Home > education, HCI, MOOCs, teaching > MOOCs: One Size Doesn’t Fit All

MOOCs: One Size Doesn’t Fit All

Can a MOOC teach course content to anyone, anywhere? It’s an imagination-grabbing idea. Maybe everyone could learn about topics from the greatest teachers in the world! Create the class once, and millions could learn from it!

It seems like an exciting idea. Until you realize that the entire history of human-computer interaction is about showing us that one size doesn’t fit all.

I went to two terrific talks on MOOCs a few weeks ago. At the GVU Brown Bag, Karen Head and Rebecca Burnett talked about their brave attempt to teach an English composition class free online, for anyone interested. They encountered intriguing challenges. The course said you needed to be fluent in English as a prerequisite, but some people who signed up were far from fluent. The course did reach people all over the world, but some of those people disapproved of Karen’s style of dress. Her attire was conservative by American standards, but the world is a big place and some people felt otherwise and said so, loudly. The instructors used YouTube as part of assignments, but YouTube is banned some of the countries their students live in. Trying to accommodate everyone everywhere turned out to be a stressful endeavor.

The second talk was by Ed Cutrell of Microsoft Research India at the ACM Learning at Scale conference. Ed has been studying engineering education in India. He explained that the top-level universities in India serve about 40,000 engineering students per year. The rest of the universities serve another 4,000,000 engineering students per year. The quality of teaching at lower tier universities in India is wildly variable. In some cases, the instructors don’t know the material themselves. Many aren’t career teachers, but are teaching on a temporary basis until they can find an engineering job. Given those constraints, Ed and colleagues are using a blended learning model where short videos by experts are followed by class discussion. With this approach, the instructor and students learn the material together. Trying to respond to the particularities of a learning context, Ed was able to design a successful solution.

If I am designing a class, even if I limit my audience to “Georgia Tech students,” that isn’t a specific enough. I would design a different class for undergraduates versus graduate students, for computer science (CS) majors versus majors in computational media (CM). I don’t always have the luxury of making those classes different—economics dictates that the CS and CM majors take almost all their classes together. Financial considerations push us to generalize.

Do you own any clothing that is one-size-fits-all? It works great for my nightshirts, but it wouldn’t work if I tried it for pants. And there are some people who don’t fit into one-size-fits-all, even for nightshirts. The analogy to online classes works pretty well. There are a few simple things where one size will work tolerably well. But in most cases we should stop trying to make one size fit all.

Categories: education, HCI, MOOCs, teaching
  1. March 27, 2014 at 5:01 pm

    Nice post. In K-12 classes I’ve observed, some teachers shape their classroom activities in an adaptive fashion in response to the class responses. This shows some of the challenges and tradeoffs of using technology to flip classrooms, and of using one live class as a studio audience. Of course, there are great, widely useful online materials. But it will take a lot of trial and error (or, less likely, thoughtful consideration) to work things out.

  2. March 27, 2014 at 7:15 pm

    Thanks Jonathan! I want to hear more about the research you’re doing!

  3. March 30, 2014 at 10:11 pm

    I also have a sceptical attitude towards MOOC and wrote a blog post that touches on (other) issues I have with MOOCs, “On the eroding status of the academy” (). Here’s a quote from my text:

    “I can also easily imagine a bifurcation with elite face-to-face universities for the select few (think Oxbridge) and online courses for the masses – because knowledgable, enthusiastic, caring teachers do make a difference!”

    I still think that’s the general direction we are moving in, led as usual by “useful idiots” who plunge ahead fuelled by enthusiasm and lack of reflection. Oftentimes it (unfortunately) pays off not to think to much, see further Alvesson and Spicer’s of “Stupidity-based theory of organizations” ().

  4. bckirkup
    April 18, 2014 at 12:52 pm

    If this is being held up as an objection to MOOCs, I must protest. This is a strawman argument. If in theory there were one MOOC that could teach everyone everything, that would be … well, it’s preposterous.

    So, the real question is something else.

    Right now, students are bound to the local insitutions, usually one in particular, which usually offer only a couple instances of any given course. The audience is between a dozen and several hundred, pre-segmented mostly by major, year, prerequisites. The instructor does not really have much opportunity to tailor the course; also, starting the course usually means investing significant time and energy.

    Imagine an alternative, in which some clearinghouse lists 20 nearly concurrent Intro CS MOOCs. Perhaps reviews of past offerings give some students an idea of which is best, but students may start more than one. They may drop some or all of them, perhaps to try others. If experience is a guide, perhaps 100,000 students start and in any given class, only 4,000 complete. That’s fine, if the friction of starting is low. They may have started 3, or proceed to find a better fit. If overall 90% or more of the students complete even one of the classes, the entire system is a raging success. Each of the 20 can be presented in the next cycle, and Intro CS can be offered to the world. This is no objection to MOOCs.

    Could some of the students have been better served by an in-person class? Sure. Could textbooks do many of these jobs? In some cases – but the question of balancing book and online materials is really semantics and profit centers more than a real debate. But perhaps rather than arguing against MOOCs, engineering better MOOCs that speak to an audience, and an effective clearinghouse to help students self-select into the appropriate courses, would be logical.

  1. April 18, 2014 at 1:39 am

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: