It’s been a pretty busy and productive time in MOOC-land. I’m simultaneously working on three different “parts” of the MOOCs In Context project with the hopes of having enough to seriously start seeing if there’s a publisher interested in whatever this will end up being. I’ve got a chapter coming out sometime in the near future (yet this year?) in a collection edited by Liz Losh about MOOCs, and I’ve got some other MOOC scholarship news on my mind I’m not quite ready to announce to the whole world yet. And my garden is completely in. So it’s been a good sabbatical, one that will end sooner than I had originally planned– but that’s another post. Anyway, more of this post after the jump.
A lot of what I’ve been writing about specifically lately has been about the “production values” of these MOOCs, particularly when it comes to things like the videos. One of my big complaints about “Listening to World Music” was the incredibly poor quality of the videos; simultaneously, one of my problems (well, “problem” might be too strong of a word choice) with the EDCMOOC was the absence of instructor videos entirely (at least in the version of the course I took and not the updated versions). And one of the many interesting things I have gleaned from the interviews with MOOC faculty I’ve done so far has been about the challenges for faulty to present themselves online, particularly to a large and anonymous audience of thousands from all around the world.
So the stuff I’ve come across lately about MOOC videos (or not), actors, and robots have all been kind of interesting. For example:
- “When Actors Replace Instructors as On-Camera Talent.” This is about an Engineering Professional Education program at Purdue (it’s “non-credit” and for “lifelong learners”) that is starting to experiment using actors to perform faculty-written lectures. I actually think this is a pretty interesting idea. A lot of faculty-types are simply not that comfortable “performing on camera” the way you need to be with videos for online classes, and I don’t see getting someone else to perform a lecture to be any less authentic than assigning a textbook written by someone else.
- “Why My MOOC is Not Built on Video.” Of course, the other way to go is to minimize video, which is what Lorena Barba is suggesting here. She is a professor in engineering at George Washington University who taught a MOOC that didn’t have a lot of video. Her (really, their– it was team-taught) reasons for this boiled down to a lack of time and the costs of production, which are both very real issues indeed, especially if you’re not sure you will be able to reuse the videos again and again. But Barba also tells the story of taking a MOOC herself and how the videos– while easy to watch and well-produced– didn’t really teach her much. “Videos are nice, they can get you exposed to a new concept for the first time in an agreeable way, but they do not produce learning, on their own. Students need to engage with the concepts in various ways, interact with ideas and problems, work through a process of “digestion” of the learning material.” Substitute “lecture” for “video” and I think that’s about right, though as Barba goes on in this post, it’s more complicated in that there are some legitimate uses for both videos and lectures.
- Okay, so what about robots? There were a couple of articles and such lately about the Twitter ‘bot for EDC MOOC recently– this from Compute Scotland and this from the Times Higher Education. As far as I can tell, it was an experiment in automating responses to some common questions from those enrolled (posted on Twitter with the #edcmooc tag). The context here makes some sense given that EDC MOOC is/was a class about “digital cultures” and educational experimentation, and also given that there were tens of thousands of “students” who were willing to play along. But really, it appears to have been more the Twitter ‘bot equivalent of answering the kinds of questions where my “automated” response is often “it’s on the syllabus.”
- But what about real (or more real) robots? Two kind of interesting bits to think about here. First, there’s “Social robots and virtual agents as lecturers for video instruction” by Jamy Li, René Kizilec, Jeremy Bailenson, and Wendy Ju (who are coming out of the Department of Communications at Stanford) in the May 2015 issue of Computers in Human Behavior. That article is behind a firewall (the link here is my access to it at EMU via the ScienceDirect database) and I’ve only read the abstract, but it sounds pretty interesting:
One emerging convention in video lectures is to show presentation slides with an inset video of the instructor’s head. Substituting a robot or a digital agent for the video of the instructor could radically decrease production time and cost; thus, the influence of a digital agent or robot on the learner should be evaluated. Agent-based alternatives for a talking head were assessed with an experiment comparing human and agent lecturers in a video from a popular online course. Participants who saw the inset video of the actual lecturer replaced by an animated human lecturer recalled less information than those who saw the recording of the human lecturer. However, when the actual lecturer was replaced with a social robot, knowledge recall was higher with an animated robot than a recording of a real robot. This effect on knowledge recall was moderated by gender. Attitudes were more positive toward human lecturers than toward robots. An initial proof-of-concept demonstrates that although a human lecturer is preferable, robotic and virtual agents may be viable alternatives if designed properly.
- And another sort of related example I came across via NPR the other day: “How A Machine Learned To Spot Depression.” This was about a “‘bot” called “Ellie” that works as a way of diagnosing PTSD; basically, the software asks subjects questions, listens to the responses, and analyzes the body language to help with a diagnosis. Right now, the idea is to develop Ellie as a screening tool to steer soldier with potential PTSD issues to human therapists.
So to me, it doesn’t take a lot of imagination to see how these kinds of tools might be helpful in teaching, particularly in something like a MOOC. A lot of the kinds of routine questions and concerns that students have probably really could be addressed by a robot (imagine the handiness of a ‘bot that could answer routine student email questions about when things are due and whatnot) and if there are ways to make lectures and other presentations in MOOC-like forums more engaging and interactive with robots, well, why not?
- Of course, no technology alone can “solve” education, which is essentially the point of the first part of my project and with the research I’ve been doing on correspondence schools, radio/TV, and “traditional” online courses. Two articles I came across about this recently, both of which I guess kind of fall into the “duh, obviously” category of things. First there’s “Why Technology Will Never Fix Education” by Kentaro Toyama— it was a commentary piece in CHE. A few quotes:
Over time, I came to think of this as technology’s Law of Amplification: While technology helps education where it’s already doing well, technology does little for mediocre educational systems; and in dysfunctional schools, it can cause outright harm.
The Law of Amplification provides one such framework: At heart, it affirms that technology is a tool, which means that any positive effects depend on well-intentioned, capable people. But this also means that good outcomes are never guaranteed. What amplification predicts is that technological effects follow underlying social currents.
MOOCs offer a convenient example. Proponents cite the potential for MOOCs to lower the costs of education, based on the assumption that low-cost content is what is needed. Of course, the Internet offers dirt-cheap replicability, and it undeniably amplifies content producers’ ability to reach a mass audience. But if free content were all that was needed for an education, everyone with broadband connectivity would be an Ivy League Ph.D.
The real obstacle in education remains student motivation. Especially in an age of informational abundance, getting access to knowledge isn’t the bottleneck, mustering the will to master it is. And there, for good or ill, the main carrot of a college education is the certified degree and transcript, and the main stick is social pressure. Most students are seeking credentials that graduate schools and employers will take seriously and an environment in which they’re prodded to do the work. But neither of these things is cheaply available online.
again, smart stuff but also kind of “duh.”
Then there’s this from TechCrunch, “Why Is The University Still Here?” by Danny Crichton. He’s writing in essentially the same vein as Toyama. For example:
However, motivation (and also patience) is the key ingredient for success in education. While drop off rates in course completion are partially indicative of the quality of the MOOC education product, they are far more valuable as a reflection of the limited dedication to continuous learning that most adults have in the first place.
I realize that this can be hard to accept in an industry like software startups where the rate of autodidacts is probably one of the highest in the world outside universities themselves. Nevertheless, there is a reason that teaching colleges spend so much time discussing how to inculcate lifelong learning, since that is simply not the default for most people. Family and job demands (aka life) can easily preclude this sort of on-going investment of time into education.
Besides lack of time though, the key challenge for open online education was connecting learning to more pecuniary outcomes, namely job performance, promotions, and new job searches. Outside of programming, which seems to be a unique area of learning with a high return on investment, few courses seem destined to transform a working adult’s job prospects.
Though one thing that I think is important to remember here about both of these arguments, that MOOCs (et al) can’t compete with/replace higher education as we know it because of student motivation and the recognized and historic (market?) value of a college degree: I’ve never met anyone who has taught a MOOC who actually believed they would compete with/replace higher education. Really, the only folks who said MOOC courses and badges were going to compete with university degrees as we know them were policy wonks and folks hoping to make money from all this, and really, those people were just wishing and hoping that if they keep repeating the same wish over and over, maybe it will some day come true.
But hey, maybe it will come true some day. For example, I’m keeping an eye on Laurie Pickard and her No Pay MBA MOOC project. Of course, she already seems to have a few things going for her– Oberlin grad, already describing herself as “an international development professional specializing in business development” with a lot of professional experience. She’s already a driven/self-motivated learner, and she probably could afford the traditional “pay MBA” program. So maybe not exactly a good test case here, and I have a feeling she isn’t really trying to get a no pay MBA. Really, this strikes me as the kind of project that is laying the groundwork for a book deal she can sell about how “you too” can go the no pay MBA way.
- On a slightly different but related note about the value of degrees: there were a couple of articles out about degrees from schools of dubious value and/or ones that are outright fakes, notably an outfit in Pakistan called Axact. Aaron Barlow has a good post about this here. He describes this as the logical end of for-profit, online education, but I think it’d be more accurate to describe it as a continuation of a process that’s been going on since similar fraud schemes emerged parallel to more legitimate correspondence courses.