Paul Stacey

The Pedagogy of MOOCs

There is a great deal of energy, enthusiasm, and change happening in today’s education sector. Existing and new education providers are leveraging the Internet, ICT infrastructure, digital content, open licensing, social networking, and interaction to create new forms of education. Open Educational Resources (OER) (including open textbooks), Open Access, and Massive Open Online Courses (MOOCs) have all gained traction as significant drivers of education innovation.

MOOCs in particular are stimulating widespread discussion around the potential to reach and serve hundreds of thousands of learners who would otherwise not have access to education. Like all of you I’ve been tracking MOOC’s with great interest.

While MOOC’s have attracted huge attention, and hype, for supporting massive enrollments and for being free its the pedagogical aspects of MOOC’s that interest me the most.

The challenge is this – How can you effectively teach thousands of students simultaneously? I’m fascinated by the contrast between post-secondary faculty and K-12 teacher contract agreements that limit class size and the current emergent MOOC aim of having as many enrollments as possible. What a dichotomy.

MOOC’s have done a great job at creating courses open to massive enrollments from anywhere around the world. But how well are MOOC’s doing at actually successfully teaching those students? Based on MOOCs equally massive dropout rates having teaching and learning success on a massive scale will require pedagogical innovation. It’s this innovation, more than massive enrollments or free that I think make MOOC’s important. Let me explain.

MOOC’s originated in Canada and I’ve been fortunate to have followed and experienced the early pioneering work of people like Stephen Downes, Alec Couros, Dave Cormier, and George Siemens. In 2007 there was Social Media & Open Education, in 2008 & 2009 Connectivism, in 2010 Personal Learning Environments Networks and Knowledge, in 2011 Learning and Knowledge Analytics which we hosted in the BCcampus SCoPE online community. For a more complete listing see Stephen Downes Partial History of MOOC’s.


All of these early MOOC’s were open to anyone to participate. Some of these early MOOC’s, taught by university faculty, had tuition paying students taking the course for university credit who were joined in the the same class with non-tuition paying, non-credit students who got to fully participate in a variety of non-formal ways. Alec Couros pedagogically designed his graduate course in a way that relies on the participation of non-credit students. Other early MOOC’s were solely offered as a form of informal learning open to anyone for free without a for-credit component.

Alec Couros produced a YouTube trailer for his Social Media & Open Education course that conveys a bit of the creative fun associated with these early MOOC’s.

The most fascinating aspect of these early MOOC’s was the pedagogical approach. Dave Cormier in this YouTube video maps out the five steps to success in a MOOC – 1. Orient, 2. Declare, 3. Network, 4. Cluster, 5. Focus.

He goes on to explain in this Knowledge in a MOOC YouTube video that a MOOC is just a catalyst for knowledge and that knowledge in a MOOC is emergent.

The “How this course works section” of the Personal Learning Environments Networks and Knowledge MOOC provided participants with the following:

PLENK2010 is an unusual course. It does not consist of a body of content you are supposed to remember. Rather, the learning in the course results from the activities you undertake, and will be different for each person.

In addition, this course is not conducted in a single place or environment. It is distributed across the web. We will provide some facilities. But we expect your activities to take place all over the internet. We will ask you to visit other people’s web pages, and even to create some of your own.

This type of course is called a ‘connectivist’ course and is based on four major types of activity.

The four types of activity are described as; 1. Aggregate, 2. Remix, 3. Repurpose, 4. Feed Forward.
I encourage you to read the full description here.

In those early pioneering days MOOC’s were exciting for their pedagogy!
Even the courses were about innovative pedagogy – Social Media & Open Education, Connectivism, Personal Learning Environments, Learning Analytics.

In 2011 MOOC’s migrated to the US with Jim Groom’s DS106 Digital Storytelling at the University of Mary Washington in Virginia. DS106 is a credit course at UMW, but you can also be an “open participant“. As described in About ds106 you can “join in whenever you like and leave whenever you need. This course is free to anyone who wants to take it, and the only requirements are a real computer, a hardy internet connection, preferably a domain of your own and some commodity web hosting, and all the creativity you can muster.”


DS106 took MOOC’s in new pedagogical directions.

DS106 has a highly innovative pedagogical approach to assignments. Rather than confidential, secret assignments created by faculty, ds106 course assignments are collectively created by course participants over all offerings of the course and are posted online in an Assignment Bank anyone can access.


Rather than specifying assignments everyone must do, participants choose from ones in the Assignment Bank. Each assignment has a rated difficulty of 1 to 5 stars and a particular assignment for a course topic might require students to complete 10 stars worth of say mashup assignments, or design assignments, or audio assignments – there are ten different assignment categories. This model of having course participants collectively build the course assignments which are then used by students in future classes is a hugely significant pedagogical innovation.

I’ll always remember ds106 as the first ever online course with its own radio station ds106 radio. I was totally impressed by the enthusiasm, connection, and devotion the radio station generated in course participants. The pedagogical potential of a course radio station is an exciting but relatively unexplored opportunity. I’ve been disappointed to not see this innovation adopted by other initiatives.


The next big step for MOOC’s came in the fall of 2011 when Stanford Engineering professors offered three of the school’s most popular computer science courses for free online as MOOC’s – Machine Learning, Introduction to Artificial Intelligence, and Introduction to Databases. The Introduction to Artificial Intelligence course offered free and online to students worldwide from October 10th to December 18th 2011 was the biggest surprise. Taught by Sebastian Thrun and Peter Norvig this course really was massive attracting 160,000 students from over 190 countries.


Pedagogically though these MOOC’s from Stanford were a step backward. The teaching and learning experience was comprised of watching video lecture recordings, reading course materials, completing assignments and taking quizzes and an exam. Gone were the rich pedagogical innovations from the earlier MOOC’s. Instead these MOOC’s simply migrated campus-based didatic methods of teaching to the online environment. Most disappointing of all was the absence of any effort to utilize the rich body of research that had already been done on how to teach online effectively.

While didactic, lecture based methods of teaching have long been the mainstay of bricks and mortar schools we know that this method of teaching does not transfer well to online. For this reason alone I’m not surprised MOOC’s have high drop out rates.

Sebastian Thrun’s experience teaching the Stanford Artificial Intelligence MOOC was so compelling that he left Stanford and raised venture capital to launch Udacity with a mission to change the future of education by making high-quality classes affordable and accessible for students across the globe.


The Udacity FAQ provides some explanation of the pedagogy. Udacity courses include lecture videos, quizzes and homework assignments. Multiple short video sections make up each course unit. Each video is roughly five minutes or less, giving you the chance to learn piece by piece and re-watch short lesson portions. All Udacity courses are made up of distinct units. Each unit is designed to provide a week’s worth of instruction and homework. However, since Udacity enrollment is open, you can take as long as you want to complete Udacity courses. Udacity courses include discussion forums and a wiki for course notes, additional explanations, examples and extra materials. Each course has an area where instructors can make comments but the pedagogical emphasis is on self-study.

In a nod to the importance of discussion in online learning Udacity courses do have discussion forums where students can post any ideas and thoughts they have about the course, ask questions, and receive feedback from other students. Forum posts can be “up-voted” by other students simply by clicking a thumbs up button on the left side of the post. When a question or answer is up-voted, the student who posted it gains “karma points”. These points serve as a rough measure of the community’s trust in him/her. On the other hand, if a post is misleading, it will be down-voted. Various moderation tasks are gradually assigned to students based on those points. The inclusion of discussion forums is a definite plus for Udacity MOOC’s as they provide a means for students to help and learn from each other. Udacity should be encouraged to move beyond using discussion simply as an informal system of help to being a key means by which learning occurs and more tightly integrate discussion into the content and assignments.

Further indication of the importance of enhancing MOOC’s with social learning can be seen by the fact that Udacity students have begun to organize their own meetups where local students physically meet to study together, ask questions, and share ideas. Udacity has put in place a community site to help students do this and 3,199 students have formed study groups across 520 cities.


In late December 2011 MIT announced edX with the aim of letting thousands of online learners take laboratory-intensive courses, while assessing their ability to work through complex problems, complete projects, and write assignments. As with other MOOC style offerings students won’t have interaction with faculty or earn credit toward an MIT degree. However, for a small fee students can take an assessment which, if successfully completed, will provide them with a certificate from edX.


Pedagogically I find edX odd. First their primary goal as stated in their FAQ is to improve teaching and learning on campus. Say what? You want to do a MOOC that teaches tens of thousands of students online in order to improve teaching on campus?

Second edX describes one of its distinguishing features as supporting faculty in conducting significant research on how students learn. There is no mention of applying research coming out of online learning to edX. Its almost as if online learning has yet to be invented. This makes it seem that the edX MOOC students are merely guinea pigs whose learning data will be collected by faculty as research data and used to benefit and improve the learning experience of tuition paying on-campus students.

A third edX oddity is that it isn’t trying to levearge MIT’s own OpenCourseWare materials by combining them with innovative online learning pedagogies for use as MOOC’s. Its almost like MIT edX and MIT OCW are from completely different institutions that have nothing to do with each other.

The focus of edX so far seems primarily to be not on pedagogy but on engineering an open source MOOC platform. In a strange twist edX announced in April 2013 that it has merged its own software development efforts with Class2Go an open source MOOC platform developed by eight engineers in Stanford’s computer science department. I think its great that this platform will be open source but I’m mostly interested in the extent to which the tool builds in support for effective, scaleable online learning pedagogies.


Class2Go supports videos from professors with in-video quizzes, formative and summative exercises, and frame extracting that lets students jump to a specific part of a video. Does this sound like pedagogical innovation to you? The only concession I see to the rich research of online learning is inclusion of discussion forums to “get people talking.” I wish edX was a little less engineering centric and way more pedagogically centric.

Coursera founded by computer science professors Andrew Ng and Daphne Koller from Stanford University launched in April 2012 as an educational technology company offering massive open online courses (MOOCs). Shortly after launch Coursera was working with Stanford University, the University of Michigan, Princeton, and the University of Pennsylvania. By February 2013 Coursera had over 69 university partners and was offering courses in Chinese, Italian, and Spanish.


Coursera is one of the few MOOC’s that actually describes its pedagogical foundations.

Coursera pedagogy involves video lectures, mastery learning, and peer assessment. Coursera is providing its university partners with a flipped classroom opportunity whereby the lecture, course reading, and to some extent assessment and peer-to-peer interaction for campus-based tuition paying students are handled in the MOOC with on-campus activities focused more on active learning. However, for Coursera MOOC participants who are not tuition paying campus-based students there is no active learning component. Although once again students are tossed a tidbit of social learning in the form of discussion forums. Lo and behold this actually improves learning as Clint Lalonde points out in “Online interaction improves student performance. Gee, imagine that.”

All of these new MOOC’s are focused on objectivist and behaviourist methods of teaching and learning. Their pedagogy is based on an assumption that when there are tens of thousands of learners social learning isn’t feasible. So instead of interaction with a person these MOOC’s focus on replacing the human social component of learning with a kind of artificial intelligence interaction with the platform. Coursera holds this up as good practice by noting, “Even within our videos, there are multiple opportunities for interactions: the video frequently stops, and students are asked to answer a simple question to test whether they are tracking the material.” Designing MOOC pedagogies based on what some are calling robot marking jeopardizes quality, learning outcomes, and ignores best practices in online learning.

I’m not the only one who thinks MOOC pedagogy needs work, see:
MOOCs and the Quality Question
What’s right and what’s wrong about Coursera-style MOOCs
MOOC pedagogy: the challenges of developing for Coursera

Students tend to find online behaviourist and objectivist learning pedagogies boring, impersonal, and not interactive or engaging. But those of us who have been working in the field and taken exemplary online learning courses know that in fact online learning pedagogies can be incredibly social even more so than campus-based courses. It is relatively easy to instructionally design online learning so that every student engages in deep discourse.

Early MOOC’s and exemplary online learning pedagogies recognize and utilize the breadth of knowledge and experience students participating in the course have. The magic of online learning happens when extensive effort is made to tap into student expertise through blogs, chat, discussion forums, wikis, and group assignments. Socio-constructivist and connectivist learning theories acknowledge and embrace the social nature of learning. Learning is not just acquiring a body of knowledge and skills. Learning happens through relationships. The best online pedagogies are those that use the open web and relationship to mine veins of knowledge, expertise, and connections between students, between students and the instructor, and between students and others on the open web.

The big new MOOC’s also seem to be ignoring Open Educational Resources (OER) and the incredible pedagogical affordances openly licensing course content brings. Many of the early MOOC’s were not just open in terms of enrollment they were open in terms of utilizing the open web, utilizing open content, and making continuous improvement of courses an integral part of the teaching and learning experience. The new MOOC’s seem intent on enclosing students in a closed environment that is locked down and DRM’ed in a proprietary way. See Coursera, Chegg, and the Education Enclosure Movement for a good description of this direction.

Like many of my Canadian brethren I mourn the loss of early MOOC pedagogical innovations and find diagrams like this that purport to show Major Players in the MOOC Universe a form of colonialism that attempts to rewrite MOOC history.

However, I do see MOOC’s as a major innovation and hold out hope that other MOOC providers will differentiate themselves by being open and by fully utilizing social learning.

NovoEd has caught my eye. New MOOC Provider Says It Fosters Peer Interaction

And OpenupEd the first Europe-wide MOOC initiative, launched with the support of the European Commission. OpenupEd is committed to Digital Openness using open source software, open access, and open educational resources for their MOOC’s.

Let me end with my own pedagogical recommendations for MOOC’s:

  • Be as open as possible. Go beyond open enrollments and use open pedagogies that leverage the entire web not just the specific content in the MOOC platform. As part of your open pedagogy strategy use OER and openly license your resources using Creative Commons licenses in a way that allows reuse, revision, remix, and redistribution. Make your MOOC platform open source software. Publish the learning analytics data you collect as open data using a CC0 license.
  • Use tried and proven modern online learning pedagogies not campus classroom-based didactic learning pedagogies which we know are ill-suited to online learning.
  • Use peer-to-peer pedagogies over self study. We know this improves learning outcomes. The cost of enabling a network of peers is the same as that of networking content – essentially zero.
  • Use social learning including blogs, chat, discussion forums, wikis, and group assignments.
  • Leverage massive participation – have all students contribute something that adds to or improves the course overall.