Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

HCPB study. ER brilliant analysts, please analyze.

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Topic Forums » Election Reform Donate to DU
 
Melissa G Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 11:30 AM
Original message
HCPB study. ER brilliant analysts, please analyze.
Edited on Tue Mar-11-08 11:32 AM by Melissa G
Got this from another list I subscribe to. Have yet to read. Seems implausible to me, but as I have said I have not read the study yet.
Best,
Melissa

http://www.krdo.com/Global/story.asp?S=7960710
Study Shows Counting Ballots By Hand Inaccurate
Updated: March 4, 2008 10:55 PM


By Political Reporter Marshall Zelingerm.zelinger@krdo.com
COLORADO SPRINGS - An El Paso County test shows that counting ballots by hand is only accurate 25-percent of the time. In January, El Paso County Clerk and Recorder Bob Balink tested three types of hand counting on actual ballots from the 2006 election.

One group of four election judges sorted their ballots into two different piles, one for Candidate or Issue "A," the other for Candidate or Issue "B". The group then counted up the ballots and recorded the vote total.

Another group of eight election judges worked in pairs. One person would read off a ballot to the other person who would record the vote on a tally sheet. All ballots were counted and then recorded.

The last group was one individual counting all ballots himself.
"The three different methods of hand counting was off from the machine count 75% of the time," says El Paso County Clerk and Recorder Bob Balink. "About one out of every four races that were counted in the hand count study matched the machine, and three out of every four of the races that were counted didn't match the machine."

Click Here To Read Complete Hand Count Studyhttp://krdo.images.worldnow.com/images/INCOMING/HandCount.pdf _______________________________________________
Printer Friendly | Permalink |  | Top
Warpy Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 11:36 AM
Response to Original message
1. Uh, maybe it's the MACHINE
I don't think there's a machine out there that doesn't miscount. Ballots get stuck together, get stuck in the works and counted twice, and get miscounted on purpose.

Counting in pairs is the best way to reduce bias.
Printer Friendly | Permalink |  | Top
 
Fovea Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 11:40 AM
Response to Reply #1
2. Yes.
interesting how the machine is presumed accurate.

Printer Friendly | Permalink |  | Top
 
Melissa G Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 11:47 AM
Response to Reply #1
3. This is a hand counted paper ballot study
There should not be a machine. Have you read the study already?
Printer Friendly | Permalink |  | Top
 
MH1 Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 11:49 AM
Response to Reply #1
4. You may want to read the study document.
(pdf link at the bottom of the op)

* optical scan machines were used
* the ballots were scanned on both machines, and the counts from the machines matched each other
* hand counts of the same ballots, performed by different teams, not only didn't match the machines, the hand count of ballot set A by one team, didn't match the hand count of the same ballot set A by the other team.

I don't want black box voting but I have no problem with using a machine when the process can be audited.
Printer Friendly | Permalink |  | Top
 
Warpy Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 12:00 PM
Response to Reply #4
5. Oh, same here, and I think a random spot check should be
mandatory before the results can be certified. Otherwise, I'm perfectly happy with the Optiscan paper ballots.

However, machines can miscount for the reasons I gave. The miscounts only become an issue in very close elections.
Printer Friendly | Permalink |  | Top
 
sonias Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 04:35 PM
Response to Reply #4
10. I second that motion
I've said this all along HCBP are no panacea for perfection. If anything you're adding more human error in the mix.

I've saved the study to take to our election reform coalition group too

From the participants in the process section at the very end of the study

1. Do you feel a hand count of ballots is an effective way to produce election results? 3 Yes 11 No Explain.
1 - No - This is a costly, time consuming process that is out-dated. Election results would be delayed.


Sonia
Printer Friendly | Permalink |  | Top
 
Peace Patriot Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 12:06 PM
Response to Original message
6. The MACHINE count is the STANDARD?!
:rofl:

Jeez, these people. We need a nationwide purge of county election officials. They are so frigging corrupt.
Printer Friendly | Permalink |  | Top
 
Wilms Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 12:07 PM
Response to Original message
7. Correct outcomes matter. Accuracy is secondary.
I'll bet a lot of handcounts are "inaccurate". If it's inaccurate by an amount much less than the election's margin, it matters little.

If I wanted to steal an election, I'd have an easier time corrupting an OpScan count than talking a bunch of hand counters to cheat.

I still think random, statistically significant, audits should be used with ANY election methodology, including hand counts. But the hand wringing notion of inaccurate hand counts is an alarmist view. If they're showing inaccuracies that effect outcomes...that's a different story. But most, I suspect, fall well below that threshold.

Printer Friendly | Permalink |  | Top
 
diva77 Donating Member (999 posts) Send PM | Profile | Ignore Tue Mar-11-08 03:45 PM
Response to Original message
8. not sure what to make of this study...just a couple of thoughts & then some background on "Bob"
Not enough info. is given regarding the people doing the counting and the circumstances under which they were counting etc. -- i.e. how many hours did each counter work for? How often did they take breaks? My experience is that no one should count more than 45 min. without a 15 min break, and then they should not count for more than 4 hours in a row in total.

Tthe fact that 2 machines were used for scanning the ballots does not apply to all machines in all elections. Software & firmware can have undetectable malicious code, blah,, blah blah.

Bob, however, has an interesting history...

Bob is gung-ho on the "voter fraud" bandwagon -- a quick google turned this up:

http://adm2.elpasoco.com/countypress/default.asp?ID=1546&itla=CAR&selectdept=CAR&selecttime=current&offset=10

Election Law Violations Being Prosecuted
District Attorney’s Office Diligent

July 28, 2005 – Colorado Springs, CO - El Paso County Clerk and Recorder’s office reports that Election law violation cases continue to be prosecuted due to a cooperative effort with the El Paso County District Attorney’s office.

Recently, one individual who was charged with voting twice in the November 2005 Presidential Election pled guilty to a misdemeanor, while another charge of declaring a false residency (a felony) was dropped. The sentence included a $5,000 fine and a one year jail sentence (both of which were suspended), along with a requirement of 48 hours of community service and 12 months probation and $198.50 in court-related costs.

According to Robert C. “Bob” Balink, El Paso County Clerk & Recorder, “It is important to note that only one of the two ballots cast by this young women was counted…the vote here in El Paso County.”

Two others cases, both involving felony forgery charges arising from fraudulent Voter Registration Drive activities, have been working through the legal process.

Today, one individual who worked for the “New Voters Project” and was charged with multiple counts of election fraud pled guilty to one charge of forgery (a Class V felony), and was sentenced to 2 years probation, 200 hours of community service, and was ordered to pay $322.50 in court-related costs.

“Election fraud undermines our democracy, and we take the matter very seriously. The El Paso County Clerk & Recorder Bob Balink is extremely diligent in reporting these offenses and providing supporting documentation, which facilitates prosecution,” said District Attorney John Newsome.

Another case involving 19 counts of forgery by an individual associated with “ACORN” is pending further court appearances. ACORN is an acronym for the “Association of Community Organizations for Reform Now.”

snip

And then this letter from the NAACP

http://coloradoconfidential.com/showDiary.do;jsessionid=76EF666297C182459AD28766EDE575E0?diaryId=3449

Proof of Citizenship Would Threaten Voting Rights

Voters should be alarmed by El Paso County Clerk and Recorder Bob Balink's proposal that people registering to vote provide proof of citizenship. Although showing proof of citizenship might sound reasonable at first glance, such policies have only one effect: suppressing the constitutional voting rights of American citizens, especially the elderly and minorities and those with low incomes. Many people simply don't have their birth certificate, which can be very difficult to obtain for people with limited resources.

Recent studies, such as those conducted by the Brennan Center for Justice at the New York University School of Law, have explored the questions of who has proof of citizenship and whether voter fraud by noncitizens is a problem that threatens election integrity in the United States.

The studies found that 7 percent - or more than 13 million Americans - do not have ready access to the documents that can definitively prove citizenship: a birth certificate, naturalization papers or a U.S. passport. Not surprisingly, the studies also found that those who most often lack proof of citizenship are poor, elderly or from racial and ethnic minority groups.

Furthermore, studies show that requiring proof of citizenship appears to do nothing to ensure election integrity while negatively impacting civic participation. In Arizona, for instance, the only state that currently requires proof of citizenship, 35 percent of new registrants in the most populous county were rejected in 2005, and 17 percent were rejected between January and September 2006. Most are presumed to be legal citizens who did not have the required documents.

We write on behalf of a coalition of local civic-minded organizations that believe the rights of so many legal citizens should not be sacrificed when there's not one single documented case of a non-citizen attempting to register and vote in the state of Colorado - much less El Paso County.

The right to vote is a cornerstone of democracy. Our elected county clerk and recorder should be in the business of defending that right, not creating barriers for legal voters. The current system of verifying voter citizenship by way of sworn affidavit is working. In the absence of any evidence to the contrary, we ask Balink to shelve this discriminatory proposal.

Rosemary Harris
President, Colorado Springs Branch NAACP

Barb Ferrill Van Hoy
Executive Director, Citizens Project
Printer Friendly | Permalink |  | Top
 
diva77 Donating Member (999 posts) Send PM | Profile | Ignore Tue Mar-11-08 04:09 PM
Response to Reply #8
9. forgot to showcase the "vendor/ operative" talking pt.
in the study, the following is stated:

"Going to a hand count of ballots requires a complete paradigm shift. Are the media and the public willing
to wait 3-5 days for unofficial results and at what cost? Many indicate that they don’t care how quickly
they get results, they just want the results to be accurate, but the reality is that people want results on
election night; even though they know the results are unofficial.


(boldness added for emphasis)

how many times have we heard this from the pro DRE rovs????


Printer Friendly | Permalink |  | Top
 
Wilms Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 11:36 PM
Response to Reply #9
12. Forget the pro DRE ROVs. Our very own autorank is among theose advocating that.
Printer Friendly | Permalink |  | Top
 
Stevepol Donating Member (1000+ posts) Send PM | Profile | Ignore Tue Mar-11-08 10:23 PM
Response to Original message
11. Stanford did a study comparing the various ways of counting votes
some years ago, tho I don't have the link now, which found that HCPB were the least likely to be wrong, had the smallest percentage of error.

But all the methods used had some percentage of error.

But nobody can convince me that HCPBs would have 18,000 undervotes in a voting division the size of Sarasota FL or that HCPBs would count a -16000 votes (that' minus 16000) as the machine did in Sancho's county in FL in 2000 when somebody made a mistake in trying to hack the result.

Ideally computers would do a better job of counting votes if the people who made the machines weren't criminals and the people who use them and pay for services related to them weren't criminals, etc.

Every election has to use human beings (who all have political biases and varying degrees of criminal psychologies) at some point and unless there is a means of verifying the machine count, you can kiss democracy good-bye.
Printer Friendly | Permalink |  | Top
 
eridani Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Mar-12-08 01:34 AM
Response to Reply #11
13. That was using a single race
If you are counting more than one race on a set of ballots, handcounting gets less accurate. I wasn't clear about whether this test featured single or multiple races.

BTW, another handcount is not an adequate check on an initial handcount. There ought to be some different method used, like weighing the ballots.
Printer Friendly | Permalink |  | Top
 
Festivito Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Mar-13-08 09:59 AM
Response to Original message
14. I imagine they did not do recounts of ONE ballot races.
That is to put on ballot in at a time and check if the machine agrees with the human.

These machines oft count marked ballots as unmarked, albeit consistently. Being consistently wrong does not make the machine more accurate.

It only shows the machine to be consistently and reproducibly wrong.
Printer Friendly | Permalink |  | Top
 
L. Coyote Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Mar-13-08 11:07 AM
Response to Original message
15. I have always advocated HCPB be counted at precincts, and machine verified.
All ballots counted twice. I do not advocate that each precinct have machine count ability, but that is also a preference. However, central (county) machine county is part of my advocacy because it places a check and balance on precincts by a different, broader level of oversight.

Printer Friendly | Permalink |  | Top
 
diva77 Donating Member (999 posts) Send PM | Profile | Ignore Thu Mar-13-08 12:22 PM
Response to Reply #15
16. machine verified - bad idea!
if you "verify" with the machine and the hand count is different, then what do you do? Hand count again with a machine number that you're trying to match? This is what some counties in CA have been doing with manual audits and they keep counting and improvising until the "hand count" matches the "machine count."

If you choose to have computers involved, then you have vendors gladhandling our votes. Not a very secure situation. No accountability once we turn our votes over to corporations.

The sort and stack method of handcounting has a 100% audit built into the process. If people are scheduled to count with hours geared to avoid fatigue, then handcounting is by far our best method.
Printer Friendly | Permalink |  | Top
 
L. Coyote Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Mar-13-08 04:14 PM
Response to Reply #16
17. There has to be a stipulation which is the official count, and what requires a hand recount.
The reason to verify is to force honesty at precionct levels.

Since PHC count proceeds transfers of custody to central tabulation, PHC should be the count of record.

What has to be controlled is possible precinct-level corruption. Counting twice for control, with distinct officials, should be a good idea.
Printer Friendly | Permalink |  | Top
 
Wilms Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Mar-13-08 09:20 PM
Response to Reply #17
19. Would Hand Counted audits of Hand Counted elections be an idea.
I agree that a mechanism for keeping the precinct honest may be a really important feature. But using an optical scanner seems so...computerized and vendor-involving.
Printer Friendly | Permalink |  | Top
 
diva77 Donating Member (999 posts) Send PM | Profile | Ignore Fri Mar-14-08 05:46 AM
Response to Reply #19
20. I like your idear!
:toast:
Printer Friendly | Permalink |  | Top
 
Wilms Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Mar-14-08 10:02 AM
Response to Reply #20
22. I've got the next round.
:toast:

:hi:

Printer Friendly | Permalink |  | Top
 
Sancho Donating Member (1000+ posts) Send PM | Profile | Ignore Thu Mar-13-08 06:05 PM
Response to Original message
18. There's a difference between being "different" in a hand count if you are ...
off by .01% . If we both hand count 2,000,000,000 million ballots, we may differ by a couple...BUT
if the machines have 10% "missing votes" (undervotes) or differ from the polls by several times the standard error...

well, that is what machines do!

It's not an exact match in counting that is important, but the accuracy of the count within reasonable confidence!


:dem: :nuke:
Printer Friendly | Permalink |  | Top
 
diva77 Donating Member (999 posts) Send PM | Profile | Ignore Fri Mar-14-08 05:49 AM
Response to Reply #18
21. excellent talking pt for pro-HCPB arguments!!
Edited on Fri Mar-14-08 05:50 AM by diva77
Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Tue Apr 30th 2024, 03:06 AM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Topic Forums » Election Reform Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC