This afternoon, the Secretary of State’s office released R-71 data in a brand new format. Apparently, the data now reflect the actual numbers of duplicates, rejected signatures, and accepted signatures.
There are some noticable differences over the previous data releases. As David Ammons explains it:
The error rate is lower than the daily and cumulative numbers that had been previously reported, because the earlier numbers included many signatures that still were being reviewed by master checkers. A prime example is that hundreds of signatures were not initially found on voter rolls by the checker, but a later check by the veteran master checkers did make a match.
He also points out:
State Elections Director Nick Handy said it remains “too close to call” whether R-71 will make the ballot, and cautioned against making assumptions based on the current error rate.
Handy is incorrect in one respect. Given a proper statistical estimate of the duplicate error rate in the total sample, and a proper projection of the other invalid signatures, we can estimate a total number of valid signatures and offer some statistical certainty about the number. (Of course, this assumes we are given the correct numbers in the first place….)
The statistical certainty only accounts for the fact that we have only a sample of the total petition evaluated so far. It cannot account for non-sampling error, biases, correlations among batches of pages, etc. Of course such error may be ignorable. I’ll get back to that issue in a later post.
The total number of signatures that have been completed is 33,214, which is just under a quarter of the total petition. There have been 3,450 invalid signatures found, for an uncorrected rejection rate of 10.39%. This rate doesn’t mean much because it doesn’t include the rate of duplicate signatures in the total petition.
The invalid signatures include 3,117 that were not found in the voting rolls, 130 duplicates, and 203 that did not match the signature on file. There are also 12 pending signatures in which a better signature card is needed. (Oddly enough, the data table includes the 12 pending signatures in the rejected totals; I suspect this is an error, albiet a minor one).
The 130 duplicated signatures from a sample of 33,314 suggests a duplication rate on the entire petition of about 1.62%.
Using the V2 V estimator, the number of valid signatures is expected to be 121,103, thus squeaking by with 526 signature over the 120,577 needed to qualify for the ballot. (The sampling error is many times smaller than the 526 margin.) The expected total rejection rate is 12.05%.
The bottom line: Unless new errors are found in the processing or counting, or some large, systematic increase in the error rate is seen for the remaining 76% of the signatures, we should expect to see R-71 on the ballot this fall.
Update: I just noticed I used the V estimator, not the V2 estimator. The V estimator is slightly biased toward too few valid signatures, so the qualitative results are the same.
So has everyone signed up for a phone bank shift yet? Since the ballot language is counter-intuitive, we need to imprint on voters’ brains that to preserve the new domestic partnership law, they must vote APPROVED on Referendum 71. Sign up with Washington Families Standing Together (www.wafst.org) to help get the word out.
Referendum 71 voters will be asked to approve or reject the domestic partnership law.
REFERENDUM 71 Ballot Title
Statement of Subject: The legislature passed Engrossed Second Substitute Senate Bill 5688 concerning rights and responsibilities of state-registered domestic partners [and voters have filed a sufficient referendum petition on this bill].
Concise Description: This bill would expand the rights, responsibilities, and obligations accorded state-registered same-sex and senior domestic partners to be equivalent to those of married spouses, except that a domestic partnership is not a marriage.
Should this bill be:
Approved ___
Rejected ___
[Deleted. See HA Comment Policy.]
They should make it simple:
Are you in favor of equality (yes)or do you prefer to side with the homophobes (no)?
(sigh) The bad guys win another round, and the world gets a little worse. Now lets stop the blacks, Jews and Irish and we’ll all sleep a little better I suppose.
[Deleted — see HA Comment Policy]
I think Mrs. Troll got confused and meant to post this
http://pamshouseblend.com/diar.....ing-points
Darryl – I realize that getting enough signatures by one has the same effect as getting an extra 100k, but I think it’s too early to say R-71 is likely headed to the ballot. “Too close to call” still seems accurate.
You estimate an end total within half a percent of the SoS-mandated total signatures needed. I’m guessing the borderline is well within the model’s standard deviation.
I know you have other stuff to do, but have done any elasticity tests? I’d be curious how much the estimated total changes if there had been just a few (say five or ten) more signatures rejected so far, especially if duplicates. Thanks.
Darryl wants to be right.
I mean when you’ve got a verification and reporting process that’s been as odd as this how certain can you really be of a post titled “R-71 Headed for Ballot”. Plus we still have 75% of the ballots to verify. It would be generous to even say “Appears to be Headed for the Ballot”! I don’t think either side can be certain of anything except uncertainty. It won’t surprise me if this is wrong and it won’t surprise me if it’s right. Get the picture?
Wingies hate. Then, they get their crap on ballots. R-71 is simply hate, bias, and bullshit. And, our state has to expend tax dollars to put this nonsense on the ballot. Can anyone tell me who wins when this crap stains our ballots?
mr. smitty,
“Too close to call” still seems accurate.
Under the assumption that the signatures checked so far are random with respect to all signatures, then it is not “too close to call.” If the assumptions are correct (and the data are given as defined by the SoS office) it is statistically highly improbable that the “win” is by chance alone.
“You estimate an end total within half a percent of the SoS-mandated total signatures needed. I’m guessing the borderline is well within the model’s standard deviation.”
Nope. Treating the error as “doubly-stochastic” (once for the estimate of the probabilities of each type of error, and a second time for applying those probabilities to the distribution of final estimates, the 95% confidence intervals is much, much smaller than the margin.
Tim @ 8,
Darryl wants to be right.
About what?
Tim @ 9,
‘I mean when you’ve got a verification and reporting process that’s been as odd as this how certain can you really be of a post titled “R-71 Headed for Ballot”.’
The only thing “odd” about the process is that the SoS office was releasing counts that didn’t match the definitions for those counts.
They have explained the misreporting, and have been very clear about what they are reporting now. While it wouldn’t defy any fundamental laws of physics for them to be wrong a second time, I’m pretty confidant they have the reporting correct now.
BTW: My assessment of “certainty” is stated in the last sentence of the post. In short: It is certain to make the ballot when only examining standard sampling error; It is not inevitable that it will be on the ballot because other types of (non-sampling) error are always possible.
“Plus we still have 75% of the ballots to verify.”
Statistically, a sample of 33,000 from a population of 138,000 makes for a very small amount of remaining uncertainty from sampling error. Don’t pin your hopes on that…you will be disappointed.
Thanks Darryl. That surprises me that the std deviation is less than 500 signatures, but it is what it is. Ideally the dynamics change for the better (worse?) in the coming days.
Thanks for showing us what a complete self righteous moron you are for not even being able to calculate numbers correctly.
Little Adolf @15,
Thanks for showing us what a complete self righteous moron you are for not even being able to follow this thread. Are you a product of home schooling?
Darryl
We need a referendum to mandate us eof the 1% probability level.
Then, any election that is not 99% certain to represent the true vote would need to be rerun.
Imagine NO MORE RECOUNTS!
I am assuming that you are taking into account that the duplicate rate will increase as more signatures are checked?
Gres @ 18
Yes
Welllll, our fuzzy math brainiacs apparently forgot to factor reality into their calculus. All that bandwidth wasted, all those fancy tri-color graphs and charts expended, in a finally futile effort to cover up one little important point: the reject signatures were not necessarily rejects.
Goldy and Darryl were blowing us away (this is why math is so important and this is why we are so important) with pre-provisional numbers from temp checkers. Those numbers for rejected signatures were then passed to the pros for validation. Some still failed to qualify, but many others were validated as valid.
We’re at ~10.4% invalid, with ~100,000 to go. Maybe your wetdreams will come true and R-71 will finally fail … maybe an ascending curve of duplicate entries will push your wishes over the top after all. But your stupid silly fancy fuzzy math and algorithms and charts and graphs will still be stupid and silly.
HA “Policy” @2 and @5:
Nevermind squashed dissent. Don’t like squash anyway.
But I mind very much the professional trolling part. I mean, WTF? Been trolling here for years at minimum wage or less, and now you tell me I’m a professional? You owe me big time at union scale, retroactive to Day 1.
Who the ‘ell do you think you are? Wal-Mart?
10 Apparently thats not the only place “crap stains”. HAHAHAHAHAHAHAHAHAHAH