It’s good to hear confirmation of the politeness issue, thanks Jeff! The time issue is an interesting one – suggests perhaps a hint of extrinsic motivation at work? A great deal depends on the pedagogical context, I think. It’s a huge problem when people are marked based on discussion contributions, something I personally avoid like the plague because extrinsic motivation destroys intrinsic motivation, but any kind of pressure that reduces intrinsic motivation can be problematic and, in a traditional course leading to accreditation, it is hard to get away from that completely. For topics that interest me, outside any formal setting with imposed expectations, you have to rip me away from them, and damn the time it takes!

Mainly because of my concerns about intrinsic motivation, I’m not so sure about the gamification idea although, like all technologies, it’s a design issue and it can be done better or worse. Gamification works well in predominantly set-based sites like StackExchange or Reddit, where things like karma and badges are meaningful tokens for when participants are unknown, serving as a proxy for reputation. Though there are those that play the game to get the points and that treat them as a form of social capital to be accrued at all costs, for the most part karma points are constructively aligned with the purpose of the site. The point of engaging is the pleasure of engagement with others (discussion, support, help, learning), not the accrual of points – at least when it works (not always – see below) points are signals of success and a mechanism to ensure adequate quality, not drivers for further engagement.

In a group-based system, those trust issues are normally already resolved, so such mechanisms are not only not needed, but they get in the way. There are big risks that the game becomes the reason for doing it – it’s an assertion of power by the teacher or whoever sets the rules. Extrinsic motivation is the death-knell for a lively group community. It also quite literally risks being gamed, especially if marks are involved too. If the game is competitive then it can also be exclusionary. It’s a particular problem when things like ‘likes’ or ratings are used as part of the gamification process. The ‘like’ is a social token, an act of giving – it’s not about quality, it’s about the person. It might help build bonds in a group, but it does not signal value in what people are saying so, as a tool for assisting learning, has limited value. A rating, on the other hand, can play the same role for bonding but, if the rating is anything less than 5-star (and therefore actually useful in determining quality), it becomes a positive critique. This either means politeness rears its head again, or it can break social cohesion quite profoundly. Anonymous rating can help but, once you know someone, a low rating can feel like a betrayal even if you are not easily identified as its originator. Teachers get away with it because it is defined in the rules and they have an explicit authority role although, even then, I think it is a bad idea for aforementioned motivation reasons. This kind of thing is also a risk in networks, which almost always exist in groups too. Once networks form it can lead to issues where cliques build power and authority and become either over-supportive of one another (excluding non-members or the weakly tied) or over-combative (excluding just about everyone). Neither is normally good for whole-group dynamics. This problem rears its head in set-based sites as networks begin to form. StackOverflow is one of my favourite social learning sites because it does do the gamification very well and really uses the crowd to great effect through its collective karma-based tools, but it is a lot less useful than it once was because cliques and power-users have come to dominate many discussions, and they are not always as wise as they might be. They get and sustain their power through precisely the gamification tools that work so well for the most part. I suspect some of these issues might go away if anonymity were enforced, or if the rewards of playing the game a lot were not to gain such power, but it’s a trade-off against the consequent loss of the benefits of social capital. SlashDot is the most highly evolved system of this nature that really does work incredibly well and overcomes all these problems, as long as you have invested the enormous time and energy needed to manipulate its toolset: it’s news for nerds, but you have to be a fairly serious nerd to get real value out of it.

I’d be interested to learn of any ways of gamifying group or network systems that overcome these issues!

Jon