A Few Thoughts On Paper Ballots
One of the items that the Georgia General Assembly will deliberate upon is how Georgia voters will cast their ballots in 2020. At this point, there will be some sort of paper trail. The question is, will the voter use a pen or a machine to mark their ballot?
I’ve had the privilege to serve on the absentee ballot review panel as the Republican representative when I lived in Walker County. I have also worked as a software developer and database administrator for a number of years. Both experiences have proven this one axiom: give a user an opportunity to mess something up, they probably will.
Paper ballots filled out by hand introduces variability into the process. I would say that the vast majority of voters would be able to fill out the ballot and be read by the electronic tabulator just fine. But! There are always “a few” ballots where the voter either accidentally votes for the wrong one and corrects their mistake or just outright changes their mind. Enters the review panel to interpret the intent of the voter.
“Most” of the time, interpreting the voter’s intent isn’t a problem because of obvious markings (they circle the name, write a note, something). Occasionally (rarely, but it does happen), the panel has to agree that an over-vote will have to be counted as a blank vote because the intent can’t be determined. Usually it doesn’t affect the outcome of the election, but there have been a few contests that come down to a single vote.
Georgia just wrapped up a contentious election for governor. There was a lot of debate over the handwritten paper absentee and provisional ballots. Personally, I’d rather not have a Year 2000 Florida hanging chad moment when both the Democratic Party of Georgia and the Georgia Republican Party sue and counter-sue with a tight election coming down to the interpretation on what a voter may or may not have meant by a stray mark. Am I being a bit over-dramatic? Maybe, but why put ourselves into that situation?
“But hand-marked ballots are cheaper!” The upfront cost “may” be cheaper, but the long-term costs of printing ballots would likely be pushed down to the individual counties. I don’t have hard numbers, but I’m operating on a hunch. I’d rather keep a new election system affordable for our counties rather than trying to take the cheap way out as a state.
Personally, I’m for ballot marking devices. It gives the voter an experience that they are used to and then prints a ballot on-demand to be fed through the tabulator that records the vote. The touchscreen interface wouldn’t record the vote, but it would have a similar experience where it would alert the voter to skipped races. That’s something a pen-and-paper ballot can’t do.
Secretary of State Brad Raffensperger made a big ask of the Georgia General Assembly during this week’s budget hearings of $150 million to purchase new voting machines for 2020. That may signal leaning towards ballot marking devices. Yes, it’s a big upfront cost, but in terms of government spending, shouldn’t a priority of government spending be fair, secure, and accurate elections?
Another idea that has been tossed around is a risk-limiting audit. I will save the details for another piece, but the skinny is that a hand count of statistical sampling of the paper ballots from the election are compared to the results from the tabulators. Few discrepancies means that there’s good confidence in the election results; substantial discrepancies would lead to a 100% hand recount of the election.
From my perspective, paper ballots without some sort of post-election verification make the paper ballots nothing more than a big security blanket–comforting, but mostly useless.
I hope our General Assembly will ask thoughtful questions and make a reasonable decision about the best way to proceed.
About Author
nsmith
Father. IT Geek. Former 14th District GOP Chairman (2015-2017). Former Walker GOP Chairman (2009-2013). Resident of Chamblee. American. Georgian. Conservative.
Add a Comment
You must be logged in to post a comment.
It’s not this complicated. Electronic Machine voting with paper receipts. Voter would then be able verify (in real time) that their vote was entered in properly. If we’re feeling frisky, improperly cast ballots could even be corrected on the spot. Finally, printable copies can be made available to the state in case of recount or audit.
The machine ballots this last election undercounted 80,000 ballots in the Lt Gov race, a few percent of the total ballots. Machines to mark ballots costing three to five times hand marked ballots are out of the question until the cause of the undercount is established.
The machines didn’t under-count those votes. The typical Abrams voter couldn’t distinguish between the Governor’s race and the Lt. Governor’s race. When they saw the LG ballot come up, all they read was the word “Governor” and thought to themselves, “Well I already voted for Abrams” and then skipped to the next page.
The undercount, based on that, would be amplified in heavily Democratic-voting precincts. There are many other possible explanations if that undercount is found to have disproportionately occurred in Democratic-voting precincts.
One is that machines may have been out of adjustment. I personally had an experience with an out of adjustment machine on one screen years ago.
Good to know that “rarely”, presumably single digits, is more important than an average 500 votes per county. And Republican wonder why the numbers of fiscal conservatism are laughing stock.
I’m not $100,000,000.00 sold on the difference between the two options. That may be just the cost of one divorce attorney to Jeff Bezos but it’s still a large chunk to this taxpayer. I’m also confused why the “conservative” party would be pushing the more complicated and expensive solution without an obvious ROI for such a large sum. Some of the reasoning seems spurious at best. Voter blacks in the wrong circle? Unlike the absentee ballots you cite, at the polls this can be remedied with a new piece of paper and a shredder.
You speak of the end user being able to screw up most anything though I have found in my 35 years of IT experience that the KISS principle is often the most prudent route to avoid these hazards. In this case this also applies to the poll workers. Without cumbersome machines to setup you don’t have to worry about charging batteries or forgetting power cords. Without machines you don’t have the worries or excuses of restricting those machines in certain precincts or counties as we had in the last election. You speak of one vote counting, but how many votes were lost with people seeing hours long waits to cast their ballot? Long lines can be resolved by providing more pencil and paper, not trying to obtain more of the limited machines available. Or weak excuses given to squander those machines away in warehouses in selected precincts.
So yeah, until convinced otherwise I see the more complicated and expensive solution as a partisan ploy to keep the party in power going regardless of the will of the people.
I hadn’t considered that equalization of waits in distributing machines is another argument for hand marked paper ballots. The limitations on the numbers of people marking paper ballots at one time are much less than those using machines. One of the reasons I nearly always vote absentee is the possibility of having to wait on people at the polls deciding how to vote on many of the things on the ballot.
Consider that Georgia Tech professor Wenke Lee, the lone computer science and cybersecurity expert on the commission, recommended hand-marked paper ballots, The only other two Committee members opposed to the machines were two Democratic lawmakers.
Anyone have a list of the other 13 on the committee besides Raffensperger and Fleming that voted to recommend machines? One or more were County election officials.
“The three dissenting votes Thursday came from the commission’s two Democratic Party legislators, state Rep. James Beverly and state Sen. Lester Jackson, as well as its only cybersecurity expert, Georgia Tech professor Wenke Lee.
So that means that the votes ‘for’ were:
The vote was 13-3, so I don’t know why there are 15 names there. I have made some inquiries but no answer yet. It is possible that the two legal representatives (Jablonski and Russo) didn’t vote.
I have been advised that the following two members were absent and therefore didn’t vote:
Michael Jablonski, general counsel, Democratic Party of Georgia
Lynn Bailey, executive director, Richmond County Board of Elections
Paper ballot or neatly generated paper ballot via input. Fine with me. Just NO bar codes. Read the individual marks, not an easily corrupted bar code that can rig the vote outcome. I often leave a blank if there is only one choice, and I disapprove of the candidate. I should have that choice to a blank vote. Although I like the warning letting me know when I do that and have another chance to cast a vote before finalizing the ballot.