December 01, 2004
What a hand recount means

From Secretary of State Sam Reed's press conference last Wednesday (Nov. 24) regarding the prospect of a hand recount:

Each time we count we're going to be moving ballots around and particularly the punch cards, in my experience, being a punch card county auditor, is each time they're handled some of these chads are going to fall out from moving around having been loosened and such and so it is not desirable to just keep recounting and recounting and rehandling these ballots. But I certainly expect that to happen
And then there is the error-prone lunacy of hand counting the touchscreen votes for which no original paper exists:
In Snohomish and Yakima counties they have electronic touchscreen machines. In both of those counties what they will have to do in the case of a manual recount is print out the result. Snohomish actually had experience with this. In 2003 they recounted countywide for a County Assessor race. So they had to print like you do out of a PC, print out the results out of their touchscreen machine and then they hand recounted those.

Question from audience: How many votes are there total in that machine?
Reed: I think they can do it for each vote as I recall ... each ballot can be printed out, right.

Yes, it really means that they will print out brand new paper for every touchscreen vote and count those by hand. Raise your hand if you think this is a sensible idea.

On the other hand, in spite of the Democrats' hysterical claims that "thousands of ballots across the state haven't been counted", there doesn't seem to be any serious reason to believe that there were any systemic errors that a hand recount could conceivably correct:

Questioner: Can you put a number on the number of ballots you might consider suspect as the result of the way they were handled and is it more than 42?

Reed: From what I have learned about the process that's been used in the counties to correct what has gone on is I have confidence in the numbers that we've received. That they have corrected the problems. I do not know of any ballots that I would view as being incorrectly counted at this point.

Posted by Stefan Sharkansky at December 01, 2004 11:17 PM | Email This
Comments
1. I dunno. The Snohomish and Yakima results can be used to see just how accurate a hand recount is. We know exactly how many votes there are going to be printed, and exactly who they are for, so any deviation in the hand total is because of the process used to count them.

Here's hoping Rossi gains a few votes in both, a little punishment for the Dems who can't complain (but will whine) with those results.

Posted by: Raoul Ortega on December 2, 2004 08:45 AM
2. I'd really like to ping these stories for trackbacks. Have you guys considered implementing them?

Posted by: Jeff on December 2, 2004 09:59 AM
3. Stefan, while you're on the subject I'd like to plug my own bit of analysis based on studies by the CalTech/MIT Voter Technology Project. I explain the difference between a voting machine's certification rate and the technology's resulting "residual voting rate," and compare the relative performance of the various technologies.

My final conclusion is that when you consider the tabulation errors introduced by even the most reliable technology (optical scan), this election is too far within the margin of error to accurately determine the winner. Thus, a "win" in any of the counts -- first, second or third -- regardless of the counting technology, is statistically meaningless.

Read it, then rant at me: Heads or tails… why we’ll never know who really won the governor’s race

I promise, it's a wonkish piece, not a partisan flame.

Posted by: David Goldstein on December 2, 2004 10:37 AM
4. For David Goldstein: I am in agreement with your general conclusion about the impossibility of identifying an actual winner, as opposed to a declared winner, regardless of the methodology employed; but I am not sure I follow the distinction you are drawing between "certification rates" and "residual voting rates." Where do I find the study to which you refer?

Posted by: Tom Rekdal on December 2, 2004 11:23 AM
5. Goldy, this analysis is flawed.

Obviously there is the margin of error of a particular technology, and there's the residual margin of error that you point out.

But, none of that really asks the important questions. For starters, what is the point of holding an election? Are we not trying to determine a winner? Now you can dismiss this as statistically insignificant using the residual blah x 3. But the reality is that a winner needs to be determined given the process as defined by law and not by a coin toss. Thus, the question that should be asked is "how can we lower the error rate of voting within the technologies that were used to hold this election?"

The best way to do this is to throw out the ballots that are not readable by machine. Once that happens, we arrive back at the 1 in 1,000,000 error rate of the machines, we get rid of the residual error of undervotes and overvotes which are simply improperly cast ballots, and we get rid of the bias of human divination of intent.

In short, if truly determining the winner with the least margin of error is the goal, then the best way to do it is to elinate the largest sources of error.

But let's be frank, you nor Gregoire really want to eliminate illegitimate ballots because introducing a large margin of error is the only way to create a circumstance that "might" produce a Gregoire win. This is shameless partisan machination.

The most accurate results we have thus far are those produced by machine, and before divination bias. That shows a Rossi win by 261 votes.

Admit defeat and concede.

Posted by: Jeff B on December 2, 2004 11:29 AM
6. Tom, there are links to the CalTech/MIT studies on my website: http://www.horsesass.org/wp-trackback.php/276

The certification rate is the error rate for counting machines using flawlessly marked ballots. This error rate is necessarily very small... less than 0.0001 percent. The residual voting rate represents the difference between the number of ballots cast and the number of votes for a particular office... those over-votes, under-votes and otherwise "spoiled" ballots we keep hearing about. Some of these residual votes are intentional. But since different voting technologies produce dramatically different residual voting rates, some are clearly not.

Posted by: David Goldstein on December 2, 2004 02:36 PM
Post a comment
Name:


Email Address:


URL:


Comments:


Remember info?