cc licensed ( BY NC ) flickr photo shared by Marina Cast.

If a course is going to be pumped up on the massiveness of its open enrollment, it ought to own how much or little comes out at the other end.

I still maintain that the idea of a “dropout” in an open course– where it take no effort or no skin in the game to drop in — is meaningless

But allow me to ponder some numbers just shared for the Coursera Social Network Analysis class. I signed up for this out of a real desire to be able to do some (and understand more) of the things Tony Hirst does.

I watched two weeks worth of videos, did the multiple guess quizzes, but flubbed out on the first assignment. I found the videos way long in places the instructor was reading my stuff I could scan on a web page, and once we got started, I never really saw any of the first grab of interest to help spark my motivation to do more. It went right to theory and tossing numbers and settings in Gephi, and there was no sense of WHY I would be doing these things.

Now I have to own my responsibility here, I gave it not my fullest attention, and once I had missed the assignment, it seemed pointless to catch that speeding train. Heck, I could not even see the caboose.

You see, the course moves at the speed it wants to, not mine. This mode does not use any of the affordances of online learning to be able to flex time and space for me to do work- it just marches on everyone rowing the boat together (or falling over).

But let’s look at the summary information sent out this past week:

Some participation stats: 61,285 students registered, 25,151 watched at least one video, 15,391 tried at least one in-video quiz, 6,919 submitted at least one assignment, 2,417 took the final exam. 1303 earned the regular certificate. Of the 145 students submitting a final project, 107 earned the programming (i.e. ‘with distinction’) version of the certificate.

61,285 students registered.. I was one
25,151 watched at least one video (41%)… I saw two actually
15,391 tried at least one in-video quiz (25%)… I did about 3 of these, but found it was more just clicking to try and guess the right answer more than learning.
6,919 submitted at least one assignment (11%)… That does not include me, and is not a stat I would be proud of.
2417 people took the final exam (4%)… the class is pretty empty now
1303 earned the regular certificate (2%).. Bueller? Bueller? TO earn this certificate, you have to achieve “80% of the points” – Its not clear if that includes the final or is just the weekly assignment points. If I am reading this correctly, a smidgen over half of the people who took the final exam earned a certificate.
145 students submitted a final project (0.24%)… I would guess someone who submitted a project did the whole class
107 earned the programming (i.e. ‘with distinction’) version of the certificate (0.17%)

Do you need to see this as chart gunk?

So in the end, we have 107 students who got the more personalized attention (doing a project, getting feedback, being part of the Google hangout presentations).

This class had one professor and 3 TA, about a 1 : 27 teacher/student ratio.

That is pretty much the size of a normal section of a class, it is the size of one of our ds106 sections at UMW.

Now there are a whole raft of reasons why people do not get to this end of the pipe, many, liem in my case, fall on my own lack of drive to really push this up the hierarchy of where I put my attention.

But I submit the methodology of this course too has a large influence as well- it did not hold the attention of the bulk of its students, like 98% of them.

Let me repeat, 98% of the people who signed up for this course did not get the certificate, or 60,059 people. NOW THAT IS MASSIVE (as in hemorrhaging).

Yet the bulk of the hyper and fervor on MOOCs is the massive numbers of enrollments whichm, frankly, when you look at these numbers, it is the wrong end of donkey (to quote Neil Young), or maybe in this case… MOOcows.

cc licensed ( BY NC ND ) flickr photo shared by iansand

Someone ought own those numbers coming out the end.

Bueller? Bueller?

Profile Picture for Alan Levine aka CogDog
An early 90s builder of the web and blogging Alan Levine barks at on web storytelling (#ds106 #4life), photography, bending WordPress, and serendipity in the infinite internet river. He thinks it's weird to write about himself in the third person.


  1. To be honest, when I heard George Siemens and Dave Cormier talk with Martin Weller about the real power of MOOCs being in the M for Massive and the C for Course I had to disagree. I continue to feel both add only a modicum of innovation of what’s happening with the open. And if this is post is a loose case study, which I take it as, then you do a pretty good job of looking at the real issues behind the idea of Massive (and I’m sure their are affordances and that people in India are learning—but I would argue they have been for a long time with the web) is that it turns the learning process into a series of stats and analytics. I want to see some real examples where the Massive part (namely having 60,000 students sign up for a class) made it better qualitatively. Is there any evidence of this? As for the course part, if this is good course design and structure than we have made no advancements in the last 20 years–in fact quite the opposite. Coursera courses dumb down and already bad LMS model.

  2. I suppose there is a small percentage in there who just took the parts they wanted or needed and didn’t do anything else (the ability to do that being more or less limited by the structure of the class). I’ve done that a few times myself. But that doesn’t make the numbers better in any real way.

    That said, what is success here? Given the super-simple registration that facilitates many “tasters” who were never going to finish, or even planning to, how many have to “finish” before it’s an acceptable percentage? Does it matter?

    1. I don’t know Chris, its a lot of guessing and supposition among 60,000 stories, and if someone waded through the thousands of posts in the forums, one might get a better sense.

      I did not find this course structure to be one that was set up for tasting- it was very sequential, and I could not see just jumping on for one small part.

      And I agree the the finishing does not matter, and what really needs to be examined are some sorting of people’s initial goals. The part that I was harping on is that there is all the rah- rah about the 60,000 on the front end- and whether we look at completion or some other sort of measure, someone needs to pay more attention to the outcome end of this, and not in just the analytic farts and charts kind of way.

      That said, I have already signed up for 2 mor Courseras. One of these days I might stick

      1. I don’t think we are disagreeing. I just suppose–and this is an aspect we are dealing with when it comes to open course materials–I am interested in the stories of those who come, get what they need, and leave. It’s not something the university, who wants the tuition dollars, considers much of a success. But I think those learners probably do.

        A course being enforced on a timeline makes this much harder (if not impossible). But for those courses that don’t…

        It’s difficult to untangle, which is why I asked the rhetorical question about percentages. I’m (kind of) surprised that there aren’t some registration questions that get at some of these questions. There’s a big difference between 90,000 of 100,000 intending not to finish and 10,000 intending not to and having 90k still drop out. And by the nature of the open registration, I have to imagine a very large percentage were never that interested in the first place.

        I’m not saying the numbers are GOOD, I’m just saying they are mystifying and thus, for now, meaningless.

  3. The “massive” and the “course” are in continual conflict. Massive encourages dropping in and out and around (and, as you say, perpetuates the LMS model when done wrong – and most of these now are indeed done wrong). Course encourages focus, and thus maybe a smaller group. The “masses” don’t want a “course” and a “course” doesn’t do well with “masses”.

  4. Leave it to you, Alan, to take a hard look at the MOOC phenomenon and give us a dose of reality. Those numbers are sobering. Unfortunately, the bandwagon effect we’re seeing on MOOCs has more to do with what I fear are CFO-type enthusiasms – as, in “here’s a way to do more with less!” I’m not buying it.
    I took a Coursera Modern Poetry MOOC this fall and had very similar questions. Like you, I started with good intentions and then, after the first assignment (3 weeks in), I fell behind. Your point about the course sailing along at its own speed, leaving the learners to row in time or sink beneath the waves, is right on. This is not adaptive learning that leverages what the web can do well. But then, to be fair, adaptive learning is not what the designers of the course I took intended.
    The “ModPo” instructor did an excellent job designing his course – a clear syllabus page, weekly readings and audio files, weekly videos (which featured the prof, with a flanx of grad students seated around a table, discussing the week’s assigment), and a regular synchronous webinar session in which you could participate via Twitter or call in. The work was serious, but doable, and the various modalities were well leveraged and implemented. It was clear that a tremendous amount of thought went into it and the instructor was charismatic and effective. Unlike your experience, I was definitely motivated and interested. But once I fell behind, it was impossible to get back into the flow (or at least, impossible for me). Like your course, this was not one where you could just drop in for the piece that interested you or easily pick up where you left off.
    The other observation I had was that the sheer enormity of the course was daunting to me. At the first live session I felt a bit lost and tongue-tied (unusual for me, ahem). Quickly, a small group of us on Twitter self-organized and formed a smaller Facebook group that worked very well. We discussed the assignments, compared notes, and supported each other. There’s power in that – but it was a self-organized artifact of the course.
    Seems to me that the real power of MOOCs is the possibility for gaining network effects, for uniting a learning community around a set of ideas and investigations, for the “course” to take on a life of its own and gain power through the collective. I think we’re getting distracted by the “M” part of MOOC (wrong end of the donkey, indeed) and seeing that as number$ instead of a network.

  5. I have been having thoughts about MOOCs for a few years now and have even participated in two, one of Stephen and George’s several years ago, and the second more recently (a Stat course from Professor Conway at Princeton). I did not finish either of them.

    The first, I started as part of a project at the school where I was the IT Director, to demonstrate some alternative pedagogy to faculty. Three others, including the Vice President of Academic Affairs joined me and we had weekly in person meetings to discuss it. That was pretty much a failure. None of them got the picture enough to want to sign up with their email address (too much email already), so it was pretty much a bust from the start. Too bad…I thought it had potential.

    The second was an attempt to reboot my statistics. I was spending time on it and pulling 80-90 on the homework and quizzes. Then I mis-read the due date on the mid-term (in a quirky sort of way, perhaps to see if we were paying attention, the due date was 9am instead of the usual 9pm for all previous assignments and quizzes). Missing the ability to get graded on the midterm took all the wind out of my sails and I ‘dropped out’.

    So the first experience was probably premature and I did not vest myself in the actual course because I was managing recalcitrant faculty. In the second, I did not pay sufficient attention and lost interest when I lost the possibility of a payback.

    That said, I believe that there were SIGNIFICANT differences in the two MOOC styles. The Downes/Siemens course provided a peer-to-peer based learning experience that was quite open and could lead to a unique and distinctive opportunity for equal exchange of ideas. On the other hand,the Princeton course was a top-down and traditional pedagogy with limited peer-to-peer exchanges. Most of these exchanges involved more experience students acting as mentors to those who did not have the background or were struggling with the technologies (R) used in the lessons. The former has real possibilities for reforming education. IMHO, the latter continues a well worn path that I rejected decades ago. Now when I discuss MOOCs with people, I am careful to make the distinction between the different styles.

    1. “[T]he course moves at the speed it wants to, not mine”: I’m curious, Alan. Was that because certain timed events would expire?
      Would it have been possible to complete the class at a slower pace?

      1. Well, it may not be a fair statement, *any* course does need a schedule. It’s more that if you fall off the pace it feels very difficult to catch up (again like any course).

        I’d gather one could go back and use the material at one’s own pace to learn the content, though one would not get the certificate.

          1. The mention of time here has me thinking the whole MOOC “model” (who knows what that might be?) seems like a continuous conversation that people actually do move in and out of. What is the purpose of a time limit in learning? Not trying to pose a philosophical question here, only why dice learning into pieces at all?

          2. I’m thinking for many people much learning is project oriented, since it allows for deeper immersion in a subject or process. Ongoing learning, if dependent on community, would require larger numbers of people

  6. Think Lisa may have nailed it by bringing up the subject of process learning. To me “process” implies participation and a sense of membership. Great numbers here might make for too many cooks in the kitchen but as members we can “allow” ourselves to split the tasks or redefine the recipe (outcomes) to accommodate a spectrum of dishes. When there’s something to be done together people adapt, adjust, take turns, form sub-groups, negotiate and generally make as much room as needed.

    Taking this natural phenomenon of emerging community and stripping it of a group purpose by returning to the spectator model of education is what these Stadium type MOOCs seem to favour. Without having to accommodate the presence of actual (in the sense of hands-on participants) people, massiveness is just an abstraction or statistic.

    Removing or interchanging audience members has a way different effect (if any) than changing players on the field. If we are looking to develop people who can act in the world over merely being there then the latest MOOC rip-offs are just entertainment.

  7. massive can become too much (of not really such a good thing anyway), courses are not indefinitely scalable upwards. when 1,500-2,000 starts feeling cozy, Houston we have a problem.

    that said, the most satisfactory part of an x is forming smaller groups outside the silo, making connections avoiding the platform except for major shopping trips to stock up.

    down-scaling works better

  8. So Vanessa do you think the value in an xMOOC arises from its being and attractor and resource? Can we be contacting many more than can carry on a meaningful conversation which might silence some while also encouraging others to split off into manageable sub-groups?

    Insisting on size as a driver of ideation might be mixing two levels of cognition. Massive input could be argued to at least return a sense of variety but that abundance could also swamp processing. We can filter only so much before we begin have too many categories to sort through for meaning. And I think we can also limit out on diversity in building meaning.

    I can “explain” virtually anything I encounter while awake or asleep but I wouldn’t fly on a cow that I’d mentally assured myself was an airplane (well, maybe one of Allan’s cows).

    And after all that, maybe MOOCs could be generators of ideas you weren’t looking for simply by being so bloody massive? Released from the harness of intentionality we soar with our bovine angels.

Leave a Reply

Your email address will not be published. Required fields are marked *