Who should recount elections: people or machines?
As the midterm elections are going to end on Tuesday, You would have thought that’s the end. But no! Recounting season is coming. With it being said, hundreds of close election results are likely to be contested even after Election Day. With it comes the ever existing debate about who should be doing the recounting, Humans or machines?
Jill stein, a Green Party Presidential candidate in 2016 requested for a recount in several key states, she made clear that she preferred human hands as the instrument for recounting the votes in an election by requesting specific number of hands in her petitions. When the state of Wisconsin refused her request she sued them by arguing that the computerized optical scanner machines, which are used by most districts in the state for counting paper ballots, were more prone to error making in the count. This was agreed by Hillary Clinton’s lawyer and even some computer security experts.
Whereas, the political scientists have a clear opinion on this matter. A professor of political science at MIT said that “There are a lot of assertions out there, and I see them constantly being made, that machines are error-prone and humans are perfect in counting, and that doesn’t bear out”.
As recounting became an increasingly normal part of the election cycle, experts are finding evidence that computerized optical scanner machines are better at raw counting. But humans are still necessary because recounts are often about interpreting human behaviour as much as they are about counting.
Edward Foley, director of the election law program at Ohio State University’s Moritz college of law said that: “As recounts are becoming more common, it’s hard to keep count on the exact number, though most recounts happen in local than state-wide, there are a lot of things unofficially called recounts when they are not.
So we can say that Americans are fighting over election results more than we did in the past, for which there is evidence. Richard Hasen, a professor of law and political science at the University of California, Irvine, came up with an estimate by searching a database of legal cases for the keyword “election” and variations of word “challenge”, and then reducing the results by removing cases that weren’t relevant. Because of this, his estimate is likely undercounting, but it showed a significant increase in the challenged election results over time. In 1996, he found 108 such cases where as in 2016, he found 337.
There are three main kinds of disputes in such cases. First, candidates and their lawyers argue over what ballots should be counted and which should be thrown out as ineligible. Secondly they argue over which candidate specific ballots should be counted. Finally, they argue over whether all the eligible votes were counted correctly which is actual recounting.
Stewart said humans are much better than machines at making decisions around the first two kinds of disputes but evidence states that the computers are better at counting. Michael Byrne, a psychology professor at Rice University who studied human-computer interaction, agreed by saying that: “That’s kind of what they are for”.
In 2004, Stephen Ansolabehere, a political science professor now at Harvard, published a study regarding the error rates over decades’ worth of recounts in the state of New Hampshire. From this study it is deduced that the scanners were actually better at counting than people.
When everything was hand counted, there was an average difference of 0.83 percent between the first vote counts and the recounts between 1946 and 1962. In 2002 general election, the hand count discrepancy was about the same 0.87 percent whereas the average optical scanner discrepancy was 0.56 percent.
When Ansolabehere and Stewart did a similar study, they found that the jurisdictions hat did their counting by computer had the lowest error rates. But that’s not to say humans are unnecessary. While computers count better, humans are necessary to spot some weird errors that humans make when they vote. And there are also trust issues with computers as they are vulnerable to hacking.
But surprisingly optical scanners and direct recording computers were trusted more than humans in 2012 cooperative congressional election study.
So even with all these risks of computer counts, we ourselves are giving us apprehension.