The printed paper "ballot" may not be what the voter intended.
There are multiple stories of electronic voting machines changing the vote on the screen. The voter corrects it, and the correct ballot is printed. But, if the voter misses the change, then an incorrect ballot is printed. Sometimes, the voter has to correct it more than once.
Here in New Hampshire, we use paper ballots marked with a felt tip pen. They are scanned with an optical reader, but it's straightforward to do a manual recount of the official paper ballots.
Electronic balloting machines are an invitation to programmatic fraud.
I attended a Free Software Foundation annual meeting some years ago. The assembled group was asked whether they preferred electronic voting systems over paper ballots. Not a single hand went up.
The most common cause of mis-vote is operator error. I've watched people use the touchscreen devices; with so many voters both unfamiliar with the mechanics of a touchscreen and unsteady on their feet (the median voter age is about 50), and with the machines made of cheap plastic resting on a table, they grip the machine to stabilize it and themselves. Their gripping thumb rests on the screen and creates phantom signal that makes it hard to localize the touch coordinate.
This is the best argument I'm familiar with against the electronic machines we used to use, and our state switched to paper oval-fill ballots that are then electronically tabulated. But, I haven't seen the statistics yet on how many mis-counts we now have as a result relative to the old system and I'd be curious to see them... Ask yourself how confident you are that 100% of voters know how to successfully fully darken the oval. And since the system is anonymous, nobody looks over their shoulder to verify they understand how to do it before the vote is dropped in the box (though in the event of a close race, a manual recount will be automatically triggered and at that point, at least it'll be human eyes deciding whether the voter circling the oval indicates their intent to vote for that candidate).
I had a paper ballot print out under plastic after I made my in person vote in california. you have to agree that everything matches up with what you actually voted for. the times that the "system messes up and the voter has to change it" it's almost always either user error or the touch screen being calibrated incorrectly.
This is why designs usually require the voter to be able to inspect the paper ballot before finalizing it. Shenanigans like that would be caught, as the machine has no way of knowing how close the voter will look.
In yesterday’s California election (Los Angeles County, specifically), it had me review all of my choices on the screen (after I was done), then it printed out the form so I could review the printed copy. Then (very confusingly) the final step was to insert the form back into the machine. Basically everyone was confused by this review and reinsert process, but still it seemed like a good system for ensuring there are no glitches.
I have had arguments with disability activists about this- apparently, some interpretations of American disability rights legislation require electronic devices to be made available to allow voters whose disability prevents them using a pen to mark their ballot unassisted.
There are multiple stories of electronic voting machines changing the vote on the screen. The voter corrects it, and the correct ballot is printed. But, if the voter misses the change, then an incorrect ballot is printed. Sometimes, the voter has to correct it more than once.
Here in New Hampshire, we use paper ballots marked with a felt tip pen. They are scanned with an optical reader, but it's straightforward to do a manual recount of the official paper ballots.
Electronic balloting machines are an invitation to programmatic fraud.
I attended a Free Software Foundation annual meeting some years ago. The assembled group was asked whether they preferred electronic voting systems over paper ballots. Not a single hand went up.