A real-time chronicle of a seasoned professor embarking on his first massively open online course.
A few days ago, I went into our campus TV studio with the two course assistants for my upcoming MOOC, to record a short video introducing them to the students. The students will see a lot of me, but my two TAs will be working behind the scenes, and the students will encounter them only through their contributions to the forum discussions. The videos were intended to compensate for that lack of human contact.
During the course of recording that video, the three of us got into a discussion about our backgrounds, our motives in giving the MOOC, and our views on mathematics, science, education, and our expectations for the MOOC format. The camera was rolling all the time, and we were able to select a few parts of that discussion and create a second video that I think will help our students understand some of our thinking in putting this course together. I posted copies of both videos on YouTube. (They are much lower resolution than the videos the registered students will see on the course website when it goes live on September 17 — the “first day of classes”.) I think the two videos provide an insight into our thinking as we designed this course.
The fact that the current round of MOOCs have a “first day of class” at all has been a matter of some debate. The C in MOOC stands for “course”, but is this the best way to go? For example, see this blogpost from a graduate student at Berkeley, who argues for a more open framework of learning resources. He makes some good points that all of us involved in this initiative have thought about and discussed, but I’m not sure the kind of thing he advocates can work for disciplines and subjects that depend heavily on student-faculty and student-student interaction, as mine does.
In fact, I’m not sure the MOOC will work sufficiently well at all in such cases; this is very much an experiment that I anticipate will continue for several years before we get good answers either way. For the first iteration, it makes sense to start with a model we know does work. And important (we think!) elements of that model are, to repeat Sebastian Thrun’s list, as quoted in the Berkeley student’s blog: admissions, lectures, peer interaction, professor interaction, problem-solving, assignments, exams, deadlines, and certification. To use the mnemonic I coined earlier in this series, our basic design principle is WYSIWOSG: What You See Is What Our Students Get.
Since these courses are free, we can, of course, do a lot of A/B testing in future years, to see which of these truly are crucial, which can be changed and how, and which can be dropped. I suspect the answers we get will vary from discipline to discipline, and possibly from course to course.
All of us involved in this MOOC movement are trying to find out the best way that works for our particular discipline and is consistent with our own style as instructors. As I indicated in Part 4 of this diary, I think it makes sense to begin by trying to implement in a MOOC as much of our tried-and-trusted classroom-based teaching as we can (as Thrun did with Udacity), and then iterating in the light of what we learn.
This is why, instead of hiring a mathematics graduate student to TA my course, which is what I would have done for an on campus class, I brought onto my team two graduate students from Stanford’s School of Education with several years of experience in learning design and the use of technology in education. In addition to helping me with the design and running of the course, they will conduct research into the course’s efficacy and try to understand how learning occurs in a MOOC. (Other than a brief, non-compulsory questionnaire at the start and finish of the course, all their research will be based on data gathered on the Coursera course platform and human monitoring of the forum discussions. One huge benefit of MOOCs is that they facilitate Big Data research.)
It’s live beta, folks.
To be continued …
MOOCtalk 
Hi Keith, great post. A couple of months ago I wrote that post about problems with MOOC and I think these platforms have become much better since then. I agree with you that it is great for the first or second iteration. And definitely appreciate that you are putting your course online for others to learn. Eager to see how MOOC systems evolve in the near future.
Cheers
Mark
WRT whether or not MOOCs should have a start date and should dispense content on a schedule, I am actually leaning toward “yes,” which is honestly not the opinion I expected to develop; I was originally in favor of all access all the time.
The reason for my change of tune: promoting patient problem solving (or, perhaps more accurately, thwarting my lack of self-control). I’ve sampled widely from the range of available MOOCs, including some on Udacity. While I am really enjoying the courses there and find them valuable, I am frustrated with my experience around problem sets; I find that I don’t grapple with them long enough before progressing to see how others solved them.
I know from past experience that the feeling of being stuck and frustrated and making no progress is essential to problem-solving (obviously; if the solutions were readily apparent they wouldn’t be very interesting problems), and that if I think about something for a few days while I’m washing the dishes or whatever, I’ll probably eventually have an epiphany, or at least come up with a creative new way to fail.
But when course content and solutions are all at my fingertips 24/7, I find myself giving in to the “stuck” feeling and deciding that I’m never going to get it anyway, so I might as well see how the professor solved it. It’s incredibly difficult (at least for me) to force myself to “waste” several days in which I could be progressing in the course on what begins to feel like the futile pursuit of a solution.
Another trade-off I’ve come to regret more and more is the loss of the community that comes with everyone grappling with the same material at roughly the same time. It’s much easier to get and give help and participate in forums when the discussions are focused around material that everyone’s been introduced to and everyone is thinking about.
In self-paced courses, I rarely follow wiki or forum feeds because the volume of information, most of it revolving around concepts I haven’t yet encountered and can’t discuss, makes it easier to learn mostly in isolation and only to visit the community in times of need rather than immersing myself in it. This is very different from my experiences with classroom courses using Piazza or with more structured MOOCs, in which I tend to be an active participant (as you can probably tell from this comment, I like to talk).
Jennifer, thanks for writing. That’s pretty well my view. Having taught transition courses on and off for most of my career, I think the need for students to work together is huge – for psychological support as much as anything. It’s really hard for a smart person to feel stuck on something that they sense is not really hard, if only they knew how to go about it. But that’s how most students feel most of the time in a transition course. It’s even worse if they go to a university that does not have such a course, and just get thrown into first year real analysis, which is what happened to me. The big question, is how well this mutual support and collaboration will work over social media. My team and i have spent a fair amount of time thinking about this — and I spent a lot of time looking at the issue before deciding to offer a MOOC transition course. I am moderately optimistic it can be made to work, at least for some people, but the chances of it working well the first time are pretty low. If you don’t mind, I’d like to have the option to refer the students in my MOOC to your comment at some stage, since you articulate well the feeling most are likely to have. –KD
Please feel free to use my comment however/whenever you like!
From my limited experience with Coursera, the start dates build anticipation but it is a big disappointment to see a class that I can’t get “credit” for because I am two months late. However, my learning style matches up with with the current style of teaching (just barely) so I would be interested to see what type of people would be motivated to succeed in a different style course.
JohmgSchroeder, You can still register for my course. I am not sure why you were unable to do so. The course does not start until September 17 and people are registering every day. If you continue to have trouble, you should contact Coursera directly. None of the Stanford MOOCs give Stanford credit; we all award Certificates of Completion (in my case with a possibility of Completion with Distinction). I hope this helps. I’d be glad to have you on board.
Sorry for the confusion; I was referring to a sci/fi literature class on Coursera that I missed because it started in July.
In the third paragraph of this post you talk about “first day of class” and I was just commenting that an outcome of strict start dates is that people who find the class late will be left out but people who find the class early will build anticipation. I am signed up for your class and counting off the days until the 17th. I didn’t mean to ask about the course and I know that there are more appropriate places to do so.
Got it. BTW, your question “I would be interested to see what type of people would be motivated to succeed in a different style course” is one of the questions my two TAs (graduate students from the School of Education) are investigating with my MOOC.
Hello Mr.Keith,
The introduction video motivates “anyone above the age of 17 could benefit from this course”. Is this the right age for mathematical thinking or it could be introduced earlier (I am close to 30)? l
Looking forward to an interesting class