Punishment for cheating in online quizbowl

[ryanrosenberg]

There's been a lot of discussion in the cheating warning thread about how people should think about punishment for Eric Mukherjee's cheating, and I wanted to start a separate thread to discuss more generally how the community should deal with cheating in online tournaments. To my mind, there are two main questions, which I'll lay out as a start to the discussion.

What should the potential punishments be for cheating in online quizbowl?
I don't have much to say here that hasn't been said in the other thread; I'd recommend reading through that (starting roughly here) for a sense of what people who have posted are considering.

How would punishments be administered?
Currently, ACF and NAQT both have sanction mechanisms that cover cheating, and each organization could theoretically ban players that were shown to have cheated in online tournaments. TDs of non-ACF/NAQT tournaments are also free to ban players as they see fit. However, there's no coordination mechanism. Different parties might have different ideas on the length of a ban, which could lead to confusion over whether a player can play a given tournament or not.

I think there's a reasonable argument against having this sort of coordination mechanism; it seems natural that TDs would follow the lead of ACF and/or NAQT in their punishments, and so a coordinating committee would be a lot of work to set up something that would happen in practice anyway. Such a coordinating committee also wouldn't have power over ACF or NAQT decision-making, and so would not necessarily be able to enforce the same sanction for all tournaments in the calendar.

[Cheynem]

I'll just lay out some of the questions that emerged from the other thread real quick. I think Ryan's analysis of who should enforce is solid.

What should the baseline penalty be for online cheating?
Almost everyone seemed to agree that at a minimum, the guilty party be temporarily banned from playing online tournaments. In general, most people thought a year seemed fair as a baseline (that can be adjusted up and down).

Should this affect a player's participation in quizbowl "in real life"?
There was not much of a consensus here. In many cases, online cheaters have not attempted to play actual quizbowl. I think this is a more difficult question, and the thoughts ranged from "not at all" to "extremely, so."

Are there extenuating factors in a player's penalty?
Most people thought yes, although what factors were not agreed upon. For example, a player's background--do they have a history of cheating or harmful behavior? Do they have a history of positive contributions? Were they contrite and make a proper confession or deny what they had done? Are they a high schooler or middle schooler or an adult? Obviously some people thought some of these factors mattered more than others. In general, people seemed to be more willing to impose slightly less strict penalties on contrite cheaters who made proper confessions and had overall good histories.

Is all online cheating created equal?
Most would probably say no, although there wasn't really a consensus as to how this affected penalties. Certainly no one is advocating wholesale bans on people who googled a question during a Discord packet reading, but there is some speculation as to if someone who googles a few questions here and there at an online tournament (as bad as that is) is perhaps not as unethical as the person who uses deceit to steal a whole set in advance.

[naan/steak-holding toll]

I would personally oppose in-person sanctions unless a person is shown to have cheated at one online event, and then cheats again after they're shown to have cheated previously. I think a one-year ban from online events is reasonable, enforced independently by TDs. Not sure that we need a formal governing body for online quizbowl, or that existing organizations need extra tasks on their plate, but the latter is certainly well beyond my knowledge.

[cchiego]

How are you determining if someone cheated or not?
This to me is the bigger problem--there are very, very few admitted cheating cases out there, but lots of "suspected" ones. What's the standard of evidence that will be used to determine if there was cheating or not? Would something akin to the NAQT reports on Watkins-gate be considered definitive in the absence of a confession? Could such evidence actually exist for most online events?

Would NAQT and ACF actually get involved (especially if the person is a retired/no-longer-enrolled player)?
I'm not sure that outside organizations would have much incentive to get involved in these situations, especially if it's for non-eligible players and didn't take place at events that they sponsored.

[Mike Bentley]

I agree with Chris that proving online cheating is going to be very difficult. But I also think it's reasonable to have an in-person ban on your first offense so long as this policy is announced ahead of time (and possibly if it's not announced ahead of time, but it's much clearer in this case). Catching someone cheating twice in an online tournament seems like a really high standard for an in-person consequences.

[No Rules Westbrook]

I suppose my opinion is that online tournaments are currently just not valid quizbowl. Until we develop some sort of framework where people can compete long-distance with reasonable controls in place to prevent cheating (at least controls that make it very difficult to cheat - in the same way that it would be very difficult to cheat at an in-person event because people can just see you looking things up), I think these events are just fun things to do and we should assume the results may be tainted. (and btw, results of online events could be tainted for other reasons too - internet connections and lag issues, technical problems etc.)

I don't wanna be ridiculous, but I'm reminded of people doing Phone-a-Friend for Who Wants to Be a Millionaire? or calling into a radio station for a trivia contest - and it's obvious that they're furiously googling answers as the question is being asked. Now, obviously in those cases, the organizers know that this is going to happen and just kind of allow it (like "Hey, if you can google this in the time alotted, I guess you win") whereas the people doing this at online quizbowl events are clearly acting in violation of agreed rules. However, I view performances in both of these settings with almost the same quantity of salt grains.

If it's an actual tournament played online (not just goofing off in Discord or something), and it's proven fairly conclusively that someone cheated (proven by "clear and convincing evidence" if you like legal jargon?), then I'm fine with "first warning = you can't play online events for one year".

But, I don't think we need to get our collective panties in a bunch over cheating at online events right now, because I think we need to just acknowledge that we don't currently have the capability to make these events legitimate by any definition.

[matthewspatrick]

Would NAQT and ACF actually get involved (especially if the person is a retired/no-longer-enrolled player)?
I'm not sure that outside organizations would have much incentive to get involved in these situations, especially if it's for non-eligible players and didn't take place at events that they sponsored.

Disclaimer: while I hold Member Emeritus status with NAQT, I in no way whatsoever speak for them. Nothing I say be should be taken as an indication of what NAQT may or may not do officially, nor should what I say be taken as a recommendation for what they should do. With that out of the way...

I can very easily see why NAQT, ACF, or any other quiz bowl body could have an incentive to get involved when someone is found to have cheated, even if it is not in one of their events, and even if it does not involve a potentially eligible player. These organizations rely a great deal on trust. They have to trust that the players in their events are competing honestly and ethically, and they have to trust that their writers, editors, staffers, and yes, even their members, are doing their work diligently, ethically, and are not skewing the competition toward any particular teams or players. It is reasonable for these organizations to ask, if a person has committed a breach of trust in one aspect of quiz bowl, how much of a risk is it to place trust in that person in other aspects?

The answer to that may well vary from case to case, but it's reasonable for non-involved organizations to be asking it.

[Sam]

How are you determining if someone cheated or not?
This to me is the bigger problem--there are very, very few admitted cheating cases out there, but lots of "suspected" ones. What's the standard of evidence that will be used to determine if there was cheating or not? Would something akin to the NAQT reports on Watkins-gate be considered definitive in the absence of a confession? Could such evidence actually exist for most online events?

I agree with Chris. In general, raising the probability of getting caught has a higher deterrence effect than raising the severity of the punishment. Given how low that probability is for online tournaments, most of the gains probably come from raising that, rather than crafting the perfect punishment.

In the other thread someone mentioned adopting measures used by universities to catch cheating for online courses. UMN has a list of examples here. Other schools probably have other ways. None of these are very easy and involve additional software, but I think that's inescapable if people want online tournaments to have anything close to the level of security as in-person. We may also decide that online tournaments are going to be very, very imperfect substitutes for in-person and tolerate a higher level of cheating because the deterrence is too costly. It probably still makes sense to have a formalized punishment for the sake of fairness, but we shouldn't trick ourselves into thinking we've solved the problem. (To be clear, I don't think Ryan or anyone in this thread is claiming as much.)