Tomorrow is the official start of the Videogames and Learning Coursera MOOC developed by Kurt Squire and Constance Steinkuehler at University of Wisconsin Madison. In this post I compare the pros and cons of Coursera's more expository "xMOOC" format with the connectivist "cMOOC" format advanced by Siemans and Downes and show how the the more modest "big" format of IU's Big Open Online Course is turning out to be a useful interim context for design-based research of hybrid formats for future massive courses that can exploit the advantages of these very different formats while minimizing the negatives.
As part of a university-wide experimental partnership with Coursera, Kurt and Constance have been leading the creation of brief (approximately 15-minute) videos and other content for their Videogames and Learning MOOC. They assembled a talented team and it looks like they are going to deliver one of the best MOOCs ever at Coursera. They have gotten tons of press; one blogger included their course in "5 MOOCs educators should take as students."
I am looking forward to following along. Kurt and Constance are leading learning scientists and I suspect they will set a new bar for learning in this wildly popular open course format. If anything just check out their awesome animated intro video. Thanks to their reputations and Coursera's reach, something like 30,000 students are registered for tomorrow's launch.
Competing Formats for Massive Open Courses
Much has been said about the so-called "xMOOC" format that Coursera has gotten very good at. While Kurt and Constance and their team had a monumental task creating such high quality content in advance, I assume they can now relax while tens of thousands of us individually follow along each week. While some of us will get onto the discussion forums, most of us will squeeze in time to watch the videos, skim the research articles, and take the quizzes. It will be interesting to see if they can boost the 5-10% completion rate that I understand is typical for Coursera and other xMOOCs.
Much of the criticism of this more expository format (i.e., efficiently exposing individuals to content) comes from proponents of the connectivist "cMOOC" format pioneered by George Seimans and Steven Downes and which continues to thrive at change.mooc.ca. These formats and recently introduced variants like the Distributed Open Collaborative Course are embodied in the ongoing Dialogues on Feminism and Technology course. A key feature of the connectivist format is that much of the content and most of the interaction is crowdsourced. This reflects the crucial assumptions of connectivism about the value of disciplinary networking.
As I wrote in a previous blog post, there are going to be many open course contexts where both expository and connectivist formats are appropriate. Indiana's current open online course on Educational Assessment includes some highly structured content on the guidelines for developing different assessment formats. We are confirming that our highly interactive wikifolio format is great for learning this content (you can try out our cool new assignment structure at our new staging site set up to protect participant privacy at the active site). This interaction will become really important in the third part of the course focusing on Assessment Policies. We can barely scratch the surface of assessment policies; our only goal is helping individuals discover and discuss how core assessment policy concepts take on meaning in particular educational contexts.
Scaling up our BOOC
We called the course at IU a "Big Open Online Course" (BOOC) to emphasize the fact that we were not ready to go "massive." Google generously supported our efforts to build some of the interaction features refined in previous small online courses into a larger course using Google Course Builder. Capping registration at 500 and assuming 75% attrition ensured that we would have the bandwidth to tweak our wikifolio format and refine the crucial peer-commenting, peer-endorsement, and peer-promotion strategies. As summarized in a recent post and to be elaborated on in future posts, my awesome team has made these features work better than I ever dreamed possible.
Most importantly, by the second unit in our course, the peer endorsement feature appears to be 100% effective. Not a single student with a completed wikifolio requested an instructor endorsement, while every peer-endorsed wikifolio appears to be reasonably "complete." Most importantly, there are about a dozen incomplete wikis that were never endorsed. This all means that the system is working and we have made a lot of progress on techniques for scaling up artifact-driven open courses. As I will elaborate on later, students are getting tons of personalized feedback while we avoid the unsustainable (and community killing) individualized feedback.
While our course is certainly not massive (currently around 100 active students), there is a relatively massive amount of both individual and social learning taking place. Most importantly, I think we have solid evidence supporting the reasoning behind the BOOC: we need interim course contexts to support design-based research to scale up interactive and participatory learning strategies that have first been refined in smaller course contexts. I learned a lot as I migrated my features, first from one course to another course in the Sakai platform, then to Google Sites, and then to Course Builder. In particular, we have learned how to build in learning analytics that automate features in Course Builder that would have seemed impossible to scale up in Sakai, Canvas, or Blackboard.
Our goal is that our interactive features will be incorporated into the new mooc.org platform that is coming out of the recently announced partnership between Google's Course Builder team and Open edX (the open-source side of edX). That platform and some more broadly popular topics should push our efforts into the massive realm (roughly at least 1000 participants completing the course).
BOOC versus iMOOC versus pMOOC?
But what would we call this new hybrid MOOC format? I was thinking about iMOOC for "interactive." But that does not highlight the highly disciplinary form of interaction that we are after, and overlaps with the highly interactive cMOOC format. It also is too close to "individualized" which is really what the xMOOCs currently are.
At the moment the best idea I have is pMOOC for "personalized." Our BOOC draws very directly from current situative theories of learning (particularly from Jim Greeno's work). These perspectives argue that one's knowledge of disciplinary concepts in a course is fundamentally defined by the personalized contexts that individuals bring with them. Contexts stop being a setting in which to "apply" abstract concepts to make sense of them. Rather the distinction between context and concept is essentially dissolved.
Whether or not the core assumptions of situativity are really true, we are finding value in the general design principle of Let concepts get their meaning from contexts. For example, students never want to critique or even discuss their classmates's perceived understanding of course concepts. But it is really easy to empathize with a student whose context is not well-suited to using the concepts. So this creates natural personalized interactions about context-concept relationships. This interactions promise to leave all participants with the enduring knowledge needed to use those concepts appropriately in subsequent contexts.
I would love to hear what people think. And I really hope the folks at Course Builder/Open edX will be able to build in some of these features into the new mooc.org platform.