Hand Counted Paper Ballot Zealots—come out wherever you are!!!

 I never cease to be amazed at how many activists and academics that are election groupies are zealots for hand counted paper ballots (henceforth, HCPB).    In many cases you won’t reach that conclusion right away.  They will tell you all about their concerns with DREs and how evil they are.  They will suggest that optical scan ballots or absentee voting with optical scan ballots are the way to go.  But what they won’t tell you is that in the final analysis, they also hate optical scan too. 

 They really dislike ANY electronic voting system, but the attention right now is towards DREs.  But make no mistake; once the DREs are on the run, they will start the attack on Optical Scan systems too.  We have already started to see this in Florida where the DREs have been decertified, and now optical scan systems are being questioned.

 Much of this is a dangerous trend that needs to be serious countered.  Most election administrators have no desire to count all ballots by hand, and this movement needs to be addressed before it gets out of hand.  Now, having said this I’m not suggesting that DREs are perfectly fine voting systems.  Nor will I say that optical scan systems are fine as well.  As many of you know, I am the developer of a system of voting using digital pens and Anoto digital paper.  We are actively trying to get it shown to the US election market.  In that effort I come across any number of activists and academics and I ask their opinion of it. 

The response is often telling.  While the system has clear benefits over optical scan systems, many of these people immediately being to question its use.  It’s as if they have already made up their minds before really learning about it.   I think that it is a shame and it is wrong.  I have learned that a few pointed questions will bring the HCPB ballot zealots out of their shell.  Of course, you can always ask them point blank: “Do you feel that only hand counted paper ballots should be used in US elections?”  See what you get after that question.

All of this leads us to one natural question: are HCPBs better than other voting systems?  I often see activists quote this article from Caltech/MIT Voting Technology Project.  The actual title is, “Using Recounts to Measure the Accuracy of Vote Tabulations: Evidence from New Hampshire Elections 1946-2002”, by Ansolabehere and Reeves. Activists claim that it supports the idea that HCPBs are counted accurately (as determined by having them recounted).  In fact, the article concludes just the opposite.  It shows that optical scan ballots are recounted far more accurately (about twice), and that HCPBs are far worse off.  What is mentioned in the article, but not discussed is that the HCPB used were in New Hampshire and used in small towns. 

I have no doubt that HCPB used in a small New Hampshire town can be counted fairly well.  But that proves nothing about their use nationwide.  Try that in Philadelphia or any other large jurisdiction with a large ballot and with less educated counters. 

There is no evidence whatsoever that we have the ability to hand count paper ballots nationwide in a major election and do it with a very low error rate.  There is a ton of evidence that suggests that this can’t be done.  If you have taken the time to read electionline.org’s article on Audits (March 2007), there are some interesting quotes in there on hand audits (either optical scan, or VVPAT vs. DRE totals).

Larry Lomax from Clark County , NV suggested that each VVPAT hand count took up to 4 minutes to complete and errors were numerous.  Yes, VVPAT paper is different from a paper ballot, but it still shows the slowness of hand counts.

If you try to find academic work that discusses the accuracy or lack thereof of HCPB you will have a hard time finding it.  Doing a Google search on hand counted paper ballots will result in tons and tons of hits from activists, shrill and not so shrill, arguing that HCPB are the way to go.  For the most part, their arguments fall short.

Let’s compare some myths and realities about HCPBs.

Myth: Ballots are counted by hand in Canada and it obviously works there.

Reality:  The truth is I’m not sure we do know how well it works in Canada.  Most people simply point out that they use HCPB.  I haven’t seen anyone provide data that shows the accuracy of their counts or recounts.  In addition, Canada is a very different nation from the US.  The Canadians manage to do all sorts of things that we can’t do it the US .  They manage not to shoot each other ( Toronto notwithstanding), and we can’t do that. If you spend any time in Canada you begin to notice these differences.  So the fact that it is done, perhaps even very successfully, in Canada is no evidence that we can do it in the US .

In fact, if you think about the size and number of our large jurisdictions, the education level of our poll workers, and the political nature of elections—you begin to realize that we probably can’t count paper.  In many of our major cities, we can’t graduate 50% of our citizens.  How in the world can we expect HCPBs to be accurately counted?

Myth: If we have HCPB I know my vote will be counted.

Reality: Again, nothing could be further from the truth.  Once I vote my ballot is put in the ballot box.  I don’t know, even if I watch the hand count, that my vote has been counted correctly.  A person could call out the wrong name, or make a mark for the wrong candidate.  Any number of errors can occur in hand counting.  I have no ability to prove or show that MY vote is included in the final tally.  HCPB do not guarantee this—despite what many claim.

Myth: Poll workers won’t mind staying late to get the count right.  If it takes all night, then so be it.

Reality: I’ve seen some activists make such claims.  The truth is that by around 7pm, most pollworkers have been at it for 14 to 15 hours straight.  Do you really expect these people to be at their sharpest in terms of counting?  It would require another shift of volunteers.  And who will we select?  Do we have to screen them ahead of time to make sure they aren’t political operatives?   We then have to hire another 2 to 4 or more people for every voting location in a county. In a large county with, say, 500 voting location that means recruiting another 1000 to 2000 people to do the counting on election evening.  Most people have NO IDEA how hard it is to get the poll workers needed for Election Day itself.  Trying to find a few thousand more people is almost laughable.  This is another case where activists spout off these remarks and have no idea about the real pressing needs of election administrators.

Myth: Once ballots are counted in the precinct, then we can rest assured that the counts are accurate.  The ballots merely need to be securely returned to the election office.

Reality: Again, think it through to the end.  Suppose that we have hand counted all the ballots in the precinct and we have reached our totals.  Not so fast.  Actually someone has to do a final sum of the results; do we do that once or twice?  Do we do it in teams or merely trust someone to sum up the hash marks?  Once the totals are put on a sheet of paper, how do we transmit or see that those results get back to the election office?  Can we be assured that they will not be altered, lost or destroyed? 

When they arrive at the election office, how are other results “merged” with your precinct totals?  Do we take the numbers for each precinct and put them in a spreadsheet such as Excel?  If so, isn’t that using a computer to total results?  How can we trust Excel if we haven’t reviewed the Microsoft Source Code for Excel?  Are we then to take the totals from, say, 500 precincts and add them by hand?  Who is to do that?  How is that to be done? 

The point is this: There is still any number of ways that errors and mistakes can take place with HCBPs and we have not really thought out the process enough to reach an intelligent decision about them.

The upshot of this is that election administrators should begin to really think about HCPB since there is a sizable academic/activists community out there that is pushing for them.  At present, I would suggest by starting to ask activists point blank if they are wedded to the HCPB concept.  If so, then they will reject ANY electronic voting system.  As a result, not even optical scan ballots are safe from their attack.  So if you are moving away from DREs on the hope that this will keep the activists happy, you are sadly mistaken.   You will be attacked for using these systems too.  Your support of optical scan can only be made if you have a good grasp of the weaknesses of HCPB.