Box Turtle Bulletin

Box Turtle BulletinNews, analysis and fact-checking of anti-gay rhetoric
“Now you must raise your children up in a world where that union of man and box turtle is on the same legal footing as man and wife…”
This article can be found at:
Latest Posts

August 24th Washington Referendum 71 Update

Timothy Kincaid

August 24th, 2009

A total of 103,898 signatures have been reviewed, or 75.5%
A total of 91,716 have been accepted, or 76.1% of those required.

The current fail rate for non-deferred signatures is 11.69%. However, this is accurate for only the first 220 volumes; the rest is subject to revision for registrations not processed in the data base before it was downloaded. I project that there may be perhaps another 700 signatures that are considered rejected at the moment which will ultimately be accepted.

At this point we are working under assumptions and guesses, but the “real” fail-rate is probably around 11.0% of all inspected signatures. Unless there is a high fail-rates for the last quarter of signatures – perhaps 16.8%, then this referendum may qualify. Keep in mind that we do not know whether the petitions are homogeneous or if the last quarter is cleaner or less clean than the first three quarters.



August 24th, 2009 | LINK

The REAL rejection rate is much higher than 11.72%. Election officials have RIGGED the results to accept more signatures than are actually acceptable by looking at rejected signatures 4 or 5 times and by accepting signatures from people who weren’t even registered to vote.

BTW, election officials have not said what they have done will accepted signatures they found to be incorrectly placed from that 222 sample.

August 24th, 2009 | LINK


At this point, there are so many variables and not enough information as to what is being included in these daily numbers. So I am not so sure your dire numbers are right. But suffice it to say that if the cumulative error rate does not move up by .19%/9,000 signatures checked, then R71 will qualify. If we are not in the 11.9% range tomorrow, assuming 9K checked, that will be a clear signal of where it is going.


There is no doubt in my mind that there are hundreds, and likely several thousand signatures that are wrongfully sitting in the accepted pile. It is simply amazing to me that the SoS hired temps to do this work, with the unsurprising consequence of a high rate of mistakes (in both the reject pile and in the sample of 226 accepted sigs).

With a very deliberate policy of checking and correcting the temp errors only in the reject pile, the outcome is being affected. Any 12-year old can figure out that even a 1% temp mistake rate in the accepteds can change the result. But to start master checking all of the accepted sigs now will constitute an admission that the practice of hiring temps was an entirely wasteful failure. They don’t want to admit that, so the result is what we have now.

I don’t know if you are a WA resident, but if you are, I think the best thing you could do (rather than vent on the internets) is contact WA Families Standing Together and insist that they pursue this issue in court. They really ought to be preparing an application for expedited injunctive relief now, to be ready for filing when and if it becomes apparent that R71 will qualify.

I can’t imagine that they would be so incompetent as to let this fall through the cracks. But then, I wouldn’t have thought that our Prop 8 leaders would go on 4 week vacations a few weeks before Election Day. So if you are a WA resident and/or a supporter, give them a call.

August 24th, 2009 | LINK

Dan, is definitely on this. You can learn a lot by looking at the requests for public records Anne Levinson has filed at But remember, just because WAFST is on top of things and presents a cogent case doesn’t mean a judge will rule in our favor. So you might also encourage your friends in Washington to get busy with WAFST right now, not waste valuable time hoping a magic judge fairy will make it all go away!

Timothy Kincaid
August 24th, 2009 | LINK

Dan and Cole,

Unless I’m mistaken, the SoS’ position is that the spot check of 226 final decisions on both side resulted in a net of one difference.

Yes I think it merits further review – especially if the count is very close – but I’m not imagining sneaky tricks.

I guess the silver lining is that if they don’t get enough, they certainly can’t claim that their rejected signatures didn’t get enough review.

August 24th, 2009 | LINK


You are right. Nothing is guaranteed in court. But in this case, based on these facts, I cannot conceive of how a judge, or failing that the appellate courts of WA, could allow R71 to be certified based on such a systemically unequal review system, based on a solitary review by temps, with a known error rate by those temps that exceeds the margin by which R71 qualified.


I believe you are mistaken. It is true that the SoS ran a separate sampling exercise based on the R71 proponents’ complaints about a few specific master checkers. The result of that sample review (of 225 sigs) when paired with the accepted-sig master check sample (of 226 sigs) resulted in a small net change, because the exercises largely canceled each other out.

But that does not mean that the two issues have the same potential impact on the outcome of the review. Our issue – the failure to master check the “accepted” sigs — is systemic and ultimately will implicate the status of 120,000+ accepted signatures. Their issue only concerned the diligence of a few specific master checkers. Presumably, those master checkers have been more diligent after having been called out by name. But even if they haven’t, whatever errors they may have caused and are causing would impact only a subset of the rejected signatures — a vastly smaller pool than the one we are concerned with.

So the two sampling exercises give a false appearance of parity because the samples were equal. But there really is no comparison. I would gladly trade off a re-re-re-re-check of the sigs that worry the R71 proponents in exchange for one full master-check of all accepted sigs.

IMO, if R71 is certified by 1000 sigs or less, we can say that the certification is almost certainly fraudulent. If it qualifies by b/t 1000-2500 sigs, then the result is open to serious doubt.

August 25th, 2009 | LINK

Tim you are naive. Those rejected signatures that were sampled were already looked at FOUR times before. So why would there be a shift in votes from those already rejected signatures? Every time they look at rejected signatures again a massive amount is moved over to the accepted pile. If the process was honest and just the accepted signatures would have been the only ones to move in the sample since they only got one look by temporary workers prior.

Timothy Kincaid
August 25th, 2009 | LINK


Our issue – the failure to master check the “accepted” sigs — is systemic and ultimately will implicate the status of 120,000+ accepted signatures. Their issue only concerned the diligence of a few specific master checkers.

Yes, good point. This does illustrate a flaw in the system.

I don’t think this is an indication of any anti-gay bias (this is the procedure with all close initiatives) but it does show a built in system bias that favors supporters of a referendum.

August 25th, 2009 | LINK

I agree, it is not anti-gay bias and it is not a sinister conspiracy. It is pretty obvious what happened:

The SoS gets hit with these referenda petitions and, because the office can’t grind to a halt to do signature checks, it developed a history of using temp workers. These temp workers – with no prior experience and low pay – unsurprisingly make mistakes.

Typically, this doesn’t matter since referenda proponents submit many more signatures than are necessary to qualify, or contrarily submit not nearly enough. So the outcome is clear regardless of how many false “accepteds” get by the temps. Real employees double check the much smaller category of rejects, and that becomes the regular practice.

Along comes R71 and it becomes one of only 4 instances in WA history that demands a signature-by-signature check. Very labor intensive, so the SoS has to hire a lot of temps. They stick to the usual practice of not worrying about the accepteds. But now the temp error is clearly material to the outcome. No one focuses on this until a week into the count.

But now it would be an embarrassment to admit that every sig had to be master checked. That would essentially admit that the use of temps was a wasteful failure. So they stick with their “decades old tradition” and continue with the count. Last week, they do the sample and discover a huge temp mistake rate. I have no doubt that the SoS was horrified, but since at that point we were 2/3 done with the count and since R71 may still fail even under these skewed conditions, they stick to the present practice. And here we are.

If the R71 opponents can’t get their sh*t together and sue over this on an exigent basis, they will be exposed as bigger incompetents than Michael Brown at FEMA.

August 25th, 2009 | LINK

J Scooter made a great post on the sos blog, here it is

Leave A Comment

All comments reflect the opinions of commenters only. They are not necessarily those of anyone associated with Box Turtle Bulletin. Comments are subject to our Comments Policy.

(Required, never shared)

PLEASE NOTE: All comments are subject to our Comments Policy.