Tuesday, June 18, 2013

A Response to Anant Agarwal's Call for Teachers to Join the Revolution

This past weekend Anant Agarwal, the president of edX, published an op-ed in The Observer titled "Online Universities: it's time for teachers to join the revolution."   There is much in his rather self-serving op-ed to criticize (and parody, as Jonathan Rhees has done so brilliantly), but perhaps the most important thing to point out is the false premise of his title.  Current teachers are cast as Luddite resisters who have dug in their heels and are opposing something that is improving education.  This is a complete misrepresentation of practicing educators at all levels, many of whom regularly use a range of education technologies as well as tried and true pedagogical methods to produce learning in their students.  In fact, education technology has been a presence in K-12 as well as post-secondary classrooms for at least a decade, and has expanded tremendously, especially on college campuses, thanks to the availability of fast broadband connections.

To suggest that the pre-MOOC university professor was teaching with chalk and yellowed notes is an absurd caricature whose sole purpose is to position MOOCs as the shiny new thing that will revolutionize education.  Agarwal declares that the "days of the old ways of teaching are numbered".  While there are certainly examples of "the old way of teaching" on every college campus, I suspect that they are vastly outnumbered by the many innovative, technologically savvy, and dedicated faculty who are working every day to improve student learning.

MOOC supporters repeatedly point out that the platforms and price point (free) improves accessibility.  I will leave it to others to explicate why this isn't quite true (and here).  Certainly, by putting courses on an open access platform, their is the potential to improve access.  But it is far from clear that MOOCs really are doing much beyond making lifelong learning more convenient; and targeting gifted learners in third world countries.

The most significant source of my irritation with the self-promoting MOOC rhetoric is centered on the claims that the mode improves learning--and, specifically, that it can support better learning than can a campus-based, large enrollment course.  In basic terms, there's no evidence to support this claim, at least not yet.  Furthermore, given the demographics of current MOOC users, it seems to me unlikely that even extensive study of  data from individual courses or users is going to be generalizable beyond a particular course or a small subset of users.   This is a big claim to make, and it requires clear data, not anecdote and declarations.

Working teachers need to know how particular students at particular universities in particular courses learn.  Much more useful would be funding projects to study currently enrolled students, especially a public universities.  As well, we need to know how best to help our beginning students learn HOW to learn, especially how to learn in a post-secondary setting where instructors aren't teaching to tests.  Since the majority of  current MOOCs users are degree holders and experienced learners--that is, they are people who already know how to navigate a course, study, be self-motivated, etc--the data being collected is unlikely to yield much of value.  This will change as more of the students are actual college students with all the complexities that college students present, when the data sets can be narrowed down to "current UT students taking X course" but that is years in the future.  As Rob Reich explains, "If MOOCs promise to enhance student learning, they must show that they deliver at least as powerful outcomes as traditional lecture classes in universities and community colleges.  If not, their virtue is their democratizing potential; but they will only be better than nothing."

Agarwal writes, "MOOCs are also improving the quality of education.  Online learning promotes active learning, where the learning watches videos and engages in interactive exercises."   I agree that MOOCs are improving the quality of education--but they aren't doing it alone and they aren't doing anything revolutionary.  They are simply hacking the flipped class model--and replacing the in class part of the formula with a (generally chaotic and ineffective) discussion board.  Furthermore, the notion that interspersing a few questions into a video creates an active learner is absurd.  It's better than simply lecturing at students; but it takes a lot more than a few questions in a video to create active learners.  Likewise, it takes a lot more than a few i>clicker questions in a lecture to truly engage students.  I've observed many a lecture class where students are shopping for clothes but still able to stop and click in on questions.

Indeed, much of what Agarwal has to say about how MOOCs facilitate learning imagines an ideal learner who uses the course precisely as it was designed to be used.  Yet teachers  know that this isn't how it works in real life.  It's absolutely true that videos allow for self-paced learning, but that doesn't mean students will use it that way.  In fact, in my own class, I found out that very few of them did.  Sure, in an ideal world, they would watch the videos carefully, engage with the embedded questions, and review confusing points.  The problem is, students rarely behave like ideal learners and so do not get the benefits that they could from the technology. We educators may understand that a key to learning is practice with retrieval; but students don't know that and will often find ways to avoid doing it unless their grade depends on it.  But again, MOOCs aren't the first to recognize the value of retrieval.  Faculty have been using LMSs like Blackboard to do practice quizzes and the like long before the advent of MOOCs.  As Glance et al. note in "The Pedagogical Foundations of Massive Open Online Courses," "MOOCs are in essence a restatement of online learning environments that have been in use for some time." 

I am similarly puzzled by the claim "students have always been critical of large lecture halls where they are talked at and declining lecture attendance is the result."  This is not at all my experience.  Quite the opposite.  Students enjoy a good lecture, one where they can scribble down the notes on the PPT, memorize them, and do well on exams.  It is unclear to me what the benchmark for declining student attendance is--this observation is meaningless without more context and data.  Student attendance has little to do with the quality of a course, in my experience, and much more to do with time of day, time of semester, whether they are commuting, etc.  In truth, one of the major challenges in moving from a lecture model to a model that emphasizes active learning is having to train students to be engaged learners.  Still, many complain that they would rather just be lectured to.  Yes, the ideal student would recognize the value of being an engaged learner; but most of us are not teaching these ideal students.  In fact, most of us have to spend a considerable amount of time and energy teaching our students who to be students, teaching them (and then repeatedly reinforcing) good learning strategies.

After declaring that the revolution has arrived, Agarwal turns to the flipped class model.  Indeed, it's become increasingly clear that the aim of the MOOCs vis-a-vis brick and mortar campuses, ultimately, is to provide content for flipped classes.  I've written elsewhere about why this is misguided and misleading.  In a nutshell, because the difficulty of a flipped class isn't moving content outside of class, it's doing everything inside of class.  Yet, by fetishizing content delivery and the professors who do it, the MOOC Incs have encouraged the belief that universities can save money by hiring non t-t faculty to run these flipped classes.  This is a recipe for disaster and demonstrates how little the MOOC Incs understand (or care about) the pedagogy of the  flipped class.

The op-ed concludes with the declaration that "it's time for teachers to rethink learning methods."  It is unfortunate that Agarwal doesn't realize (or won't admit)  that this was happening long before MOOCs came on the scene.  MOOCs are doing nothing new, apart from experimenting with large scale delivery.  Certainly, they have encouraged an engaged discussion about technology and pedagogy; but thus far, they are using methods that have long been used on brick and mortar campuses--indeed, if anything, MOOC pedagogy feels a bit retro to most of us (breaking up lecture with questions; discussion board; quizzes).  I can imagine a time when MOOCs might advance our understanding of student learning and make real contributions to pedagogical theory, but they aren't yet there.  The MOOC Incs would do well to recognize this point and refrain from perpetuating ridiculous and unfair caricatures of teachers. 


  1. I agree with much of what you say, Jen, especially the fallacy of basing the pedagogy on the behaviour of the ideal student. However, I think some of your rebuttal is based on the behaviour of the ideal instructor, or at least, the better-than-average instructor. In my experience, it's a minority, not "most of us" who "spend a considerable amount of time and energy teaching our students how to be students, teaching them (and then repeatedly reinforcing) good learning strategies." A majority of professors, I think, should sit up and listen to Agarwal's plea to "rethink learning methods." I'm not saying MOOCs are the answer. Instead, I'm happy to help these professors adopt research-based instructional strategies like flipped classrooms with peer instruction.

  2. I definitely think professors need to sit up and realize that the days of phoning it on are over. What's less clear to me is how many, really, are still doing that. I feel like I have such a limited view of all that happens even on my own big campus. I know my department is filled with very strong and thoughtful teachers; I am sure there are plenty of not very good teachers around, though. I guess what I'd really like to see, in place of these caricatures, is some actual data about what is happening on specific classrooms on specific campuses. Over the past 5 years, teaching on college campuses has received a lot of attention and support and has improved a lot. That is a piece of the narrative that is getting absolutely no press because it's not convenient to the MOOC narrative. What I'd prefer is, as you say, a call for professors to adopt research based instructional strategies like the flipped class. But also an acknowledgment from Agarwal and other leaders of the MOOC Incs that great progress is already underway; and more could be done if $$ and other resources was being focused on redesigning campus courses instead of developing MOOCs.

  3. You're right that our estimates of how many profs are "strong and thoughtful teachers" and how many are "phoning it in" are often anecdotal. Not always, though. Henderson and Dancy (2009) performed a nice survey of physics instructors. They found that while many had heard of research-based instructional strategies like peer instruction, they often implemented it with their own modifications (at the risk of lessening the impact) and then discontinued using it. So, on the one hand, damn. On the other, people are performing the surveys needed to give us data, not anecdote, about adoption of RBIS. My friends at UBC are doing something similar, to evaluate the impact of the Carl Wieman Science Education Initiative (cwsei.ubc.ca).

    Here's the Henderson and Dancy paper:


  4. Thanks for this reference. I'd love to see something similar with liberal arts instructors--I suspect rather different outcomes. Part of what irritates me in the call for revolution is the assumption that the quality of teaching is universal across disciplines. It's not, in no small part because t-t liberal arts faculty teach at least twice as much as most of their peers in the natural sciences, engineering, computer science, etc. As well, more attention is paid to our teaching at tenure time, even at R1s. As someone in the College of Liberal Arts, what I see (for the most part) is very innovative, energetic, thoughtful teaching. But I gather, anecdotally, that the situation is not so glorious in the sciences. So, perhaps, the message might be better directed to Agarwal's colleagues in disciplines more like his own?

  5. On one point you made, "As well, we need to know how best to help our beginning students learn HOW to learn, especially how to learn in a post-secondary setting where instructors aren't teaching to tests." a good set of videos are "How to Get the Most Out of Studying Video Series" http://www.youtube.com/playlist?list=PL85708E6EA236E3DB . The author, Steve Chew, is a cognitive scientist and former U.S. Professor of the Year (partly given for this work). I've shared it with people more knowledgeable about cognitive science than me and they were very impressed. Also, my students have liked it and tell me that it helped them. His "Improving Classroom Performance by Challenging Student Misconceptions About Learning"
    http://www.psychologicalscience.org/index.php/publications/observer/2010/april-10/improving-classroom-performance-by-challenging-student-misconceptions-about-learning.html can be eye-opening.

    On what is happening by discipline, mine, economics, it dominated by "chalk and talk." Watts, M. and G. Schaur, “Teaching and Assessment Methods in Undergraduate Economics: A Fourth National Quinquennial Survey,” The Journal of Economic Education, 2011, 42 (3), 294–309, is the most recent work. As a followup, I and a colleagues surveyed 275 economists; roughly 1/3 felt that "lecture is best," 1/3 that lecture wasn't, but was "cost effective," and the other 1/3 felt that non-lecture methods were best (but by our data lecture still dominated their classrooms).

    In math, nearly 2/3 of calculus instructors agreed with “Calculus students learn best from lectures, provided they are clear and well-prepared.” (see http://www.maa.org/columns/launchings/launchings_07_11.html ).

    On the Henderson and Dancy paper, I'm pretty sure that this is a "high" point among science instructors given the relatively high profile of people advocating non-lecture methods like Carl Wieman (Nobel) and Eric Mazur (Harvard).

    It would indeed be very interesting to see similar surveys done across disciplines. I didn't realize that liberal arts was so different from the sciences (or, apparently, social sciences).