A real-time chronicle of a seasoned professor embarking on his first massively open online course.
From an educational perspective, my goal in offering a MOOC on mathematical thinking is very modest. I have not approached the task as one of developing a whole new pedagogic model. That is a future goal — for me or for others. Rather I set out to see how much we can take current university teaching (of transition mathematics material) and make it available to a wide audience. Indeed, almost all the Stanford MOOCs currently being offered are free, online versions of regular Stanford courses, in many cases running concurrently with a physical class on campus. (As I noted in an earlier post, the technology that supports these MOOCs was actually developed at Stanford in order to facilitate flipped-classroom learning in on-campus classes.)
The underlying assumption of university education — at least at major research universities (as Stanford is) — is that the principle value for the student comes from studying with a world expert in a particular domain. Though many professors at research universities do in fact put enormous effort into their teaching, what is really being offered (sold) to students is the expertise (and reputations) of the faculty. (Other parts of the value proposition, such as the prestige of the university, stem from the faculty, both past and present.) It’s a method that works well for very bright, well-prepared, and highly motivated students, but it is not ideal for everyone.
In fact, even at less prestigious universities, where there are fewer leading research faculty, and at liberal arts colleges, where the primary focus is on undergraduate education, field-content knowledge hugely outweighs pedagogical content knowledge — how to teach the subject and how students learn it. (A Ph.D. is usually required for a faculty position.) That makes universities and colleges very different from high schools.
One of the implicit purposes of a math transition course, such as mine (as well as many other first-year courses in different disciplines), is to help incoming students adjust to the different approach to teaching. More precisely, it is to help them adjust to not being “taught”, but having someone help them learn. This is particularly significant in mathematics — at least in the US — because of the hugely formulaic, procedures-focused nature of K-12 mathematics education in this country.
My challenge then, like that facing most of my colleagues offering their first MOOC, is to figure out how to take an existing educational model, hitherto used to teach (or help to learn) twenty-five or so students in a classroom, and make it available to thousands, spread around the world.
Since my topic is mathematical thinking, the biggest, and most obvious challenge is how to compensate for the complete absence of regular interaction between the students and me, the instructor. Sure, I give lectures when I teach a physical transition class, but the lectures are one of the least significant components. They really just set the agenda for learning. In order to help the students develop the ability for mathematical thinking, I need to see them in action at the board, to read their work, and to discuss their attempts face-to-face. Learning to think mathematically is more like learning to drive or to play tennis than soaking up knowledge. You have to do it alongside an expert or coach.
It’s a challenge I think cannot be completely overcome in a MOOC. The question is, is it possible to get part-way there? I suspect it is, but we’ll only find out for sure by making the attempt. So here we are.
One thing a MOOC does offer that is not possible in a physical class — and hence is a plus — is that all the instruction and professorial-learning-assistance can be on a one-to-one basis. Sure, it’s all one way, but if you set it up right (and if your voice/personality/whatever work over an ethernet cable), then the student can get that sense of working alongside the instructor — the expert.
Though by no means the first to discover that, Salman Khan, by virtue of his huge following at Khan Academy, demonstrated just how powerful is that sense of “working together, side-by-side”. Though I share the dismay of many of my colleagues at his less-than-expert content knowledge and his almost non-existent pedagogical content knowledge (neither of which he could be expected to have, given his background), where I seem to part company with many of them is the huge significance I attach to the way he pulls off that human-connect. For online learning, I suspect it trumps almost all other factors.
(BTW, in developing my MOOC, I soon lost track of the number of times I made a decision based on a “suspicion” — or a “guess” or “hunch”. MOOCs are generating enough research questions to sustain several generations of doctoral dissertations in education research.)
Based on that suspicion (admittedly a suspicion comfortingly buttressed by a Khan Academy user base that numbers in the millions), Khan’s format was my starting point, as I observed in my last post. Not just the physical aspect of “sitting alongside in a one-on-one tutorial” but the associated human connect (and with it reassurance and encouragement) that Khan delivers.
In Khan’s case, his now widely familiar format originated with him informally helping his school-age relatives (who lived a long way away) with their math homework. What the viewer gets on their computer screen is, well, just “Uncle Sal”, doing what he would have done if he were really sitting alongside one of his relatives. For my MOOC, I wanted to achieve a similar outcome. Not a slick show, not a polished, rehearsed performance. Just me doing math.
Of course, the logistics of putting together a complete course that has to run automatically, and be scalable to many thousands of students around the world, many of them not native English speakers, meant that there had to be a lot of detailed advanced planning. Everything had to be scripted. But when it comes to the bits where I explain some mathematics, I put the script to one side and just start to work through the material as if I am sitting next to a student.
You might not like it. It might not work for you. You will surely despair at my handwriting. You might hate my accent. (I did cut down drastically on my jokes and puns, in deference to a multilingual audience.) But as far as I can make it, absent being physically in the same room, it’s what you would get if you were taking the course with me here at Stanford. [Some time spent in a campus video-editing studio made my into-camera segments look a lot smoother than they were when we recorded them! If it’s digital, it’s plastic. But the goal there was to reduce the length of those segments.]
Which brings me back to my starting point: seeing the extent to which we can take existing university education and make it available to the world.
Once we can do that — and it will surely take several iterations to iron out all the kinks and make an altogether better job of it — we can look at how to change the underlying model. In addition to MOOCs making accessible to the world some aspects of university education, I think that the act of designing them, mounting them, and analyzing the results, will lead to changes in the way we organize learning within our universities.
It is because the current goal is to see how well we can deliver (current) real university education to the world for free that most of the MOOCs being offered have an unpolished, unrehearsed look. By deliberate choice, to the greatest degree we can achieve, what you see is what our (on-campus) students get. (I think this WYSIWOSG philosophy — I just made up that term — is also one of the reasons for the success of Salman Khan — including the fact that in his case, unlike university MOOCs, he does not even lesson-plan his instruction sessions.)
So much for the most visible part of the MOOC: the instruction. But instruction is still just instruction. As I’ve said before, the learning takes place elsewhere, through other mechanisms, none of which we understand very well. So where is that educational meat?
Now we are about to really enter speculative territory.
To be continued …
COMMENTS: As always, comments are welcome, provided they remain on topic.
MOOCtalk 
Hi Keith,
I am teaching a transition course this semester (today is our first day!) and am excited about this experiment you’re doing. Since my classes will have already been meeting for a month before yours starts, I’ve been wondering if I could offer my students some form of extra credit if they participate in your course in a meaningful way, by helping others out in some capacity with material they’ve already seen. Is this feasible with the format your course will take? Would I be able to easily see what my students have done, by searching for their names to find their contributions or something along those lines? Regardless, I’m looking forward to following this blog and the course as it develops.
Molly, I’d love to have your students participate in the way you suggest. In fact, I believe that kind of participation is the key to making a mathematics transition course (or indeed any math course that involves reasoning and proofs) work. Through your students, the MOOC students around the world without access to good higher education (or any higher education at all) will get the benefit of your knowledge and expertise, and the same is true for the students of any other instructors out there who will be giving transition courses this fall. I don’t know of an obvious mechanism for you to monitor your students participation, other than for them to let you know their public identifiers so you can read what they say on the course forums, where they will be flagged as “Tutors”. But I’ll check with the folks at Coursera who are building the platform and see what they say. It’s possible that they could build something in for a future iteration of a course such as mine. I look forward to welcoming your students online starting September 17.
Molly, a follow up to my previous reply. Coursera says they will be able to provide a mechanism to report on forum participation some time in the future, but for this fall any instructor who wants to give their students extra credit for helping others in a MOOC like mine will have to arrange a mechanism with their class to meet their needs — perhaps having students who act as tutors keep a record of their contributions and self-report. Simply taking screen shots of forum posts might be enough.
Thanks for looking into this for me. Having students self-report is a good idea for now, I’ll think about how to implement it with my class.
There was a recent New Yorker article on Stanford and what the on-campus value proposition is (http://www.newyorker.com/reporting/2012/04/30/120430fa_fact_auletta). It is networking for job opportunities, both between students based on their class and connections, and between faculty with their industry connections and students. The MOOC effectively gives the intellectual content but not the corporate seal of approval (“I am a Stanford graduate”) or the career network.
Yes, I saw that article. I think there are other reasons why online learning won’t come close to replacing living-on-campus learning besides the ones you cite. (Day students miss a lot as well.) It’s the “a MOOC is my only option” aspect that sends shivers up my spine. And for some, a MOOC could be the gateway to more. Udacity already passes details of good students to recruiters at top tech companies. And I can imagine Stanford student recruiting having an interest in outstanding students in a MOOC.
Don’t get me wrong. I’m enrolled in your course! I am 53 years old and I have a PhD in Computer Science. I probably don’t need the course but I haven’t been in college for 32 years and I very much appreciate the opportunity to hear the same story twice or hear a new story. I took the Vaccines course at Coursera and I’m also enrolled in the Machine Intelligence course and a number of other upcoming courses. I think Coursera is really great and it really builds on what MIT started with Open Courseware. The fact that you’ve got so many new universities signing on means that individual professors also love the exposure and the chance to try something new.
But I also work for a big company and I see what they do when they hire young people. They make lists of Tier 1 and Tier 2 (no Tier 3) institutions, and focus 99% of their effort on Tier 1. Tier 1 is predictable and includes Stanford. It’s a branding thing and Stanford has a good brand. That brand is worth $50K a year for those that can afford it. For those that can’t for whatever reason (too old, too distant, too poor), the MOOC is truly a wonderful thing, and the training it gives is intrinsic and can help someone who doesn’t have the connections to acquire skills that will allow them to make their own connections and break through the lazy hiring practices of places with their Tier 1 and Tier 2 lists.
Yes, and program managers and HR folks at those big companies will tell you over a drink that the quality of the school is not a good predictor of a good employee — it’s just a convenient filter, and leads to a “safe” (for the recruiter) hiring decision, not unlike the old “no one got fired for buying IBM”. With some controls, the Big Data aspect of MOOCs might disrupt that hiring practice. Thanks for writing.
I agree with you about the human connection element being absolutely essential. It’s why I’m loyal to Khan Academy, despite sharing your doubts about Sal’s pedagogy and, to a lesser degree, content knowledge. I started playing around on Khan Academy because, with a B.A. in English and a single 3-hour Trig course to my name, I felt mathematically illiterate. Eight months later I’m still visiting KA every day and I’m looking forward to trying my hand at your Mathematical Thinking course.
In my opinion, many of the pedagogical criticisms of Sal Khan’s videos completely miss the point. I’ve seen teachers and professors post their own videos, allegedly improving on his by employing sounder teaching methods and more consistent explanations of the content. This sounds great–except for the fact that I was completely turned off by them, either because the instructors flat-out lacked charisma, or because the presentation lacked a certain something.
For instance, one of the videos I watched had text that just popped into existence. There was no human hand writing it; it just appeared. I hadn’t realized until that moment how much the act of watching Sal write things out (somewhat laboriously at times) affected my engagement with the material. When the text in the “improved” video popped into being, my brain stuttered in a way it never does with Sal’s videos. What I find valuable about KA videos is not the procedural instruction, but the opportunity to observe how someone I can relate to approaches an (unscripted) problem or calculation.
I appreciate that you’ve decided to focus (for the moment) on how to translate current university education into MOOC form, rather than trying to revolutionize education in one fell swoop. I do think that MOOCs (or at least online education–perhaps not in MOOC form) has huge potential as well as some crippling drawbacks, but I think perfecting it will be an iterative process (something many of the critics should keep in mind, if you ask me–things don’t have to be perfect to be valuable, especially for those with limited options).
Anyway, I look forward to your class and to seeing how the whole e-learning experiment evolves.
Jennifer, I agree with everything you say — particularly your last paragraph, albeit from a different perspective on this occasion! 🙂 Glad to have you on board.
I translated to my native language MOOCs Why Look Unprofessional: MOOC planning – and I am amazed open thinking Dr Devlin- . No doubt we will all learn a lot from this eminence. Thanks for allowing us to be in class and be part of the 33 600 students (I think we will be more) Congratulations!!!
Hi Keith,
I know of the famous Stanford humour from Mehran Saham in his videos about Java. Sticking to this example, the worst thing about the lectures were the custom Stanford libraries used in the programs while teaching general OOP.
To your advantage, maths is universal and should be the same anywhere. I am really looking forward to the course, its between my other courses and overlaps a little at the end but that’s OK.
I read a lot of books about maths at the moment to get up to speed after all the programming. The book ‘Introduction to Mathematical Thinking’ should be available as Kindle edition 😉
Marcus
Marcus, As I’ve written in my blog, I think MOOCs are “e-books on steroids”. So that leaves little role for a Kindle edition. Hope you enjoy the course.
What math would I need to know to follow your program? I’m 17 and my goal is to study math in Germany as an international student. But I’m scared that my knowledge in math might not be enough, which is why I’m following your course to see if I can keep up.
Zander, You really won’t need to know much mathematics at all. It’s about learning to think about things a whole new way. Viel Glück!
I’m very curious on how your course will change my thinking in mathematics.
I’ve been a medical specialist all my life and started computing a long time ago. The first thing I did was programming in Pascal. What struck me was the very precise definition of what you wanted to use in the programming. That forced me to think about what I reely needed and how to use it Since that time I was more and more thinking in algorithms and realized more than ever that a sound base was needed. I qualified this as scientific thinking. The only problem, if you see it that way, is that you always end up with a question, or a bunch of questions. I really can’t wait to start with your class. 😉
blutch1, I know how the course is intended to change your thinking; for most students of this material, the big question is whether it will. Many don’t make it – they discover that mathematics is not for them. (And among those who give up are many who did well in math at high school.) And you are right in supposing it’s more about raising and analyzing questions than finding answers. Good luck.