51 pm: |
|
McNeil Lehrer, PBS, interviewed Debra Bowen, California Secretary of State, about the machines in California. Apparently, there is concern here since the hand written ballots are processed through an optical scanning device which can be tampered with. Now What? |
Don Deppeller
Voting Rights Forum Participant
Username: Dondep
Post Number: 13
Registered: 1-2008
Best of Black Box? N/A
Votes: 0 (A keeper?) | | Posted on Thursday, January 17, 2008 - 12:08 am: |
|
"Now what?" If that's the case, we've been had. If the hack was done to generate the same paper trail to reflect the machine, the recount won't change a thing. Do voters get to review the paper ballot, and see the names next to the tally? And as I said on the new thread Bev started with her report on what happened today (rather yesterday, as it's about 1am here), if there's been ANY TIME AT ALL the paper ballots (assuming they accurately reflected the voter's intent and not a hacked choice) have been out of the custody of a state elections employee and a 'reputable witness' (NOT an LHS employee), the the results could still be tainted. The recount data for Manchester is showing, for the most part, an extra 1 to 5 votes for nearly EVERY presidential candidate.....but without changing the 'official' results in any substantive way.
We may be engaged in an exercise in futility, if any thieves have covered their tracks already.
Chris, in case you hadn't heard, the exit-polling data by Edison Research (that's provided to all the media, see above) is "weighted, post-election, to make up for discrepancies." Wow; when I read that casual remark on the Edison link, I just felt the life go out of me if they could be that 'willfully negligent', as Bev put it. One can only now point to how early the other races have been called, with projections based on less than 25% of the votes cast! Yet the Obama and Clinton race was held over because they kept expecting the exit-polling results to show up, and they never did. They may do the "raw exit polling" but never want to relinquish what those numbers exactly are......unless of course you can 'pay-up'. |
NH Town Demographic Data
Excel Spreadsheet
NH Districts for Square Mileage