MOOC Research about peer interaction #Futurelearn

Steve Draper presented a paper by him and Sarah Honeychurch, MOOC Research about peer interaction at the FLAN meeting in Edinburgh which I'm liveblogging (so, again, these are my imperfect notes on what was said). He reflected on Honeychurch's experience of a MOOC where various platforms were used for peer interaction, none of which were on the official platform. This has implications for research: just looking at learning analytics from a main MOOC platform would be poor at capturing all peer interaction.
Draper identified types of platform and issues/usage of each Twitter, Blogs, Facebook, and Google docs. They can be used for different things, have different privacy issues, and also might be used in sequence, migrating between types e.g. chat in Facebook, then someone might move this to a longer piece on a blog or Google docs. Draper identified that learning is already amiguous with relation to public or private - the amount of privacy needs to vary (e.g. what students keep private, show to peers, show to tutor, show to the public). In terms of siftware/apps issues of changing privacy do not seem to be managed very well.
One way of classifying peer interaction is about roles and types of interaction e.g. collaboration on a joint project, discussion which leads to "different personal gains in understanding" or peer review/support. "Confident (life-long) learners move fluidly and fluently" between these different modes, able to undertake the different roles and seeing the benefits of engaging in each.
Draper identified that you can also classify the communication by "content type" e.g. aiming to get information, around understanding things, to vent feelings, to socialise. Sometimes this might be aimed at no specific audience (but rather to feel better) or it could be aimed at different audiences (peers, tutors etc.)
Draper then discussed the Vygotskian approach of scaffolding learning through interaction with an expert, and also about (my labeling) learning to think and practice in a discipline. In a MOOC the interaction is more likely to happen with peers; since there is a wide range of expertise, there will also probably be people above the level of a learner who could interact with him/her to progress. However, this interaction has to be designed and scaffolded.
Summing up, there was interaction by platform, by task/goal interaction and by how peer interaction supports the neo-Vygotskian learning process, and each type has different research implications.
Photo: blurry impression of Draper anwering questions, as seen over my netbook.