Hacker News new | comments | show | ask | jobs | submit login
MOOC Completion Rates (katyjordan.com)
33 points by jnazario on Mar 3, 2013 | hide | past | web | favorite | 28 comments

I guess people see 10-20% completion rate and think: this is bad.

Let's do some math.

At my university (one of the biggest in Poland), there were 150 people majoring in CS per year.

According to the article, an average MOOC enrollment is 50.000. Let's assume 10% completion rate.

That's 5.000 people. It would take my university 33 years to graduate that many people (assuming 100% completion rate).

The cost per person? Several orders of magnitude smaller per student for MOOC.

Scalability? Almost effortless for MOOC, almost non-existant for the university (to double the number of students they would have to double the number of professors, double the number of buildings etc.).

What happens when you fail? At MOOC, try again. At university - you're out.

There are many reasons why completion rates are much better at university (you paid for it, you value it more; more external pressures (your peers, your parents), motivational support from your fellow students; the way you think about it (university: I have to do it or else it's really bad; mooc: no biggie, I can always do it again).

MOOC destroys traditional education at almost every other metric.

It's a classic Innvator's Dillema: a product that is not as good as what exists but so much better at some important things (cost, convenience) that it'll grow like weeds and will become better at things it's not good at (like retention) faster than incumbents becoming better at matching MOOC at price, convenience.

"According to the article, an average MOOC enrollment is 50.000. Let's assume 10% completion rate. That's 5.000 people. It would take my university 33 years to graduate that many people (assuming 100% completion rate)."

You are comparing a completion rate for 1 class to the completion rate for graduation at a university (consisting of a LOT of classes). If the completion rate for 1 class is 10% I cant imagine how low the completion rate would be for a whole degree assuming it would consist of somewhere around 20-40 classes.

Yes, the reasoning is flawed, but I think the GP is correct about the innovators dilemma that MOOCs pose for traditional higher education. Many universities, behind closed doors, are scared shitless about what's going to happen over the next couple of decades.

The availability of MOOCs only reinforces the feeling that traditional undergraduate education is losing its value. Couple that with tuition costs greatly outstripping income and inflation, and other sources of funding (grants, state support) flat or waning, and it's easy to paint a bleak picture for the future of the traditional undergraduate university.

On the other hand, the genie is out of the bottle. Universities can't pretend that MOOCs don't exist, or they will lose out to others who are investing in them. With near zero marginal cost to scaling, the price (if any) is going to be under huge pressure to stay low.

A dilemma indeed.

Your nitpicking is correct. I don't think it changes the big picture, though, so it's kinda impotent nitpicking for nitpicking's sake.

If I were to be pedantic about every little detail, I would end up with unified theory of the universe.

But since you brought it up: MOOC will also dramatically alter how we look at higher ed.

For historical reasons, my CS department emerged from EE department, since electronics existed before computers. As a result, I had to take courses in analog electronics, semi-conductors and what not. Out of 5 years of study, about 3 were relevant to CS.

I hated those courses, they proved to be useless in my programming career and yet, I had no choice.

At the same time I couldn't have taken a course in Greek history, even if I also liked Greek history (Spartans, those buff, semi-naked men!) in addition to programming.

The future will not be like that. The future of higher ed will be à la carte.

People will no longer get a major in CS, they'll take whatever courses they like and think are useful in finding employment.

It's happening already - Udacity already made a big deal (rightfully so) about people who found employment at good companies after completing merely few courses.

(as an aside, the rarely spoken truth about programming is that you don't really need education; I don't credit my university with teaching me much about what's actually relevant to real-life programming)

While I agree that MOOC is absolutely going to disrupt traditional college/university schooling, I disagree with a couple of your arguments. Actually no, scratch that, now that I think about it, if I had been able to pick my own curriculum then yeah--(snip a bunch of stuff about what I myself learned, did not learn, would have picked, etc. nobody cares about that!)--maybe I would not have learned AS much, but like you say, at least most of it would have been useful to my personal development. And I could have patched some gaps in my knowledge (Analysis) that later on gave me trouble with advanced courses that I really could have gone for otherwise.

And the GP comment was not nitpicking. Nitpicking is when someone makes a big deal out of a small thing. Just because the error they pointed out does not invalidate your argument, does not make it nitpicking. Yours was a pretty big error: comparing dropout rates for a single class to those of an entire education. And it deserves pointing out, because such a "little detail" has a good chance of causing confusion later on in the discussion. And you calling it "pedantic" really does injustice to the GP calling out a genuine error. It's not like they were being rude or annoying about it. And I admit, this paragraph can be considered both nitpicking and pedantic, I still want to say it because you could have been more polite about it. "Right. While that doesn't invalidate my argument, thanks for the correction."

> I cant imagine how low the completion rate would be for a whole degree assuming it would consist of somewhere around 20-40 classes

Nowhere near as low as 10^-20 to 10^-40, since completion of each course is most likely extremely not independent (i.e. if someone finishes 4 courses, they are likely to finish 10).

I don't understand why we are comparing the completion normal rates with MOOC's completion rate. I think that they can't be compared yet. While I am doing by degree full time, MOOC is an additional learning for me. If students could complete all courses online without attending a single class offline,the conversions would be much better. You cannot say MOOCs are good or bad just on the basis of completion rates.

He's not, though.

He's not taking his school's completion rate into account at all, only the total number enrolled in the CS program.

"That's 5.000 people"

But its not 5000 people because they aren't graduating - they are just passing one class. If 5000 people pass that one class and then they have to stay motivated to pass 20 more... its not going to be 5000 people that graduate in the end. Even with the massive number of enrollments in MOOCs the graduation rate could be so low that his CS school is producing way more grads.

Arguing about low completion rates when projected across 20-40 classes is only showing you're stuck in an old, clunky way of thinking.

Points to think on:

1. You're acting like dragging students through a muck of 20-30 TA or GTF taught classes, that are outside the student's major, and often outside their interests due to strict guidelines for class choices for gen-ed, is the gold standard of a good education. Check your premises as to the value of such a program and the resulting degree. I would argue, quite easily, that a student completing 10 Coursera classes in the field their interested in, already outweighs, in usefulness towards personal and career skills, those 20-30 gen-ed classes and probably provides a higher quality of education for the remaining major courses.

2. My experience is with Coursera, so I will only make points based on that. Coursera is finding the nation's (maybe even world's) best teachers for each subject. You may be able to go to Stanford and get a couple of the best experts to teach you on their subject, but any given bachelors program can only hope to have 1 best-in-class teacher. I have taken 2 Coursera classes, and can tell you the quality of the teaching is much, MUCH higher than my bachelor's program. There's really no comparison of the level of quality of the education from a Coursera class, to any non absolute top of the top tier schools.

3. These are retention rates with absolutely no external rewards. You might get a certificate (of no explicit value) if you're lucky. But there are no degree programs, there really is no external motivation (or negative reinforcement) around class completion. To compare the figures with traditional universities, where students have numerous external motivations, consequences for failing, etc. is just naive, inconsequential, stupid, etc.

And how many people are passing 1 class at his school?

It can't be more than the 150 in the program.

I sign up for just about every one that looks interesting, obviously without the time to finish them all. The reason is quite simply because I want to look at the materials on my time, occasionally take part in the discussions, and try out the assignments here and there that look particularly challenging. Last I checked, you had to be registered to view any of the materials on EdX and Coursera which is pretty much why I do it this way.

Yeah, I'm pretty much the same. Which should be completely fine with anyone, since it's putting me way ahead in terms of learning from where I would have been without the course.

Completion rate is only one measure of these courses, and I'm not at all sure it should be the most important one.

The last MOOC I took was MIT's introduction to solid state chemistry (great course, BTW). Before an actual MIT student can take that course on campus, the go through a pre-screening process that ensures:

1. The student has the necessary background knowledge that the course assumes,

2. The student has the necessary skill required for the course,

3. The student has the necessary motivation to do the work,

4. The student has the time to devote to the course.

There's also a pre-screening process for people who wish to take the online course through edx.org. This pre-screen process ensures:

1. The student can successfully create an edx.org account and click a sign up button for the course.

Given this huge difference in screening of students, and the fact that the online course is for the most part not "dumbed down" compared to the on campus course[1] , I would absolutely gobsmacked if the MOOC version of the course came within an order of magnitude of the completion rate of the on campus course.

[1] Online students were given 72 hours on the exams, to allow for computer and network outages, and to give more flexibility in accommodating their lives outside of the course. On campus students were given 3 hours for the exams. I think they said the content of the exams was the same.

MOOCs are still in the experimental phase. I finished 1, and about to finish another soon. I believe, there is still a lot to innovate when it comes to giving the lecture and evaluating the student.

I also believe, a major goog/fb scale company will raise for everyone to get continuous education.

Its silly to criticize them for rates under 20% since many people sign up just to see what it's like without actually planning on finishing it. What would be more relevant is to see what percent of serious committed students finish a course. E.g. what percent of students who pay for a course end up finishing it?

Of the MOOCs I've participated in, the student surveys have never asked the most relevant question I can think of with respect to completion rates: "do you (or did you) intend to complete the course?"

Anybody here ever been asked that?

Very good point. Of course, for a completely fair comparison you need to ask the same question (before the course starts) to the traditional students. I expect it can't be much more than a percent or so, but it depends on whether the university penalizes students for signing up for a course they don't finish (in my years of college, the CS department didn't seem to care very much. It was very different for other departments, though).

Of course, without penalties, you will get students that start out a semester following more courses than they can finish, only to decide later on which one to drop. I'm not entirely sure how that translates to MOOCs, though.

That's a big point - I have joined both courses that matter to me and courses that seem fun or relate to my side interests, where I just want to watch a few of the lectures.

In the first part, I have 4/5 completion (and I'll finish the missing one when it will be run next time); but for the second part, naturally, I have 0 out of 20+ courses.

This is interesting, but I think completion rates (and enrollments, to a degree) are bogus statistics. All the incentives are aligned toward signing up for lots of courses, with or without any intention to finish. I'm "enrolled" in 13 Coursera classes at the moment, but only active in two (and something close to this has been my moving average for the last six months).

The only cost to me is a few too many automated emails. And since course quality still varies widely, trying out lots of courses and sticking with the best ones is a good strategy.

Yeah, I've signed up for about a dozen Coursera, Udacity and CalTech courses over the past year, and finished 3 with enough work for certificate of completion, and put in enough work on a few more to get something out of them.

I think the data is interesting, but the point it infers is absolutely moot: I have personally derived a a HUGE amount of educational value from several MOOC's that I never completed. That's part of the beauty: you can learn what you need to learn, and you're not forced into a box that the course prescribes. You can pick-and-choose your own learning process.

Could it also be that many people will sign up in order to get access to certain materials, but then not end up completing all of the assignments/exams?

I don't have any experience with MOOC's, so I'm curious as to how these peer graded systems work and why they result in lower completion rates.

I have signed up for several courses on coursera, but have completed only one. Because of time constraints, I just sign up, and then go through the materials at my leisure.

I don't care about Mooc completion rate at all. If for example only a few hundred people completed the course, this means that the course had brought educational value to thousands.

For some professors the kickback value might play a bigger role. Some courses on coursera are just not taught as widely as the professors would like to have it. For example Hinton's Neural Network course, or the one about Quantum Computation. In these cases, a few hundred completers would be plenty to justify the effort of the authors (in my opinion at least).

Interesting in what that might say about job candidates who have actually completed a number of MOOCs. That would put them in the top 10% or so by that measure, for whatever that's worth. It at least says something about their ability to start something and finish it.

Are MOOC completions something that job interviewers actually look out for?

I have always used MOOCs to learn, not to get certified.

I don't know why people are seeing low MOOC completion rates as a problem. It just makes completion more valuable as a credential. That plus the fact that MOOCs have wider accessibility should make traditional institutions very scared. It's the MOOCs' game to lose.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact