\This post began as a comment to the post: Study shows new voting machines to be widely used in 2020 are vulnerable to hacking. As it became longer I decided to post it separately.
The new voting machines discussed in the original post have the voter make their choices on a digital screen. The machine then prints out the paper ballot and the voter is supposed to review the paper ballot to make sure it matches what they entered on the screen. The study cited in the original post showed that only 40% of the voters actually reviewed the printed ballot, and of those many missed intentional errors that had been inserted into the research machines.
My question is: Why do voters need to enter their choices on what is essentially a digital data entry screen and then have a computer fill out their paper ballot for them? This process seems less efficient than having voters mark the paper version of their ballot themselves. If nothing else these “data entry” machines add a step with no discernible added value to the process, a step that also introduces one more place for an error (or a hack) to occur.
When filling out a paper ballot themselves, the voter has to pay attention to the marks on the actual ballot that will be counted. During this process they are much more likely to catch mistakes than they are when reviewing a computer generated copy of choices they made in a separate step. Yes voters do sometimes miss-mark ballots, but at least those errors are more likely to be random than systematic.
Some of the best voting systems I have seen are where the state mails out the ballots to all registered voters. The voters then use a black or blue ink pen to mark their ballots themselves. Generally the voter has the option of returning their ballot by mail, dropping it in special locked boxes at public buildings, or by returning it to their polling place on Election Day.
This solves several problems. First, since voters get their ballots in advance, they have some warning time if their name has been “inadvertently”removed from the voter rolls. Second, it removes unnecessary hackable machines and software (and all voting machines are hackable) from the process. Third, the focus for possible hacks can be concentrated on the machines counting the actual paper ballots. Fourth, a sample of the ballots processed by each tallying machine is generally required to be verified by another machine. If there is a statistically significant difference, all the ballots processed by the suspect machine are counted again, often by hand.
Digital technology does not have to be used for everything. When it comes to voting, individuals filling out their own paper ballot still remains the least expensive, most secure, and most verifiable approach. I also would bet that even when individuals hand mark their paper ballots at the polling place the voting goes as fast as when they have to enter their choices on a computer screen, have the computer then print the ballot, then have the voter verify that the computer printed the results they actually wanted.
So why do we need computerized voting machines at all? Well two reasons come to mind. First, the companies that sell the machines would lose that revenue stream. Second, certain political entities would find it harder to hack elections to their benefit. I find neither of these compelling reasons to use electronic voting machines for data entry.