R-71 UPDATE: The error rate on signatures for R-71 has continued to climb throughout this week. The following is from a report filed by the Secretary's office.
The daily update showed these cumulative totals: 88,191 checked, with 77,637 accepted and 10,554 rejected. The rejection rate has now risen to 11.97 percent. We can sustain an overall error rate of no more than 12.4 percent to make the ballot.
Open Letter to Secretary of State
I have included two very important letters. The first is a letter written by Larry Stickney with my full concurrence on the facts and the conclusions. It includes notes from our observers.
I have also included a letter from attorney Steven Pidgeon which was sent to the Secretary of State last night.
The communications are very candid and we feel, accurately describe the current conditions in the processing of the signatures we delivered to the Secretary's office.
Click here to read letter to Secretary of State
Open Letter to Friends of R-71
From Larry Stickney and Gary Randall
Thank you to the many thousands of you who have responded to our plea to contact Secretary of State Sam Reed's office. While that office has made a furious defense of their R-71 signature counting process since then, we continue to see a lot of unaddressed problems as the numbers continue to move in the wrong direction for R-71.
Even left-wing blogger/demographer Daryl Holman of Horsesass.org has expressed curiosity and surprise at some of the statistical abnormalities of the count in recent days. As Daryl notes in his Wednesday article (Shifting fortunes for R-71?), "This could, conceivably, reflect reduced effort by signature checkers to find the signers in voter rolls." In Thursday's article, he wrote the following: "in the last several days, there appears to be an unexplained, systematic increase in the rate at which signers are not found in the voting rolls."
Those of you who e-mailed Sam Reed may have received a response from Sam's executive secretary, Ms. Ingrid Pharris. She is, interestingly, the daughter of James Pharris, the attorney from the state AG's office assigned to defend the right of the militant homosexual organization (WhoSigned.org) to post the names and addresses of R-71 supporters on the internet so other militant homosexuals can harass them. Ingrid's letter was sent to assure you that the Secretary of State is going out of his way to make sure that the referendum count is being carried out in "a fair, transparent manner".
Here is an example of one of the statistical aberrations that has many of us scratching our heads: Of the first 72,273 (Wednesdays count) signatures tallied, only 9.5% were rejected because petition signers name was not found on the voter data base. Of the last 6,922 signatures counted the percentage jumps dramatically to 12.8% not found!
Meanwhile, the Secretary of State continues to tell us that the increase in rejection numbers are due to the increase in duplicates found. However, the increase in duplicate signatures has only increased by 0.52% over the last week whereas our rejection rate has increased by 1.29% during this same time period. Today, 8996 names were counted, which easily surpasses the normal rate of 7500 or so by about 20%, yet the Secretary of State assures us that they "did not decide to speed up the process".
To put it bluntly, the "spin" we are hearing from the Secretary of State's office is deception pure and simple. Larry was personally assured just last week after a meeting with Secretary of State official, Shane Hamlin, that they would not deploy more than 4 master checkers at a time. This is an important consideration because we are allowed only three observers on the floor at a time. When you have more than four master checkers at one time, the observation team of 3 simply cannot keep up with what is happening. We are now told that for most of this week, they have been using 6 to 8 master checkers at a time. No phone call or explanation came our way from the Secretary of State, just more frustrated messages from our observation team, wondering how in the world these folks get away with this.
Incompetence or skullduggery?
One issue the Secretary of State's communications team has not been anxious to write about in detail, is that they began the count utilizing an outdated voter data base. The effect? New voters who registered between July 2 and July 25 didn't show up. Protect Marriage Washington had sent out several thousand voter registration forms along with the R-71 petitions and signed up many new voters that were not picked up by the initial checks! Here is the story you haven't heard: The problem was discovered when one of our observers saw one of the R-71 petitions that just happened to be the one several of Larry's family members signed. When our eagle-eyed observer contacted Larry and told him that his daughter's signature was rejected, Larry and wife Polly remembered that their daughter had registered to vote in July and that she had her voter registration card to prove it. Larry brought this up to Shane Hamlin and was told that he would look into the matter. Shane sheepishly reported back to Larry several days later that he had "discovered" the older voter database was still being used and that the Secretary of State would be looking into ways to change over to the new one.
Fortunately, this has been a blessing in disguise as all of the invalidated signatures must be run through for a third check against the updated data base and we stand to pick up a large number of signers not validated before. Unfortunately, the Secretary of State decided to limit the third check to a quick check as opposed to the standard in-depth search utilized when counting other signatures. We believe that many signatures will be discovered on this third check, but even more would be found by employing the standard check process. Why the quick check, Sam?
Our concerns are based not only on provable numerical abnormalities, but by those raised by members of the Protect Marriage Washington observation team, who are working tirelessly for the cause on the front lines of the R-71 battlefield in Olympia.
Please take the time to review some of their comments at the end of this letter. Their eyewitness accounts of the hasty signature checking process and the negligent, uninspired, and shoddy work of some of the checkers is a reoccurring theme and in complete contrast to the rosy picture painted of the counting operation by Ms. Pharris and the Secretary of State's spin doctors. Sincerely, Larry Stickney, Protect Marriage Washington
Gary Randall, Faith & Freedom Network
Comments by Observers
"I have seen firsthand how thorough investigating finds many valid signatures that are being ignored because of the rushing. Most of the checkers are college kids; I have seen the same kids spend 2 minutes investigating a signature, finding and validing it; and after they come back from a break complete with a supervisor pep talk, they resume their job spending less than 30 seconds - and rejecting the signatures. It is very frustrating. If the kids are encouraged to take their time and investigate, they find 90% of the signatures to be valid. When they are encouraged to rush because of a deadline, they only do a cursory look and then reject them. The bias is palpable in the room. They are now spacing the master checkers further away from each other so that we cannot watch them. Please call the number listed below and insist that they slow down and do a thorough investigation of all the signatures. Then pray. Thanks!"-- K.M.
"On 8/18/2009 I have volunteered to work as an observer for the R-71 signature verification. I immigrated to the US from Ukraine in 1998 being 20 years old at the time, so I am fully aware of the culture and how names in Russian are spelled out and translated into English. As I was watching one of the workers on the 3rd final check, I took notes for any mistakes in the process. I have noticed at least three cases when names of the people of Russian/Ukrainian (Slavic) origin were wrongly rejected. On one instance, the first letter of the last name was incorrectly read resulting in no database findings; I could see it from the Russian signature that the first letter was misunderstood. On the second instance, the Russian signature was only a half comparing to that of the database with the identical handwriting, the address was different though and as a result, another name got rejected. On the third instance, the name was found in the database, but the hand writing was different. Now, we have to remember that Slavic people who got used to write their signature in Russian a certain way for many years, may change their handwriting in English because not all of them really know English, so, being afraid to make a mistake, they spell out every letter resulting in a handwriting difference. Another point I want to make is that signatures in Russian don't have to include both first name and the last, it could be the first letter of the first name together with the full or partial last name, or simply the full or partial last name alone. And the last thing I want to say that, unlike American names which could be more than a few of the same in the database, when you come upon a Russian name, it is very likely that that is the very person who wrote it down, and there is no other with the same first and last names in the state of Washington. So even if the handwriting doesn't completely match, it would still be that same person."--A.S.
"It is interesting that "their" search page is set to 25 results per page which is an ARBITRARY number. Example: Say John Smith is being searched for his signature. He registered say King County...then relocated to Spokane...there OBVIOUSLY is going to be MORE THAN 2 PAGES at 25 per page, therefore, JOHN SMITH would NOT be found in King County ....if he moved to SPOKANE...and although he IS a registered voter...by this policy set today...his registration would NOT BE FOUND..Hence, this new "policy" appears.... (deliberately or mistakenly)....to lead to missed and/or INACCURATE search results."--V.H.
"Thank you for giving me the opportunity to help you last night. I wanted to give you my opinion and great concern about the different checkers I watched last night. L.N searched very well, using just 1 letter of the name or the number of the street address and tries lots of combinations. I thought she is a good Master Checker. P.T is very quick with the computer, but was the most thorough Master Checker I watched. He tried over and over again, probably 12-15 times to find a name. He took into account changes in signatures over the years. I liked his work too. I began watching M. at 8:58 and was VERY concerned with her searching. She should NOT be considered a Master Checker. After watching her do 7 searches in about 10-15 minutes, I was so shocked I started documenting exactly what she was doing. She may have been tired because it was so late, but I really think it has more to do with needing knowledge of searching a database. Her first search was always a full, first and last name search with her best guess as to the spelling. Then she would try the whole address alone. Twice I saw her get 8 and 9 exact name matches but she didn't check but 2 of them for signature matches. It seemed she was looking for the names to match the address. I jotted these thoughts down on my paper. Not thorough, very few searches, does not search by initials, fast, not enough combinations, usually full name searches (first and last), only tries once or twice on some, always used full name (all letters) except where noted in my notes. Although I think she is trying her best, she is a great concern to me. If I was to talk to her supervisor, I would say I don't think she understand how to search a database, because she is using all of the letters of first and last names, and that she seems to be looking to match the name to an address. We will be contacting the men coming up tonight to bring these papers up to you. Thank you."--S.K.
While observing the signature verification process last evening, I observed 2 noteworthy problems.
1) In two instances there were problems with the scanned images in the state's database. In the first instance, all of the voters data was a match - his name, address, etc... However, the scanned signature was for that of a woman with a completely different name (wrong signature scanned in).
2) The voters signature card was placed too low on the scanner so that only the top tips of the letters were visible, not sufficient to confirm the signature. Otherwise, all of this voters data was a match.
In perhaps 4 - 5 other instances, I felt that there was a sufficient match to confirm the signatures, but they were rejected. Most often there were people of the same last name at the same address, but not the voter (signature) in question. In looking at the date of registration on the bottom right of the screen, I could see that they were relatively new voters (registered within the last 3 years).
I also observed a couple of other rejections of signatures also of relatively new voters of which I felt there were adequate grounds for acceptance - unique or distinguishing attributes in the formation of certain letters. In these cases, all of the voters data matched and the signatures were rejected. It is likely that from ages 18 to 25, a persons signature will change somewhat as they have to sign an increasing amount of documentation after entering the workforce and (presumptuously) independent living,. They might consider age when confirming signatures if all other data is a match.--J.M.
Here are my thoughts and observations from Aug. 19, 2009 session, approx. 7 p.m. through 9:53 p.m. I was observing master checker P. the whole time. On the whole I am favorably impressed with this checker. He uses both single and multiple database entry points and gives very conscientious scrutiny to all questionable signatures, following all the election divisions guidelines for comparison. Nonetheless, I sensed that fatigue during this day's session might have lessened his performance somewhat.
Notes on Red marks upheld:
1. On Vol. 224, Page 7, Line 13, female name: There was a male of the family name at the address given on the petition. It was possible that the petition signer was newly married and newly living at that address. A search was made by first name, but checker decided there were too many entries of that name to search for signature match. When a search was made just by family name, I saw a middle name of the same initial as the first name go by, but far too fast to see anything else, and no stop was made to check the signature with that registration entry. He went so fast, he must have been looking for only one other factor, perhaps just the first name itself. (I believe that the middle initial possibility should have been checked because when signer married, perhaps she stopped using her given first name as her legal first name and started using her middle name as her legal first name, the same thing that I understand my own mother did.)
2. Vol. 224, Page 5, Line 10: Checker tried only one interpretation of the street number, whereas I considered there were grounds for two interpretations. The use of only this number in conjunction with other data points means that the other possibilities using the other possible number were excluded.
3. Vol. 216, Page 15, Line 15: The petition signer's first name was different from the printed name shown on the registration, but it appeared to me that the signature on the registration was indeed the name signed onto the petition. I suspect that an error was made in entering the man's first name into the database somewhere along the line, and this signature should not be excluded.
Other notes: Vol. 224, Page 3, Line 7 (kept green): While the petition signature does not look much like the signature on the registration, the printing of the name on the petition looks very much like the signature on the registration, suggesting that this person's signature has simply undergone considerable change since his registration. I hope these notes are clear. Yours in Christ, --J.B.
These are my observations from the morning shift on August 13, 2009.
I observed master checker Z. checking Volume 227 and did not observe anything unusual. Most of them he confirmed as rejected and reversed only a few. Page 15 included many Russian names, and of 7 initially rejected, only one was reversed and accepted.
I also watched master checker M. as he worked on Volume 228. On page 9 he rejected line 16, which looked to me like a match. His numbering on page 13 was different than the actual petition; his line 20 was the petition's line 19.
They both finished their volumes about the same time (around 9 a.m.); M. took a personal break, and when he returned he was doing initial checking. Z. began master checking Volume 229. On page 2, line 5, he rejected this signature, which looked to me as though it matched. On page 10 he accepted a previously rejected line 2 after consulting with M.; the non-sponsor observer took lots of notes when this happened. At about this point T. said she needed 2 guys; she chose Z. and C. Z. went over to her to see what she needed, and she told him to go back and finish his volume first. He started working faster so he could finish up. Page 13, line 20 appeared to me to be a match but was rejected. A lot of the names on page 13 were Russian. He especially rushed on page 15, taking 20-30 seconds to check line 2 before he rejected it.
I walked around to find another master checker to observe. I watched N. for a few minutes while T. gave instructions to Z. and C., which I did not hear. Z. then took a computer station next to C., so I went over there to observe. They both had volumes that already had 2 signatures next to each rejected line. I tried for a few minutes to figure out what they were doing, unsuccessfully, then went to the break table to get my sweater. I mentioned to V., who was working at the break table, that they were doing some sort of triple check, and I went back to resume my observations. The non-sponsoring observer was standing between the 2 checkers, clearly crowding them, leaning in, and obstructing my view. I tried moving around to get a better view, but through my glimpses I could not figure out what they were doing; the screen didn't match any of the rejected names, and I couldn't get a better view to observe any closer. This took place about 10 minutes prior to the lunch break.--B.C.
My experience today was primarily with two checkers who were doing the second check on initiative signatures. One of them was considerably faster than the other which led me to question why. Was the reason for faster speed due to knowledge and skill or was it due to negligence. The appearance of things was that the quicker man did not employ as many means for finding a signature and also did not spend as much time studying the signatures he found for similarities. Should you have any further questions please let me know.--D.C.
When I observed M., I noticed some things he did differently than some of the other master checkers. One for example Is he would put his hand over the signature lines to make it harder for me to see the signature line numbers. I don't know that he was doing it on purpose, but it seemed strange.
He also would not search in a consistent pattern as I saw some of the other checkers do. He seemed to search thoroughly on some, and others go fast through the checking. He also seemed to have more acceptance of rejects then some of the other checks like D. or N. I also saw him search for names that would bring back huge results, then just use his down key to scroll through the records, but not stop long enough to analyze. It seemed like he was trying to fill time. He would also stop and look at something for awhile, that was not even close to a match. These are just presumptions I made as observing. When I observed M., I noticed some things he did differently than some of the other master checkers. One for example is he would put his hand over the signature lines to make it harder for me to see the signature line numbers. I don't know that he was doing it on purpose, but it seemed strange. He also would not search in a consistent pattern as I saw some of the other checkers do. He seemed to search thoroughly on some, and others go fast through the checking. He also seemed to have more acceptance of rejects then some of the other checks like D. or N. I also saw him search for names that would bring back huge results, then just use his down key to scroll through the records, but not stop long enough to analyze. It seemed like he was trying to fill time. He would also stop and look at something for awhile, that was not even close to a match. These are just presumptions I made as observing.--J.
1. The average check is NOT 3 minutes but 1.5 to 2 minutes that includes MORE than just the Russian and Asian names.
2. Some of the newly hired checkers from last weeks hiring trained only ONE HOUR (we were present for their orientatioin and training!) are now being used as MASTER CHECKERS.
3. The ...seasoned...regular....on call college age checkers are allowed to listen IPOD's that have been loud enough that I can hear it while standing next to a checker...and I have a partial loss in my rigtht ear! This is going on while they are checking signatures. This is NOT democracy in acton.
4. Last week an observer reported to P. and I that he witnessed and documented a female checker "singing" out loud with her IPOD on unaware she was disturbing the other checkers. TWICE she was told by the other checkers to stop singing. The Supervisors have done NOTHING about this....INCLUDING Ms. K. B. who SCOFFED at this when we tried to raise this serious concern of accuracy!
5. "K" a master checker (I personally documented racing through master checks was rocking and bouncing with one knee, tapping his pencil on the desk with his right hand while listening to his IPOD spent an average 2 minutes or less on his master check. All his average 1 - 2 minute checks were REJECTED. As of yesterday, S. (Director) has not upheld his promise to keep the Master Checkers to a minimum of 4 "sitting together." When asked by me personally on Monday that the Master Checkers are not sitting together...Shane replied, "I relaxed this with Teresa last Friday and as a matter of FACT I am going to be INCREASING the master checkers this week. He has.... Personal documentation from Pastor R this am:
8:10 - 4 MASTER CHECKERS
9:15 - 5 MASTER CHECKERS
10:53 - 6 MASTER CHECKERS "MASTER CHECKERS ALL SPREAD OUT - 2 ARE DIAGONAL FROM EACH OTHER, OTHERS SPREAD OUT. VERY DIFFICULT FOR OUR OBSERVER'S TO KEEP UP WITH AND WATCH."
Secretary of State's office did not keep their word to accomodate our observers
1. S. did NOT inform us he had "relaxed' this [policy) with T.!
2. The Master checkers are scattered all over...sitting where they choose,
3. And, the Master checkers as of Monday has INCREASED to SIX - AGAIN.