Part of this should probably be in the Pit, because I’m just about screaming-mad about the inept way that elections (to Chair and various committees) of my academic department have been run over the years (basically they change voting requirements, eligibility, vote-counting norms and about eight other aspects of departmental elections so randomly, it’s a wonder that someone hasn’t sued the university on one legitimate cause or another), but I’ll try to calm myself in the hope of soliciting some ideas for improvement. [/deep cleansing breath./]
How should we in the 21st century organize departmental elections? I’m proposing that we somehow use our computers to vote, but my technophobic colleagues look at me when I make this proposal like I’ve imploded myself. They give me blank “How-would-we-do-that?” looks, so before making suggestions I’d like to ask how other people have done this, or would do this.
At this point, we’re photocopying paper ballots (that of course could just be photocopied themselves ad infinitum) and asking people to stuff them in a ballot box that sits unguarded all week long. I’ve tried pointing out that fraud is practically inevitable at some point, and the possibility of fraud is present in every election we’ve ever had under these conditions. (More blank stares.) I need a voting system that
allows multiple options (i.e., several different offices, some with different erms of office)
allows anonymous voting
restricts voting to eligible voters
is otherwise protected from voting fraud.
and anything else you’d like to suggest as necessary. We’re a department of about thirty members, and we all have e-mail addys that identify us by our real names. I was thinking of having people vote by attaching an anonymous ballot to an e-mail, but sending that ballot-with-attachment to one member of the Election Committee, who would make sure that there was no more than one vote from each eligible voter and then forward the attachments to the other members of the election committee (without the identifying e-mail) who would then open the attachments and count the votes. But maybe this is a problem because people wouldn’t feel that their anonymity was protected since the person the first e-mail was being sent to could open the attachments before forwarding them along.
Anyway ,that’s only one idea. Does anyone have any other ideas for a fair, anonymous voting system, electronic or otherwise, on a small scale?
Wouldn’t a lot of your problems be solved if you just shortened the amount of time people had to vote? You could just have a departmental meeting, hand out the ballots, and have people fill them out and put them in the ballot box right there. That would drastically reduce the possiblity of fraud.
I know you might not be able to do that, for scheduling reasons.
Paper ballot papers are the easiest. You can have a secret ballot by mail by using two envelopes: put the ballot paper inside one envelope, then that inside a second envelope that identifies the voter. Check the outside envelopes, then put all the inside envelopes together, mixing them together before you open any of them. The whole thing should be done in front of witnesses to verify the whole procedure, including the preservation of anonymity. On a small scale, it has to be cheaper and easier than any computerised system.
It’s really hard to get us all together in the same room at the same time. We all have schedules that put us in the classrooms at different times of day, and that have us on campus on different days. (Asking people to come in on their off-days might not seem like a big problem, but it is when some people commute from very long distances–coming in on an off day is something that people really hate to do, and on “on” days, they’re often in class all day long.) So I’m really hoping someone has an e-solution, which would be elegant (and make me look very smart in the bargain for suggesting it.)
Also one other problem I didn’t mention: distribution of ballots. We’ve been putting blank ballots (in envelopes) in people’s mailboxes, which anyone with two feet has full access to, 24/7. Unbelievable as it sounds, when someone complains that he never got a ballot, which happens every year, we just hand him another. If we ever had one more ballot than we had voters, we would have no alternative, as I see it, than to declare the entire election invalid (assuming it was close) and just start over again from scratch. Makes me nuts.
Why is an academic department using anonymous ballots? The parties involved are supposed to be responsible and should be accountable. I suggest you challenge the paradigm here.
What kind of fraud are you suspecting? I just suggest having the election ballots in the hands of one or two people who cross your name off when you pick it up. The ballot could then placed in an envelope and those given to a third person, or slid under their door. Just how much distrust is there and what can the department officials do that has people so concerned? Or is it jut you? :o
I don’t believe that there is a simple and elegant e-solution. The main problem with elections is people casting votes fraudulently, and that is easier to do in cyberspace than on paper.
You can have a ballot box (as kunilou suggests), or a mail ballot. In both cases:
(1) You need to check people off as they vote, so no one votes twice. With a mail ballot, that can be done with a declaration attached to the ballot, which is separated from the ballot before it’s counted. (That gets around the problem of ballot papers not being received, and duplicates being issued.)
(2) If you use a ballot box, you need to seal it so that ballot papers cannot be removed during the ballot, and you need to douible seal it so that ballot papers cannot be added which the ballot box is unattended (e.g., while the office is closed, if you have a balot over several days).
(3) You make sure that ballot papers cannot be forged by the simple device of the returning officer signing each ballot paper, and keeping a count of the number of ballot papers signed and issued.
I might add that I’ve had some experience running elections, including preselection ballots for state and federal members of parliament in Australia (the equivalent of primaries in the US, except that they are run within the party, not by the government). It’s easy to build in double and triple checks for paper ballots, while preserving a secret ballot at the same time; I don’t believe it is so easy to do the same thing for electronic ballots.
It just seems very wasteful of our time to take trouble to distribute ballots in people’s mailboxes if we respond to “lost ballot” complaints with “Here, have a couple extras while you’re at it.” If it’s so easy to commit fraud, and we’re okay with the honor system, why not just have a stack of ballots available and say to people “Please, only ONE vote per person.” It would be a lot simpler, no?
I think the principle of a secret ballot is pretty basic. We have hand-ballots, at departmental meetings, for routine issues, but whenever something is potentially serious, we vote by secret ballot. Not to have secret ballots is to open oneself to intimidation, no?
[ol]
[li]Number or code each ballot, keeping a list of the numbers issued.[/li][li]Shuffle the ballots and issue them randomly to maintain anonymity. Anybody that loses their ballot…tough luck. [/li][li]When you collect the ballots, check off each ballot against the list. Any not on the list, or any duplicates, indicates malfeasance.[/li][/ol]
Or, you could make a Web-based voting app based on using public key encry…never mind, I’ll get off my hobby horse.
Just due to the whole mistrust for electronic voting, it seems unfeasible. I have no doubt that an elegant online solution exists - heck, I’m pretty sure my school does online voting for student elections, but since I don’t bother to vote in them I’m not positive :). But it’s not a system that’s easy to win converts for. I think your best bet is just maintaining adequate security over the ballot box - is there a department secretary who’s not already so overworked that he or she couldn’t handle signing off on those who’ve voted and those who haven’t?
Or do this: give out each ballot in two envelopes: an inner one, preferably with something official-looking printed on it (just to make it ever-so-slightly harder to counterfeit) and an outer one, printed with a code specific to each voter (use a long code generated by a mathematical algorithm to make them impossible to fake.) When they fill out their ballot, they seal each envelope, and the envelopes can thus be kept intact until the counting. If anyone complains that they lost a ballot, or that theirs was stolen by someone else, you can pull the one corresponding to their code out of the stack, destroy it, and hand 'em a new one (also with a code, and the list of codes and names will have to be updated). When votes are counted, the outer envelopes are checked against the list, so that voters (and abstainers, possibly) can be accounted for. The outer envelopes are then removed and destroyed, thus removing any tie between the vote and the voter. The inner ones are shuffled, then opened and counted.
Just off the top of my head. It seems like a method that could dispense with the need for each ballot to be signed and countersigned for, though it may well end up being more trouble anyway. Sort of an interesting thing to think about, though.
Interesting idea, since trying to copy votes in this system would give a ballot (code) duplicate to another and both could just be destroyed and the person asked to revote once.
You know, I’m amazed that there are elections in the first place. Most math departments i’ve seen either let the guy who wants a given position (chair, director of grad or undergrad, etc.) have it, or more often pick someone in a sort of rotation and tell them it’s their turn in the barrel.
Well, it’s often a thankless position that requires some persuasion to find willing candidates. But we’re also ideologically divided, and concerned that maybe “one of THOSE maniacs” will be in charge of setting policy for the next few years.