R-71 checkers process over 5,900 more signatures

R-71 checkers process over 5,900 more signatures

State election workers have now checked over 35,000 Referendum 71 signatures, including 5,926 during the Friday day-shift. The daily rejection rate was 12.91 percent, which was higher than the newly revised cumulative error rate.

R-71 backers, hoping to force a public vote this fall on the state’s new “everything but marriage” domestic partnership law, need 120,577 valid signatures to earn a ballot spot.   They submitted 137,689, and Friday was the sixth day of checking by election crews.  The day shift ended with 5,926 new signatures processed, with 5,161 accepted and the rest either rejected or pending determination. The rejected signatures included 604 people who weren’t found on the state voter database, 33 duplicates, and 116 whose signature on the petition didn’t match the one on file.

Overall, the totals showed 35,296 checked so far, with 31,199 accepted, 4,063 currently rejected and 34 pending a response from their home counties with an electronic signature for the voter. 

After adjusting for a variety of changes made this week by senior checkers who review the work of the crew, the state Elections Division adjusted downward the percentage of signatures currently rejected, not including those pending the electronic signature from the counties. The new rate, as of the conclusion of the Friday day shift, was 11.51 percent. A crew was also working a swing shift later Friday. Election officials hope  to complete the full check by August 17 or thereabouts.   

For more information, check in at the division’s webpage .

26 thoughts on “R-71 checkers process over 5,900 more signatures

  1. Why is Washington State trying to rig the election? Everyday the numbers and math keeps changing so as to validate more signatures so this anit-gay ballot initiative gets on the ballot.

  2. Charges of dishonesty and the like are nothing more than overblown internet rhetoric. Let’s give that no weight.

    However, there is one very valid question, which was posed several times yesterday and which Mr. Ammons has pointedly not addressed: If a master checker is reviewing the initially disqualified signatures, why is he not also reviewing the signatures initially determined to be valid? It does seem to be unfair to give a double-check (and thus a second chance at validation) to the rejected signatures, but to accept the junior checkers’ initial conclusions about the valid signatures, without giving those signatures a second look.

    Whatever procedure applies to one group of signatures should apply to the other. That may prolong the process and keep the master checker busier than he would like, but since the status of R71 may come down to a few score or a few hundred names, I can’t see how a comprehensive master-checker review this can be skipped.

  3. Wait, now the new update shows 600 more than what’s listed as “checked” at the bottom table. Seriously, what’s going on? Does anyone there have the correct numbers to correct all the tables?

    There’s something up with these numbers, and it’s not kosher.

  4. David, How many “master checkers” are there per other checker? Is it 1:1? And if not, do they use the same process for checking every signature that the other checkers do?

    Thank you.

  5. David, How many “master checkers” are there per other checker? Is it 1:1? And if not, do they use the same process for checking every signature that the other checkers do?

  6. And my second part to the question, is that we’ve seen nearly 20% of the rejected signatures moved over to the accepted side. Have there been any that have gone the other way?

  7. Can someone please explain how the chart on the main page that is updated has a cumulative error rate in the top box (11.51 %) that is lower than EVERY single daily snapshot number in the lower chart? An average number shouldn’t be lower than ALLof its individual components……what am I missing here?

  8. Dear Blogreaders — I was off for the weekend, and i am away from my office this morning. After conferring with the Elections Division, i will address your questions/comments. I’m told that the main point is that the top section of the R-71 website is the one to monitor and that the daily snapshots quickly become out of date in subsequent day or days, primarily by overstating the rejections. There is zero intent to put out confusing or misleading or biased information — that was the exact opposition of trying to get information out on a daily basis. As I previously blogged, even though i’ve covered I&R for nearly 40 years, I assumed that a signature was either determined accepted or rejected, period. Typically the check is done without daily work papers being released, and we’d just find out after the random sample testing or the full count how it turned out. I didn’t know about the master-checkers and how missing signatures on the database are handled. In releasing the daily work papers, it appears we have confused and concerned people more than we have enlightened. for that i’m sorry. We’ll know soon enough whether the sponsors got enough valid signatures.

  9. Thank you for your email David. I like having the dailies released and appreciate the transparency. I still wonder about my second question. How could 20% of the rejected signatures get moved over to accepted with none moving the other way. Are the master checkers not checking the accepted ones as well. This is extremely disconcerting. And, frankly, something that seems could be challenged.

  10. Also – 11.51% apparently represents the best rejection rate ever. This seems a little surprising and a little unbelievable.

  11. David your efforts are very much appreciated and the daily updates are really enlightening and informative. While we are very much on an emotional roller coaster with the outcome of this process I should hope that people in WA would want their signature to be double check if it were tossed aside – I would.

    If the results (%) were leaning the other way all of this complaining and mistrust would be cheers. Folks, let the process play out. No one wants to screw you over. We are lucky to watch this process with twice daily updates and to have this communication with the Sec’y of States Office. The sense of entitlement to get the news/blog first sort of reminds me of armchair quarterbacks. They are doing to best they can with the information they have and are granting us a peephole into the daily process. Since these are not the official numbers, but a daily count, ease up and work on restoring your faith in the democratic process.

  12. David Johnson has an excellent point. An ‘initial checker’ (the more junior staff) could presumably make the same type and percentage of errors in rejecting a signature as in accepting a signature. But even if accepted signatures are wrongly accepted at a lower error rate than rejected signatures are wrongly rejected, it’s a human process — so there has to be some error. If the referendum qualifies by only several hundred (or a thousand) signatures out of 138,000 submitted signatures, then all accepted signatures — out of fairness — ought to be reviewed by a more senior reviewer. In essence, referendum proponents will have had 2 bites of the apple to get signatures ACCEPTED (the initial reviewer, then the senior reviewer) — but accepted signatures are considered conclusively accepted after only one initial review, which means referendum opponents will have had only 1 bite of the apple to get signatures REJECTED (only the initial reviewer). In other words, it’s like a recount of only rejected signatures– rather than a recount of all signatures (an analogy to what the Supreme Court found to be a denial of equal protection of law in Bush v Gore, and though the factual situation is a bit different, the concept is the same). That simply doesn’t seem fair when something is this close. Can the Secretary of State’s office please comment?

  13. To Dave Ammons. KnowThyNeighbor.org here. I am concerned about signatures that are being validated/invalidated based on authenticity of the sig itself. You say there is no provision for challenges by the public, however, I have a scenario for you. If the names of signers had been released during this process, KTN/WhoSigned.org would have been able to provide the public with a searchable database of all signers, showing non-matches against the voter reg list but also showing a pdf of the actual petition (ie the actual signature) see our Arkansas lists and click on the last name of a signer http://www.knowthyneighbor.org/arkansas/
    But more importantly to your signature checking, if the public had been allowed access via our work, voters would be able to see the pdf’s and know if their signatures were forged. This happens alot especially when the signatures were collected in churches–people will sign for family members and even friends and neighbors. Are you telling us that if I as a voter, saw that my name was forged and I contacted your office while you were still validating signatures that you/the Sec of State would not remove my name (a forgery) from the number count?

  14. Mr Ammons, were the Friday swing shift and Saturday numbers not reported because you were out of the office, or is there some other reason why the initial reporting schedule given us isn’t being followed? No problem if there’s a scheduling snag, but it would be good to know the story since our expectations have been raised to view updates twice per day.

  15. J Scooter:

    I was thinking the same thing. If it is the case that the SoS is giving one set of signatures a second bite at the apple and another set of signatures only 1 bite at the apple, then that raises a Bush v. Gore (i.e., an Equal Protection Clause) issue. Bush v. Gore dealt with the process of determining “voter intent” on cast ballots. But the principle is the same: you can’t subject citizens to apply different standards within the same process. Thus, it was unconstitutional to have a recount in some which FL citizens’ ballots were subjected to a rigorous standard while other citizens’ ballots enjoyed a liberal standard of review.

    So too it seems to me there would be a very strong case that it would be unconstitutional (not to mention unfair and distorting) to subject rejected signatures to a more rigorous review process than accepted signatures.

    We don’t yet have confirmation from Mr. Ammons that this is what is happening. But I am sure that Secretary Reed, a public servant of great integrity, will rectify this issue if it in fact exists. If that extends the process by a couple of weeks, so be it.

  16. Again I am having some concerns about the math. It now appears that the total listed as checked (35,296) is fewer than the total of the daily checks (35,866).

    There was a good explanation as to why Friday’s total was MORE than the sum of the days (some counts early on Friday), but I am having difficulty thinking of a scenario by which the total checked is LESS than the sum of the daily checks.

    And on Friday we were told that there were 1,111 more valid signatures than the daily valids, but today it is only 155 additionals corrected valid signatures. Is it possible that some of the supervisor corrections have been re-corrected back to rejection?

  17. Hi, Lurleen– we’re actually in the process of designing some new reporting and a better website. think launch will be tomorrow afternoon, reporting what we know firmly by that point. we’re recasting so that we report better refined information — that is, after firm decisions have been made by the checker and master checker and the accepted-rejected categories are much less subject to change. i’ll be blogging on this in a bit. I hope you’ll see more useful information. Releasing the daily worksheets ended up being more confusing than helpful, i’m afraid, since so many signatures were mid-process.

  18. AJ– the numbers have shifted around because that’s the way the signature-checking process works. you’ve been seeing data based on many signatures being mid-process in the checking. we’re at work now on a better way of laying it out — entering the signatures checked and the outcomes as we nail them down, batch by batch. ETA is 4-5-ish tomorrow afternoon. We’re still hoping/planning to be done checking by late next week.

    Neil–no intent whatsoever to do anything but check the signatures with integrity. please don’t presume to know the hearts and minds of 30 checkers, the professionals at the Elections Division, the Secretary of State, and the government. nothing is rigged or fixed here. i swear.

    David Johnson– i checked and my original understanding, which i previously posted, is correct, that our master checkers have had the policy for years of double-checking the rejects, but not the accepteds. The rule of thumb is that you can be qualified as a signer with one person, but rejecting you take two sets of eyes. Our attorney says the long-established case law is that we conduct the process to faciliate access to the initiative and referendum process guaranteed in the Constitution. This is the longstanding process used in checking signatures.

    AJ–numbers are shifting around, per above.

    John Sutherland–we have 5 master checkers and about 25 workers, so roughly 1:5. The master checkers are working from “volumes” or batches of 15 petition sheets that have been completed by a checker. They are senior people who sometimes find the voter properly registered under a slightly different name, address, etc. Per second question, the accepted ones are not master-checked.

    Mike–cumulative error rate accounted for master-checkers’ decisions and counties reporting back with electronic voter signatures. daily error rate and data were works in progress.

    KL– thanks for your comments.

    J Scotter–see above for our policy on double-checking the rejects.

    Tom Lang–Signature-checking is central to what our checkers are doing, Tom. per public records request, i presume you know that we have a longstanding policy that petitions are releasable public records. Only a federal court order is blocking us from release. We have at least three requests so far. A court order on the merits of the sponsors’ First Amendment case is Sept. 3. As you note, that’s scheduled for after the signature-check is complete.

  19. David Johnson– i assume you have now read our confirmation of our longstanding policy of double-checking only the rejecteds. this has never been challenged in court as a public policy at the state level, i’m told. case law sides with access to the ballot.

    Timothy–hang tough. we’re working on a new and improved reporting that doesn’t use the apples and oranges of signatures that are only mid-way through the vetting process.

  20. Thanks for the response David.

    I think we’d all agree that the person who knows the voter’s signature best is the voter. And I think we’d also have to agree that the person who knows with absolute certainty whether they signed the petition is, once again, the voter. What is difficult to understand is why the most qualified people to determine with absolute certainty whether signatures are valid, the voters, are the people who have been prevented from engaging in the process.

    Our plan has been to publish a searchable database where each name links to the petition sheet where the name appears so that the voters can verify their own signatures. Prior to the Temporary Restraining Order (TRO), we expected to have about two weeks before certification for Washington State voters to check and see if their name was on the petition and whether their signature had been forged or report any other irregularities. With the TRO preventing the Secretary of State from releasing the signatures to any requesters until after September 3rd at the earliest, whether the referendum will be on the ballot will be decided before voters have had any opportunity to engage.

    It’s unfair that the voters of our state have been locked out of the verification process when they are the best people to judge their own signatures.

    Is there a process by which the certification period can be extended to allow time after the TRO is removed for us to publish and allow voter self-reporting of signature forgery and other irregularities?

  21. Brian — I’m unaware that copies of petitions have ever been used, before, during or after the Elections Division check, to challenge a signature that has been officially submitted to the Secretary of State for verification. I know that the practice is to receive any written statements about ones own signature (such as wanting to withdraw it) as part of the public record, but as previously reported, a WAC says the signature cannot be subtracted from the total. If a court-ordered re-check occurred, I can only speculate what the judge would allow to be considered.

  22. Dear Mr. Ammons:

    Thank you very much for your response to my concern! While I respect your office’s intentions, and I appreciate that it has been regular practice to double-check only those signatures that were initially rejected, I think the SoS’s position is untenable.

    You say that the SoS “conduct[s] the process to facilitate access to the initiative and referendum process guaranteed in the Constitution.” But the same WA Constitution that allows for referenda *also* provides hurdles to the referendum process. Both the right to referendum *AND* the hurdles to qualifying a referendum are constitutional concerns, and both must be respected. (Otherwise, why bother checking the signatures at all? Why not just “facilitate access” to the referendum process by counting the signatures on the petitions and then certify based on the raw count?)

    Skewing the review in order to minimize the constitutionally mandated hurdles (i.e., by making it easier for signatures to be accepted by junior staff, but harder for signatures to be rejected by that staff), is no more constitutionally valid than would be imposing greater and unequal scrutiny on the accepted names.

    I sincerely hope that this turns out to be an academic question (either because R71 fails or passes muster by a margin that exceeds the number of signatures resurrected by the master checkers). But if the end result is close, I think you are certain to have a lawsuit, and that lawsuit will succeed.

    I am not sure what precedent your lawyers were referring to, but I highly doubt that it is caselaw that addresses this unequal “master checker” system that you have described. And, while I could be wrong, I doubt that the precedent your attorneys have in mind applied Bush v. Gore, which cut new federal constitutional ground in December 2000. I hope your office reconsiders its current practice and either master checks all the signatures or none.

  23. All,
    After September 3rd, if the public’s right to access of public information prevails and the TRO is thrown out, you will give WhoSigned.org and KnowThyNeighbor.rog about 2 weeks and you will be able to see all the signatures, including the non-verified ones and the actual scanned petitions sheets (corresponding to the names) for public verification. The way it should be. It may not help the overall cause, but it certainly will give a very clear cut, truly transparent look into the process. We (the public) will have many of its questions answered about accuracy in the process as well as the exposure of fraud, misinformation, trickery etc that may or may not have occurred with paid signature gatherers and “activists” collecting the signatures. What I know to be an important way to help maintain the integrity of the initiative petition process.

  24. David Johnson– I don’t know if the checker-master checker reject-accept question has been squarely litigated, but I suspect not. I was talking about the general principle. We do, obviously, require the person to be found on the Washington state voter database and for the signature to match. Thank you for your thoughts on this point, including the reference to Bush v. Gore.

  25. Mr. Ammons has stated two things that aren’t necessarily consistent: (1) “the long-established case law is that we conduct the process to faciltiate access to the initiative and referendum process guaranteed in the Constitution” and also that “this has never been challenged in court as a public policy at the state level ….” Basically, what this means is that although “long- established case law” GENERALLY access to the ballot, the precise issue of double-checking only rejected signatures and not double-checking accepted signatures has never been challenged. But there may be unique circumstances here that support such a challenge on statutory or constitutional grounds.

    First, and perhaps more important, past signature verifications were apparently conducted in situations where the signatures themselves were always publicly available — and that meant, as Brian Murphy has suggested, that every voter could see for him/herself whether his/her name was submitted without authorization by someone who improperly knew his/her signature. Yet in this situation, given the TRO, not a single voter has any idea if his/her name was submitted without authorization (if nothing else, this is a reason for the State and other interested parties to ask the federal judge to dissolve the TRO — the so-called privacy / retaliation concerns of some voters need to be weighed against the right of all voters to assure that their names were not improperly submitted, something that is a particular concern now that we know that the State double-checks ONLY rejected signatures and does not double-check any accepted signatures). Brian Murphy, don’t you think that once the federal judge knows how the signature checking process is conduced that he might be inclined to balance the interests somewhat differently and dissolve the TRO? Or maybe the 9th Circuit will dissolve the TRO (usually TROs are not appealable, but in a situation like this — where the non-disclosure affects the signature-counting process, the grant of the TRO may well be appealable because it will be difficult if not impossible to undo the harm created by the acceptance of signatures that should have been rejected).

    Second, because the signatures are kept from public view during the qualifications process, it is particularly unconscionable that the State is not double-checking accepted signatures where success or failure of qualifying for the ballot could be determined by 1% or less of the signatures. It is unclear from Mr. Ammons’s comments whether the long-standing precedent of not double-checking accepted signatures ever took place in a situation where the signature count is bound to be super-close. At a minimum, the State could randomly double-check 20,000 accepted signatures and see what it finds — if no errors at all are demonstrated, then perhaps double-checking of all accepted signatures would be a waste of time. But if even 1% error is found in the accepted signatures, that itself demonstrates the need to double-check ALL acceptable signatures.

    Third, as David Johnson has noted, there are two sides to the referendum process — the right to get on the ballot, and the right not to have a validly enacted law improperly submitted to a referendum. Once a law is enacted, the burden should be on those attempting to overturn it — and that is inconsistent with giving the law’s opponents two bites at the apple with the signatures, versus denying the law’s proponents the same opportunity to assure that accepted signatures were properly accepted. It’s actually a somewhat different situation from the typical voting situation where — AFTER a candidate or issue has properly qualified for the ballot, the eligibility for voting itself is interpreted charitably in favor of the voter’s right to vote. Here there is a strong offsetting reason, i.e., the enactment of the law, to be considered when determining how “charitable” to be in accepting signatures.

    In short, unless the State conducts — at a minimum — a random double-checking of the accepted signatures to determine if there are ANY errors in the accepted signatures — there seems to be a strong constitutional flaw in the qualifications process that is being used, whether that flaw exists due to Bush v Gore or some other court decisions.

Comments are closed.

Comments are closed.