That "paper trail" won't help

From Renee Kovacs:
In the discussion of electronic vote verifiability one VERY critical element has been overlooked when calling for a “paper trail.” What if the voter simply doesn’t use it, does not look at the paper print out to verify it? This is not a trivial issue – as you will see from my poll watch experience described below. I was a poll watcher with Election Protection in California in a county that used all Diebold DRE machines. Below is an excerpt from my narrative to the EP people, with my description of a critical problem with the “paper trail” that is supposed to be an independent verification of the voter’s intent. This problem has nothing to do with the technology, but is a user issue and so much harder to correct.

… None. of the workers knew that the plastic flaps over the printers were supposed to be open so the voters could verify their choices, and did not understand why. I opened the flaps on machines several times in passing, and at 8:00 AM reminded the poll workers that they were supposed to be open. They did not check after every voter and, unfortunately, many voters unthinkingly closed the flap (if open) when they finished, so the next voter didnât even know he was supposed to see the printed votes. I brought this up several times in the day and 1:30 PM had a conversation with (a friendly poll worker) about the importance of voters verifying their vote. Both the supervisor and he believed that the voters could verify on screen, and so clearly they had not grasped that the purpose of an independent printed ballot was a cross check against the screen display, that it was technically possible that the screen and printed votes would not match (due to glitches or software tampering), and that the printed ballot was necessary to have in case of a recount. Once I explained that, he understood the situation and at 2:00 he quietly taped the flaps open for his precinct, and they stayed open until closing; a very friendly worker at (the second precinct) also began checking the flaps after voters beginning around 4PM.

I would guess that only about 20% of voters looked at the printed paper, so who knows how many printed ballots do not match the screen ballot. In other words, the “paper trail” is not working as planned, and if a recount was done based on these paper ballots it might not reflect what the voters intended at all. I would consider this paper trail useless, in theory. If one must use DRE in future, the remedy for this is to REQUIRE the computer screen to instruct the voter in large clear RED letters to verify the printed votes and include a statement on screen that this is necessary for recounts. (I do not know what the computer screen says now – I have always had paper ballots.) In addition, there should be large posters on the walls explaining the need to verify the printed vote, which the poll workers can tell the voters to read. before voting

T he vote machines as set up were actually quite short and many men had to stoop way down to see the screen, so reading the small printer located below the screen level off to the side was even more awkward for them. Moreover, the print is not especially dark, making it even less likely that the voter will review every item on the printer, in this case 15 offices and 15 propositions! (Average time to vote was 6-7 minutes.) Raising the machines 6 + inches would still be comfortable for the average voter (5 ft 4 – 6ft), and a big help to taller, AND relocating the printer to be up at eye level next to the screen would be a HUGE help for everyone . If the voters don’t verify the much prized “paper trail” it will be just so much paper…


So the apparent victory in demanding a paper print out for DREs may not be such a victory after all. Strangely, ergonomic issues seem rather important.
Paper ballots are a really nice, simple solution. One size fits all, what you see is what you get.

And much easier to monitor and verify one optiscan than multiple DREs.

Leave a Reply

Your email address will not be published. Required fields are marked *