The course has presumably been designed so that a motivated and intelligent student can understand it in this time.

But what happens when life interferes with a student's plans? I can think of lots of reasons why a student may be distracted by legitimate life events in this period of time, e.g. illness, or a family wedding/birth/death etc.

Online courses do give you the flexibility to work around things like this. They also allow people with full time jobs to access the material at a pace which suits them.

But it doesn't do that. It merely shows you were able to perform well in a contrived testing environment that has little to do with how you'd actually approach problems in the "real world" (that is, everywhere but a test).

>12-13 weeks is plenty long to learn the material for a course. A student is at a college to learn. It should be their top priority to learn the material during the course.

I think you're missing the author's point here. Yes, a student is at college/university to learn. However, the job of the institution should be to teach, not to weed out students, and that's where the true failure is taking place. And again, nobody is always in full control of their life. Shit happens, as they say, and just two weeks of illness (or really any unforeseen circumstances) can cost you a whole semester. Some students need to juggle studies, work and getting enough sleep while pretty much having no life because of the horrendous costs of attending higher education (which is a whole other problem entirely).

>As for "failure #3", lectures have been used for hundreds of years because they work.

That's an appeal to tradition. They haven't been used because they work, but because there was no viable alternative. But now, we have such an alternative - video lectures and online courses.

>He also seems to miss the fact that for science courses, you need labs. You cannot do chemistry labs online. You won't be able to do field work in biology by watching a video.

That's true indeed. I don't think it's productive to attack universities as a whole, either. They serve a lot more purpose than just holding lectures. You can meet people in the same or adjacent fields as you, you can form connections, work in research (and maybe even stay there after graduation) and many other things. But that doesn't change the point of the article, which is that lectures as a teaching format suck when compared to the alternative.

>You lose the overall atmosphere of a learning environment through online courses, and that is not something I'd want to give up.

You'll have to excuse me for being a bit snarky now, but if by "atmosphere of a learning environment" you mean cramped lecture rooms, long and drawn-out lectures without much (if any) interaction with the lecturer with the overall process often being more of a hassle than anything else for all involved parties, then no, I don't think that's something worth clinging to because we've done it this way for the last thousand years.

I think that at some level an A or B indicates something related to knowledge and capabilities. My personal experience is that tests are an effective method of determining who should proceed and who should not. It's not fair to other students and teachers if a student takes Calc II without knowing Calc I. Some mechanism is needed to weed out those who shouldn't move on.

Lectures do work. They have been working for hundreds of years. This isn't an appeal to tradition. This is a statement about evidence that they work. Lectures work so effectively that people want to put them online so anyone can watch them.

But guess what happens with students in online classes? My anecdotal experience is that they won't watch the lectures. They won't sit through a 20 minute lecture on how to solve quadratic equations. (Except a core group of good students.) The passing rate of students in online math courses at my college is lower than for the face-to-face courses.

But that's not a counterargument. Your simply restating your own point, which I disagree with. It's pretty obvious that someone who would score well in an artificial testing environment A would score well in such an environment B if B is a followup of A. That doesn't disprove the point that grades are a fundamentally flawed concept in education.

>My personal experience is that tests are an effective method of determining who should proceed and who should not.

Again, based on what? Conformance to a completely artificial testing environment? Again, tests have nothing to do with reality and are a flawed concept. If anything, they show if you can work under an extreme time constraint with limited resources available. Are we really basing success or failure in all of our education on this limited and arguably pretty unimportant skill-set?

>Lectures do work. They have been working for hundreds of years. This isn't an appeal to tradition. This is a statement about evidence that they work.

You're still ignoring the context, and that is that we have a better alternative to physical lectures now. Let me present an analogy:

"Books work. They have been working for hundreds of years. This is a factual statement about evidence that they work."

While principally true, that doesn't change the fact that printed letters on dead trees are no longer a viable model of distributing knowledge. It's an appeal to tradition because you ignore the context at hand. Your only argument is "we've always done it that way, therefore it must be right, why change it?".

>Lectures work so effectively that people want to put them online so anyone can watch them.

What are you trying to argue with this? Yes, online lectures are better than traditional lectures. To extend on the book analogy, while it's fundamentally still the same as reading a physical book (you're parsing words), an electronic representation of text is in every way superior to a printed book. The same holds true for online lectures.

>They won't sit through a 20 minute lecture on how to solve quadratic equations. (Except a core group of good students.)

Let me present you with my anecdotal experience as a student myself: Next to no one but a core group of good students will sit through any physical lecture unless it's absolutely mandatory to attend, either. As I've put it in another comment, if you're not at the university because you want to learn, what the fuck are you doing at the university?

The main problem with his argument is not that the "testing environment" is flawed. He is talking about learning as if it were a race, and suggesting that only the hares should be allowed to run it.

Often those who do extremely well in math classes have a pretty good head start, due to circumstances that are not a result of their abilities, or even their own choices. And, of course, such students will do better later; they can maintain a more comfortable pace than those who are racing behind b/c they are trying to make up for their late start.

It also disfavors those who are distracted not by parties and social scenes, but distracted from the main course by branches of the subject under study which catch their interest. That is, it favors the compliant student who is after good grades, and disfavors those who desire knowledge, skill and understanding for their own sake, or for any purpose not foreseen by those who designed the "course".

The great benefit to these on-line study programs is that it serves the interests of all of these kinds of students. Neither necessity, influence, affluence nor preference need induce the educators to favor one over the others.

I can give them a test and when they get an 'F' on it there is no argument. The test is a good measure of who knows the subject and who doesn't. For the most part. It's the most effective tool that I know of.

The absolutely mandatory part is the crux of the issue with regard to face-to-face classes. It forces students to meet at a set time and work on a given topic. In the absence of this forcing mechanism human nature takes over and most students fail miserably at the time management aspect of online classes. This is all anecdotal though. I don't have any research to back this up.

But does this have to be the case? In the current educational environment, I would agree that a number of students benefit from being made to attend class. The question is if this is just a result of human nature, or something else. Where that something else could be having students go to college when they are not ready for it, or high schools and earlier killing students intrinsic desire to learn, or some other reason.

Really? I give non-compulsory lectures, and get > 80% attendance rates. I also encourage audience participation, and get it. I think my lectures would be less useful via the internet. Not useless of course, but not as useful.

>No, no its not. Its to show that you've learned the material.

I disagree with both of you. It's to motivate the student to learn.

In my opinion, if your not at the university to learn first and foremost, then what the fuck are you doing at the university in the first place?

People, by and large, need artificial encouragement to help them do what they want to do in the first place, but lack the willpower or procrastinate too much. People subscribe to classes so they can learn in a formal setting material they could perfectly well study on their own before, but didn't. People take out gym membership so they have no excuse not to go. People ask relatives and friends to help force them prepare for something important. You may be a champion of iron will (though I doubt it), but most people aren't, and they're still quite capable of learning nevertheless.

In the recent ML/AI course offerings from Stanford, the exercises were a joke - they were painfully easy. But if they didn't have graded exercises with deadlines, probably five times fewer people would have finished the course. In fact, similar courses with entire lectures and homework have been posted on the web for years, with little engagement, while this new offering, with a formal setting that graded you on exercises, has seen tremendous success.

Your rhetorical question is exceedingly naive.

The contrived setting of the classroom is much better than an online setting. This may change as technology gets better. Right now, online is not a better format than face-to-face. Online might be the most cost effective format but it isn't the pedagogical optimum.

I don't think you've made a compelling argument on that score. It is precisely because life is complicated that lower cost and more flexible method might be considered optimum.

There are however a lot of students who see academics as something of a game and measure their achievement by score, the fact that they may learn a few things along the way is secondary to their desire to be able to say "hey , look at me I'm a high achiever please pay me lots of money". These people would probably be lost without a grading system of some kind and a lot of them are actually very smart people who probably should be at university.

And while this more of a philosophical point than anything, I think that this form of competitiveness ("look, I'm better than you") is a major problem of today's society. Education should teach to work together, not against each other.

Experiments in reducing the grade-centricness of universities have mostly foundered on employer objections. UC Santa Cruz for a while was famous for giving narrative evaluations instead of grades. Clases were formally graded pass/fail to indicate whether you met the minimum requirements to move on, and then in either case the professor wrote a one-page or so explanation of what you did well and poorly in the class, some advice, etc. In a sense the evaluation was still quite rigorous, just not boiled down to a single letter.

Companies hated this because they wanted to know which students had the 4.0 GPAs, and which had the 2.0 GPAs: a string of passes plus a stack of narrative reports didn't fit their evaluation model. So UCSC moved to allowing students who wanted to receive a grade along with the evaluation to opt into an A/B/C/D/F grading model alongside the narrative evaluation; eventually they made that mandatory; and finally just dropped the narrative evaluations.

You would probably just have recruiters who would grade candidates on the basis of how many times the word "excellent" appeared on the report or similar, no doubt startups would pop up offering machine learning type solutions to these problems as well.

Not to mention it would be easier to corrupt the system as there would be more plausible deniability if a professor simply "forgot" to include some negative factors on the report.

Much easier just to filter numerically based on grade. Also most of the smart people who do enjoy the learning as an end in itself should know the material well enough to do well in an exam anyway and just view the process of revising for the exam and tweaking their coursework for good marks as "playing the game".

I'm not convinced that academics are better qualified to predict the needs of society regards knowledge and education any more than the free market is.

Regards to helping students evaluate themselves , there is no reason that a professor cannot provide feedback in addition to assigning grades indeed I'm sure many would be happy to and it's more likely a case that many students are simply happy to get their grade and move on rather than evaluate themselves.

On the contrary, I'm pretty sure the free market is the last thing qualified for that. It will always be a race to the bottom, and that's not something education should be.

Would Stackoverflow.com have the participation it has if it didn't have badges, numbers, and karma? I think your perspective ignores a great deal of research on games and human behavior.

And that's exactly the problem with the entire system. We need other ways to give incentives and motivation. The point is that plain numbered grades suck at giving motivation, they suck at actually judging the performance and/or understanding of a student and they suck at being anything but a completely arbitrary method of weeding out people who are unable to conform to the contrived requirements of an artificial testing environments, whether they are actually intelligent or not (and being a slow learner doesn't mean someone is dumb, another thing this system sucks at).