Monday, May 21, 2012

Online education and the latest hype from Thomas Friedman


upload.wikimedia.org
by Massimo Pigliucci

Americans claim that education is one of their top concerns every time they are polled before an election. Yet, most of them don’t take education seriously at all. This is obvious from a series of elementary facts: a large portion of them thinks that creationism should be taught side by side with evolution; the public education system — one of the fundamental backbones of any democracy — has been under relentless attack and has been in the process of being systematically demolished for decades; and many American parents don’t really seem to think that “educator” is a real profession, since they are convinced that parents ought to dictate what is and is not taught to their kids, and how.

Enter New York Times columnist Thomas Friedman, who never misses a chance to write enthusiastically  about new technologies and the flattening of the world economy — even when the technology in question is of doubtful use, or when it becomes clear that there are at least some down sides to globalization.

In his “Come the Revolution” piece, Friedman waxes poetical about Coursera, a new Silicon Valley backed company that will take “the next step” in the daring world of online so-called education. Friedman approvingly quotes Andrew Ng, associate professor of (surprise!) computer science at Stanford: “‘I normally teach 400 students,’ Ng explained, but last semester he taught 100,000 in an online course on machine learning. ‘To reach that many students before,’ he said, ‘I would have had to teach my normal Stanford class for 250 years.’” Except, of course, that lecturing to 100,000 students has precious little to do with “teaching.” Teaching — at its best — is an interactive experience between a necessarily small number of students and someone who has both knowledge to impart and an effective way of imparting it.

Lecturing — which is all one can do with 100,000 students — is just about the worst way to teach ever devised. It’s problem are well known in the pedagogical literature, which I guess would be too much to expect Friedman to have read before lecturing us on how to get a better, faster, cheaper education. To give one-way talks to an audience (which is what lecturing is about) is an effective way to communicate a large amount of information to a large number of people. But communication represents a small fraction of what teaching is about. Real teaching must include guided discussions, interactions among peers, and a great deal of exercises. The ideal model is that of the Renaissance workshop, where one learned from the Master and his best assistants, day by day. In modern education, this is what is done in the best graduate schools and when using the Montessori method. A far cry from what Friedman and Silicon Valley are proposing.

Showcasing his typically bombastic prose, Friedman boldly claims that “Big breakthroughs happen when what is suddenly possible meets what is desperately necessary. ... the need to provide low-cost, quality higher education is more acute than ever. ... getting a higher-education degree is more vital than ever.”

Right, except that low-cost and high-quality usually don’t go hand in hand, unfortunately, and particularly so when they involve unavoidable human labor (as opposed to mechanized one). Yes, there is no doubt that the cost of higher education has exploded at an indecent rate over the past several decades, rising even faster than health care (the other thing Americans claim to be on their list of top priorities, but about which they keep doing next to nothing). This has generated an absurd situation in which total student debt in the US is actually higher than total credit card debt. But the reasons for this sorry state of affairs are multiple and complex — and therefore don’t fit into Friedman’s hyper-optimistic view of the future (I think that’s a general malady affecting techno-optimists, but that’s another discussion).

Contra popular perception (especially among state legislators), the problem hasn’t been caused by over-inflated salaries of university professors who go home at 2pm to mow their lawns. Rather, education has been commodified and made into a for-profit enterprise (even public education), just like a lot of other things that shouldn’t necessarily be so treated. University administrations have become bloated and more powerful and self-serving, offering stratospheric salaries to people with Wall Street-like management skills even when they have no idea of what they are managing or, apparently, why. Not to mention that many state schools nowadays are only so-called, since states throughout the country have been slashing their higher education budgets while keeping their political power over universities’ boards of trustees.

And let’s look for a minute at that “getting a higher-education degree is more vital than ever” bit. This is increasingly true only because pre-college education has largely become a joke (again, for a variety of reasons, one of which surely is the wrong type of parental involvement in what and how teachers teach, and — again — the dearth of proper resources, including well trained teachers. Yes, misguided union policies are also part of the mix, but only part.). Let’s be frank, Friedman: a good number of college students should never have gotten to college, and they get out of it with a minimum ability in reading, comprehending and writing. But the answer to the problem isn’t to push for higher and higher degrees (a nation of PhD’s, anyone?). It’s to go back to square one and restructure elementary, middle and high school education so that our kids can enter not just the workforce, but society at large, with the necessary critical thinking tools. Ah, but that ain’t gonna be done by simply putting a bunch of lectures on line, is it?

Friedman is so enraptured by his own pindaric flights that he doesn’t seem to notice pretty glaring problems intrinsic in his dream scenario: “[Coursera will be] awarding certificates of completion of a course for under $100. (Sounds like a good deal. Tuition at the real-life Stanford is over $40,000 a year.)” Yeah, sounds like too much of a good deal, no? Does anyone really think that a Coursera certificate will represent anything like the sort of education that one gets at Stanford? (On the other side of the equation, does anyone really think that Stanford-type education is worth $40,000 a year?)

And here is another issue: “[Coursera] operates on the honor system but is building tools to reduce cheating.” The so-called honor system doesn’t work in an environment in which students are constantly told that they are in competition with the rest of the world, where the goal is not to further one’s education, but to beat the crap out of all your competitors — just like on Wall Street, or in a video game. At any rate, I have experience with teaching online courses and assigning online tests, and I can tell you that if you can devise a system to detect cheating, the cheating industry (yup, there is an industry for that, and it’s legal!) will design a way around your defenses, and so on in an escalating war that has increasingly less to do with anything resembling education. At any rate, I’d like to ask Prof. Ng what sort of assignments he gives to his 100,000 students. Certainly not long papers that he would have to carefully read and then thoughtfully comment on — that would really take him centuries. Unfortunately, the evidence does show that the only things that significantly improve student learning are courses that are reading and writing intensive. Oops.

Does Friedman see local, brick & mortar colleges heading for extinction? Not at all: “these top-quality learning platforms could enable budget-strained community colleges in America to ‘flip’ their classrooms. That is, download the world’s best lecturers on any subject and let their own professors concentrate on working face-to-face with students.” Setting aside the bizarre qualification of what Coursera and similar companies are doing as “top-quality,” Friedman here can be read in one of two ways: on the one hand, he may be suggesting that community colleges (but why not Stanford itself?) turn into glorified tutoring companies. I doubt that’s what he meant. On the other hand, he may be saying that the real business of colleges is to teach via personal, continuous, and varied interactions between students and teachers. That’s correct. We call that education, and it would help if people took it seriously.

47 comments:

  1. It is amazing how misinformed the author is about on-line education. Apparently he assumes that all it consists of is video taped 'lectures!' As to cheating, there are a considerable number of tools to avoid this. First time, and probably the last, to read this blog as it is hopelessly ignorant.

    ReplyDelete
    Replies
    1. One thing that I'm guessing "My Head is Flat" (as I call him on my blog) Friedman ignores is ... accreditation.

      People made such a to-do about MIT doing so many online courses ... then noticed that none of them is for credit. And per his note on "certificate of completetion," yeah, Teapot Tommy, I'm sure that's going to look really impressive on a resume.

      I'm surprised he's not volunteered himself yet to be the Americans Elect presidential nominee.

      Delete
    2. This is also why the NYT doesn't have a real paywall, rather than the fake that's easily defeated by NYClean or a Greasemonkey script. Friedman might find out that not a lot of people would pay to read him (or Bobo Brooks).

      Delete
  2. Are you really that indifferent to the possibility of mass online education? Obviously, there are many problems left to tease out, but it's definitely something I've been very excited about.

    I've taken courses at Udacity, and found them extremely rewarding, and a great learning experience. One of their goals is actually to get rid of the 'lecturing at students for an hour' format, in place of something much more intereactive. At Udacity, the videos are kept to just three or four minutes in length, with programming quizzes in between. This is a much more active and intimate experience than most of what is currently on offer. In the sciences, automated grading for numerical and multiple choice answers could work just fine, and when you couple that with an active community of online students, there is lots to be gained.

    Humanitites will probably struggle to get as much out of the internet, and accreditation from these place isn't likely to mean much for the time being, but still, don't you find something exciting about an innovation where thousands of people around the world have the chance to learn subjects at an advanced level from excellent teachers at little to no cost?

    ReplyDelete
    Replies
    1. No, because it's not really learning. You talk about automatic "interactions" as if they are equivalent to face-to-face time with a professor, and that's simply not the case. It IS exciting that so much information is available to so many, but the internet has been around for a while and it hasn't solved all of our education dilemmas yet. What we need is to develop a deeper, more systematic method for teaching our teachers-to-be, thereby increasing quality of education, not just availability. Like it says in the blog, we need to restructure education, not just make it cheap.

      Delete
  3. When talking about online education, it's too easy and tempting to generalize across all subjects. Some subjects are well suited to be taught online, and some are not. I'm often surprised that this point isn't talked about more.

    For example, I don't doubt that one can learn how to program computers solely through an online education. Online education seems to offer an effective and low cost approach for those wanting to learn technical disciplines. However, an education need not necessarily include learning technical disciplines -- education is something different.

    In my opinion the humanities cannot be taught online and it's more important than ever to give our students the proper exposure to them. The techno-optimists who think that a classroom based education is overrated, and too costly, have failed to realize that ignorance cannot be measured in dollars.

    ReplyDelete
  4. I think you're right, Massimo. I view lectures like reading books. They're great to learn basic information, and to find where your interests are. But actual learning and mastering a subject involves interaction and feedback that you'll never get from these types of online lectures. I can see them being useful for the interested layperson, but we shouldn't kid ourselves that they pass for a "real" education.

    ReplyDelete
  5. Before you dismiss online teaching out of hand, spend some time looking at the Khan Academy.

    Yes, there is a lot of unsupportable hype going around about online education, and also some very sinister aspects in the for-profit sector. But I do think that technology may, over the next couple of decades, be completely disruptive and "disaggregate" the current functions of education.

    The four main functions as I see them are 1) providing quality content, 2) motivating students to study it (including at a very basic level, eg "pay attention, no texting in class"), 3) providing feedback to assist student learning, and 4) assessing how much the students know. Except that fourth function should be broken down into two parts: 4a) content-free signaling that a student good at learning one arbitrary thing will likely do well at something else too, and 4b) has the student acquired skills and knowledge with true practical value, from machine operation through to coherent writing.

    The linkage of assessment 4) with "teaching" (1-3), especially in an era of student evaluations, is intrinsically prone to grade inflation and other forms of corruption, so ultimately this disaggregation would probably be a good thing. Then we could maybe have a proper conversation about the true purpose of education and how to assess whether it is successful or not, and construct assessments accordingly.

    Math-intensive subjects provide a remarkable opportunity to partially automate function 3). Clever uses of technology for this purpose are finally being developed, and should be welcomed. Unfortunately, reading and writing etc. do not lend themselves to such automation. However, online education does offer possibilities for peer-to-peer interactions that could reduce the total cost of education via the professor's time. Ultimately, 3) requires as much one-to-one feedback as possible, which is expensive from a teacher/professor, but has negligible cost if computer-automated or peer-to-peer.

    While 3) needs expert knowledge, it may need less expert knowledge than 1). For example, in my own field of population genetics, I WISH that subjects like Hardy-Weinberg and genetic drift were taught only via recorded Khan Academy style snippets, only done by true experts. Khan unfortunately gets some stuff about Hardy-Weinberg wrong http://www.khanacademy.org/science/biology/v/hardy-weinberg-principle, although no more so than most textbooks. See http://onlinelibrary.wiley.com/doi/10.1002/bies.201100178/abstract for my thoughts on this curricular matter, but I mean it just as an example.

    Given the economics, online education is coming. Those of us who care about quality education need to engage with the conversation, and not dismiss it out of hand. There are good opportunities to embrace as well as appalling pitfalls to avoid.

    ReplyDelete
  6. Check out http://www.forbes.com/sites/joshuagans/2012/05/07/what-my-11-year-olds-stanford-course-taught-me-about-online-education/ for a really thoughtful blog post about where online education via Stanford courses currently stands.

    ReplyDelete
  7. The literature may *say* that lectures are ineffective, but it's pretty apparent that in the courses I take (grad-level physics), they are the only remotely realistic option. I'm inclined to think that the literature either focuses on a few subjects, or focuses on introductory courses, and then extrapolates beyond where it should. Not having read the literature, it's hard to tell where the evidence ends and extrapolation begins. Admittedly, the online courses in question are likely intro courses, where the literature may be applicable after all.

    Similarly, it is also obvious that reading and writing are not the best way to learn every subject. Sounds like more unjustified extrapolation. I think the subject in the news story was computer science? Clearly, not writing, but coding is the way to go.

    ReplyDelete
    Replies
    1. Miller,

      You should not be inclined to think either way on this matter if you are ignorant of the relevant literature and/or have not conferred with those who are knowledgeable on these matters.

      Delete
    2. If I'm inclined to believe neither way, that would mean Massimo failed to persuade, would it not?

      Does Massimo count as someone who is knowledgeable in this matter? If so, he may assert that the literature indeed does not rely on extrapolation, and I would believe that on his expertise. He may assert that writing intensive courses are the only effective courses even in math, physics, and computer science, and I might believe that too. He did not make these assertions, instead treating them as hidden assumptions. Which should I accept, personal prejudices, or unacknowledged assumptions?

      Delete
    3. Miller,

      I did intend my comment to be taken as contentious, so apologies if it caused any offense.

      Re: "If I'm inclined to believe neither way, that would mean Massimo failed to persuade, would it not?"

      No, not exactly. In general, one should rarely be persuaded by one source of information, whether that source is prima facie reliable or not. If one has no good prior information on a subject, one should remain agnostic and resist forming a view subject until one has surveyed the best relevant information. Massimo's piece could be a good starting point, but whether it is or is not, I cannot say.

      Re: "Does Massimo count as someone who is knowledgeable in this matter?"

      I do not know. His field of expertise is indirectly related to the issue, so that's an initial strike against him. However, except for his views on political and moral theory, his views on matters tend to be informed, so this counts in his favor. But this initial reasoning is all for naught unless one delves into the literature in more detail.

      Re: "Which should I accept, personal prejudices, or unacknowledged assumptions?"

      Neither. The epistemic point is simple and it is this: One should apportion one's beliefs according to the best available evidence, and if one does not have any evidence on a matter, one should remain agnostic about issues pertaining to that matter.

      E.g., I have no opinion one way or another on the efficacy and social benefits / costs of online courses. My anecdotal experience is that such courses are suboptimal for learning, but I do not place too much on that.

      Delete
    4. P.S.

      Miller,

      So, rather than disagree with Massimo (which implies that you are privy to some relevant information on the matter), you should have inquired into how, why, & upon what basis he formed his view, which amounts to asking him questions, not making assertions identifying your own ignorance.

      I should add that I only picked your comment out for reply because it was the most conspicuous (i.e. it contained this sentence: "Not having read the literature, it's hard to tell where the evidence ends and extrapolation begins.") and it embodied a nasty habit all too prevalent: forming beliefs and views on matters without any good reason, which can only serve to bias one in the subsequent consumption of evidence.

      Delete
    5. Eamon, I think I am arguing with Massimo through you, and I should instead be addressing him directly.

      Delete
  8. First of all, I find it strange that several people thought I was dismissing out of hand the entire idea of online education (I'm looking at you, Joanna, who should have known better — more about your comments below). My criticism was specifically tailored to the hype in Friedman's NYT article.

    SBissell,

    > First time, and probably the last, to read this blog as it is hopelessly ignorant. <

    Good riddance to you, sir. You clearly did not pay any attention whatsoever to what I was writing, and obviously have nothing to contribute to the discussion.

    Alan Go,

    > Are you really that indifferent to the possibility of mass online education? <

    No, I'm not. I'm just weary of the hype, which will play particularly well with the deadly combination of push for profit by private companies, push for budget cuts by politicians, and push for getting more out of faculty by university administrators.

    > In the sciences, automated grading for numerical and multiple choice answers could work just fine <

    Yes, has Greg has pointed out, some disciplines lend themselves better than others to automated grading. However, even in the sciences, the student needs feedback, and the teacher who relies on easily automated grading risks of lowering the standards significantly.

    > don't you find something exciting about an innovation where thousands of people around the world have the chance to learn subjects at an advanced level from excellent teachers at little to no cost? <

    Yes, but first of all that doesn't translate necessarily into degree-worthy education. Second, the evidence shows that online courses are best for students who are already highly motivated and independent. That is far from the majority of students, unfortunately.

    Joanna,

    > Then we could maybe have a proper conversation about the true purpose of education and how to assess whether it is successful or not, and construct assessments accordingly. <

    How about we have that conversation now, before the "revolution," rather than afterwards?

    > Math-intensive subjects provide a remarkable opportunity to partially automate ... Unfortunately, reading and writing etc. do not lend themselves to such automation. <

    Agreed, though you may be overestimating even the automation that can be done for math teaching.

    > online education does offer possibilities for peer-to-peer interactions that could reduce the total cost of education via the professor's time <

    Sorry, I don't believe in peer-to-peer "education." Discussions and feedback with your peers is necessary, but nowhere near sufficient. If it were, why have a teacher to begin with?

    miller,

    well, if you have not read the relevant pedagogical literature, how can you be so sure that it was a matter of bad extrapolation? And anecdotal evidence (such as your experience in graduate school) doesn't really get you very far (not to mention that I'm surprised a graduate program relies that much on straight lecturing). And no, the NYT article did not refer specifically to computer science, it simply quoted a computer scientist as a source of support for Friedmnan's ideas.

    ReplyDelete
    Replies
    1. Massimo,

      Just to be clear, the bit about your views on political and moral theory being uninformed was a playful jab.

      Delete
    2. Massimo,

      I said the graduate courses rely on lecturing, not the entire graduate program.

      You said that lecturing was "just about the worst way to teach", and that only reading and writing intensive courses significantly improve student learning, and you did not add any qualifications about subject matter or subject level. I used an anecdote to show why I find this incredible (but very interesting if true!). I am genuinely curious if I am wrong.

      Therefore, I find it frustrating that in your response, you can't even be bothered to make a statement of whether the pedagogical literature uses extrapolation, nor be bothered to say whether writing intensive courses are in fact effective in computer science. Could you clarify this issue?

      Delete
    3. The pedagogy literature would have greater weight if a) there was any decent agreement about what constitutes "student learning" and how to measure it, and b) there were randomized trials that used such metrics of learning as an endpoint. Until then, the evidence base of the pedagogy literature is extraordinarily weak.

      Delete
    4. Does anyone here read anything (I'm fairly certain Massimo does as I've seen him publish in Science & Education) from The Journal of Research in Science Teaching, Science Education, or Journal of the Learning Sciences?

      Delete
    5. @Stewy: I know, right? So annoying reading these arguments that start out "I've never read any pedagogical studies, but I learn just fine from lectures so lectures work awesome" and are coming from *scientists.*

      @Joanna: Point (b). Please explain how you can pull off a randomized trial of a college-level pedagogical technique. I'm guessing students at your school are also allowed to choose their own course schedules, just like they are everywhere else.

      #self-aggrandizement: Here's an article (look! from the pedagogical literature! which exists, and is a pretty rich field of study, FYI!) about the difficulties in randomizing students in pedagogical studies, and how pedagogical research is similar in this way to clinical research:
      http://jmbe.asm.org/index.php/jmbe/article/view/360/html

      Delete
    6. Nice article Rachel! My feeling is that both physicians/medical researchers and educators/educational researchers tend to overestimate their confidence as to what works best. I base this on a long history of being wrong in medicine (I recommend the book "Taking the Medicine") and on likely universality of human nature in the parallel case of education. Overconfidence systematically creates a tendency to see equipoise objections where none, in fact, exist.

      I actually did randomize some of my college students once. I taught a large undergrad Genetics class, which was divided into 6 TA-led discussion sections spread across 3 TAs. I wanted to test the benefits of heterogeneous vs. homogeneous groups for small-group discussions. Unfortunately, my group allocation method was hopelessly flawed: I used prior GPA, which turned out to correlate very badly with success in Genetics. It did correlate well with attendance at discussion sections though: heterogeneous groups consistently had 1 member missing, while homogeneous group consistently had 1 entire group meeting, so the whole thing became confounded with group size on top, and I didn't even bother to analyze the data. But still, this is an example of how I pulled of the randomization, albeit at a large research university where few of my colleagues would care.

      Delete
    7. I meant "1 group missing" not "meeting". And just to clarify, students picked one of the 6 sections to fit their schedule, then within 3 of those sections I assigned students to heterogeneous groups, and in 3 I created homogeneous groups. Each TA led one section of each type. N=the total number of students in the class for most purposes, with the multiple sections helping control for time of day and other possible pseudoreplication factors.

      Delete
    8. @Joanna: Thank you!

      That's really cool that you were able to run a randomized study, and I'm sad that you don't have the data analyzed. Still, I would love love love to see the results that you do have. I'm trying to do a similar study this coming academic year. I have four lab sections of a 200-level microbiology course, that's open to biology majors and required for nursing majors. With kind of stunning levels of administrative help, I have two homogenous sections and two heterogenous sections (kind of: one section is 100% biology majors, one is 100% nursing majors, two are mixed nursing and biology majors). These four lab sections are tied to two separate lecture sections. Because this is a class that sells out routinely, students already have to get signed in by a person in the registrar's office (we can do this, we're a tiny college). So the registrar was randomly assigning students to sections when they came in to enroll, within these constraints (um, and of course lots of other constraints, like if students had other conflicting classes or were in sports and such. So there's that.) But even though this class is one of the very largest on our campus, I'm still only going to wind up with 20-24 students per lab section, plus when you figure in drop-outs/nonresponses, it's probably going to be only about 15 participants per arm of the study. :(

      I completely agree that both educators and physicians can be way too confident about "what works." (I'd also recommend Atul Gawande's 2005 New Yorker essay "Desperate Measures" as a creepy human interest story about overconfidence and being totally wrong in adhering to equipoise.) But the solution to this is more research--which, weirdly, can only be done by nodding to people's equipoise objections, even if they're wrong. I think the alternative is the much worse overconfidence of "I know I'm a good teacher, my students get good grades and I can just feel they're learning so much" that was completely unchallenged for generations (and is still, I fear, the dominant way of looking at assessment).

      Anyway, would you mind terribly sending me what you have on your genetics class study? In the future, when I write up my only-sorta-randomized, N of approaching zero, microbiology study I'll totally cite you (as unpublished data, but still).

      I'm at robson@morningside.edu

      Thanks in advance! :)

      Delete
  9. > How about we have that conversation now, before the "revolution," rather than afterwards?

    Massimo, my point is that a successful redesign of assessment would profit enormously from the disaggregation of assessment and teaching. Otherwise, the assessment tends to become corrupted by the motives of the teacher, which are different from the motives of the assessor. Conversations about assessment will look very different "after the revolution" if that revolution includes disaggregation. Entrenched interests make that conversation hard to have in the absence of disaggregation. I do not believe that assessment comes cheap, BTW, in fact I think it should be done much more thoroughly and expensively than is sometimes currently the case (eg multiple choice tests). I think the primary cost savings in assessment should come from doing it much, much less frequently. But multiple choice tests may turn out to be great for automated feedback, directing students to content targeted at addressing specifically identified misconceptions, even as the same tests are lousy for assessment. We need to stop confusing the two aims.

    > Sorry, I don't believe in peer-to-peer "education." Discussions and feedback with your peers is necessary, but nowhere near sufficient. If it were, why have a teacher to begin with?

    You are taking my comments out of context. I reject the whole idea of a monolithic concept of "education". I agree that peer-to-peer needs to be kept far, far away from delivery of quality content (#1 on my list). But in the feedback department (#3 on my list), it can often be great. One of the best ways to learn is to explain something to someone else, and realise in the process which part of it you don't quite get yourself. As the listener is puzzled, the speaker needs to work harder to be clear, and in the process can reach insights. This doesn't require a teacher, it can work just fine peer to peer. Other types of feedback do require a teacher, eg I don't expect many qualified peer volunteers to provide quality feedback on spelling, grammar, and essay structure. That is no reason not to work to reduce costs by disaggregating each component and delivering it in the cheapest appropriate way. Content may come via recorded snippets from experts, a motivation coach without specialized content knowledge may help keep the student focused and ensure that they are in fact studying, peers may provide discussion forums in which students act as sounding boards for each other's ideas, and a writing coach with less expert knowledge of the topic may provide feedback on essay drafts. Finally, maybe a year or two later, a specialized assessor may examine the students knowledge of the subject and/or writing or other pertinent skills.

    ReplyDelete
  10. I guess my overall point is that rather than dismiss the hype, we should "surf" it, laying forth an articulate alternative vision of how technology can and should improve education while reducing costs. I don't think criticism, however well-grounded, is an effective response to the hype: it makes us professors sound just like conservative entrenched interests.

    ReplyDelete
  11. Good article Massimo. Quick question could you point out to me books on the philosophy of education, and/or how to reform America's flawed educational structure?

    ReplyDelete
  12. I very much agree with this post. A reasoned criticism like Massimo's is probably the only way to get the positive aspects of online teaching discussed without the all-or-nothing thinking that seems to be so pervasive in the US these days. (That's the problem with hypes - you can't really "surf" them at their peak.)

    When Friedman writes "At the same time, in a knowledge economy, getting a higher-education degree is more vital than ever", he shows exactly the (systemic?) problem: the degree is actually not vital at all, the education itself is. (The degree is just supposed to be proof that you had the education, yet that seems to have become lost at one point.)

    When a college degree become mandatory, it is no surprise that you have a lot of unmotivated students. Here in Germany, vocational training is a perfectly acceptable alternative to university - the demand for craftsmen is phenomenal at the moment. (Our universities have their own problems, but just using my own scores from the US and here as a comparison I dare say that our quality level is higher - at least it was 20 years ago.)

    ReplyDelete
    Replies
    1. "When a college degree become mandatory, it is no surprise that you have a lot of unmotivated students."

      I've thought about this problem quite a bit in the past and had trouble thinking of a satisfactory solution. Vocational training could be part of the solution (for those not interested in a general education), but usually that is something very different than an education at a college or university. I think in the US this alternative not quite as acceptable, but I think the bigger obstacle is that a person must decide to learn a particular occupation. Going to college or university can be a socially acceptable way of delaying a decision, but this is problematic and expensive for those uninterested or not ready for this type of education.

      I guess part of the problem is that too many people don't value education outside of its impact on future employment. This can be illustrated with the question -'What can you do with that?' a common response to a student declaring a major. Given the huge increases in cost of education, I assume that this type of thinking will only increase.

      Delete
  13. Eamon,

    > Just to be clear, the bit about your views on political and moral theory being uninformed was a playful jab. <

    Thanks for clarifying that, sometimes my sense of humor is under par, especially before caffeine... ;-)

    Miller,

    > Does Massimo count as someone who is knowledgeable in this matter? <

    Somewhat knowledgable, as someone who has been teaching graduate and undergraduate courses for quite some time, reads at least occasionally the pedagogical literature, and has even published a few papers in that field.

    > He may assert that writing intensive courses are the only effective courses even in math, physics, and computer science, and I might believe that too. <

    That statement was a reflection of the findings in the book "Academically Adrift," which is linked in the post. Obviously it doesn't apply to math (or logic) courses. Though their equivalent presumably do: try learning math without intensive written exercises, or without spending a good amount of time studying the material.

    > I said the graduate courses rely on lecturing, not the entire graduate program <

    I understand, my comment was aimed at the courses. Not sure how it could have been interpreted otherwise.

    Joanna,

    > my point is that a successful redesign of assessment would profit enormously from the disaggregation of assessment and teaching. <

    Maybe, though that seems to me an entirely different conversation from the one Friedman started and on which I was responding. He certainly didn't even bring that point up.

    > You are taking my comments out of context. I reject the whole idea of a monolithic concept of "education" <

    I don't think I did. I have serious concerns about peer-to-peer in the specific context cited by Friedman. Yes, it does have a place in the sense you discuss, but even so I doubt it is very effective without direct teacher supervision. After all, peers can simply explain things wrongly and convince other peers that they know what they are talking about.

    Unknown,

    > Quick question could you point out to me books on the philosophy of education, and/or how to reform America's flawed educational structure? <

    I would start by checking out the above mentioned "Academically Adrift," but the literature in this field is huge. Here is another large resource:

    http://www.everythingphilosophy.com/philosophy-of-education/

    ReplyDelete
  14. I teach online. I know it's not all about lectures. I also know that students walk away with less learning than they do in an on-site class, and retain it for a shorter period of time.

    Yes, online education fills a gap, and online lectures are a good thing. But they do not substitute for a quality education. The author makes good and relevant points here. We are talking about apples and oranges. Friedman is up to what he's always up to: blurring over critical distinctions to come up with 'bold' statements that sell books.

    ReplyDelete
  15. The bottom line is that online education is here to stay. You can't stop technology. If you want to survive, you better learn how to adapt.

    ReplyDelete
  16. Alastair,

    that has never made sense to me. Do you think "technology" has its own brain and willpower? Besides, there are examples of technologies we got rid of (the Concorde) or stopped using (the atomic bomb) because we thought better.

    ReplyDelete
  17. Massimo,

    You can't stop technological development. Online education is here to stay for the very reasons Friedman has put forth. I understand why some educators may feel threatened by that prospect. After all, their lifestyle (if not their livelihood) may be radically altered. But everyone here is in the same boat. No one is immune to technology and its impact on the work environment.

    ReplyDelete
    Replies
    1. Alastair,

      you have not addressed my comment, you are simply stating a generic article of faith. Besides, nobody is talking about "stopping technological development," we are talking about what to do with education -- not at all the same thing.

      Delete
    2. Massimo,

      You are right that we ought take tests in person, and you are right that the optimal method of teaching involves copious quantities of student-teacher interaction. However, just because a lecture is a live performance rather than a recording doesn't mean it is interactive to any significant extent. That all depends on the teacher and a great many, especially in more technical courses, rarely offer added value over a recording. To the contrary, since they do not interact, those who are additionally poor presenters are inferior to a recording of a good presenter. So, a recording often is better than the teacher you have, but this conclusion, though perhaps practically useful, doesn't threaten your depiction of an idealized education.

      What does is your seeming rejection of the idea that recorded lectures can and should provide a labor saving tool to allow professors to spend more time actually interacting with students and less time endlessly giving the same lectures over and over again.

      And, though that's ideal, how much is it worth? How much more do you learn from actually attending a calculus class than you do from just studying and learning the exact same material from a book?

      Teachers are expensive. That isn't some new thing, and there are plenty of other reasons why school is expensive, but you can't tell me that school wouldn't be far cheaper if those of us who are capable of learning from a book could just do so and get equal credit as those who learned differently but perform equally.

      The way things are organized is ridiculously impractical. It isn't about optimisation and efficiency. It is about momentum and entrenched interests.

      Delete
  18. F2F beats on-line, still. If this were not so, there would not be such capital and interest rushing to each new small step in videoconferencing. I agree pre-college does not minimally instruct much less educate, nor does college in general educate. There will never be an economic reason for a predominantly educated workforce: what would be the purpose? Education is the drawing out of the best one is capable of. This is a minority persuit.

    ReplyDelete
    Replies
    1. long number,

      but one of the issues here is that education is not just for the purposes of the workforce, we educate citizens of a democracy, not just workers for the industry. I am constantly amused by how especially Americans don't talk about citizens, but rather about taxpayers or members of the workforce. Life is broader than work and taxes.

      Delete
  19. And here is a report on a randomised trial comparing "traditional" statistics teaching to a hybrid traditional+automated approach, and the hybrid comes out well.

    http://www.insidehighered.com/news/2012/05/22/report-robots-stack-human-professors-teaching-intro-stats

    It's hardly the last word, but I think the key is to test every online course in trials such as these, and make each and every one of them justify itself. Given the extreme ease of shortcuts and shoddy online options, the onus is on the online option to prove itself, but that doesn't mean it won't (for some subjects, courses and implementations).

    ReplyDelete
  20. Many universities also see on-line education as a lucrative revenue generator and don’t want to miss the money boat. There is a bit of if-you-can’t-beat-them-join-them attitude.

    On-line education is doable on low level courses, and today’s kids learn differently. Just as traditional classroom learning, the quality depends on the instructors and students.

    ReplyDelete
  21. Massimo,

    Concerning your comment above: […] you may be overestimating even the automation that can be done for math teaching.

    I entirely agree.

    I have taught math at university level (that was some time ago) and, although multiple choice questions were becoming popular, I never used them in my exams. I felt the only way to really measure understanding was to look at the process leading to an answer and not at the answer itself. How a student thinks about a problem is much more important than the factual answer.

    There is also the fact that a stupid error can lead to a wrong answer while most of the reasoning is correct.

    ReplyDelete
  22. I found this a remarkably poorly-argued article, and I'm not a fan of Friedman at all.

    I only wish I could have emailed the author instead of publicly commenting. I have no intention of trashing his work, as it looks like a great deal of thought, effort, and no doubt studying went into it.

    ReplyDelete
    Replies
    1. Daniel,

      well, sorry to have disappointed you, but your comment isn't helpful at all, as it is. Feel free to elaborate on why the post is "remarkably poorly argued" (really? It's not just that you disagree? It's that badly written?). Before you do, make sure you read the comments above and my replies, you may get a better feeling for why I think the post is well argued...

      Delete
  23. Professor Pigliucci: I won't fault anyone for giving Thomas Friedman a good rap on the knuckles. He deserves it.

    What I took away from your post is that the education system has poorly motivated, poorly prepared students, and college has become a way of extending high school. In other words, college is remedial. And the management is ineffective. Correct me if I am wrong.

    I doubt if online education will fix the system. But it might provide opportunities to help those who are prepared and motivated. Which I think is what you pointed out.

    Cheers,
    Scott

    ReplyDelete
  24. Massimo,

    I'm skeptical about online education taking over the traditional college system anytime soon — and Thomas Friedman annoys me almost as much as he does Matt Taibbi — but I think you're underestimating the value of lecturing as you put it. I've learned a lot from Khan Academy and MIT Open Courseware (and the Rationally Speaking podcast!) I don't think they're in any position to test, grade, or certify anyone, but the lack of face-to-face conversation between teacher and pupil doesn't leave online education completely hamstrung.

    These websites have comment sections which take questions — the crowd can vote up or down questions — the teacher can answer the ones with highest votes. Not a perfect system, I know. But it will help many.

    You say, “Teaching — at its best — is an interactive experience between a necessarily small number of students and someone who has both knowledge to impart and an effective way of imparting it.”

    I agree. But must we doom people to choosing between no teaching and best teaching? There’s Stanford for those who can afford it — The classroom sizes at even the big schools suggest there isn’t much personal interaction — and for those who want to learn just something, there’s online education.

    Friedman seems giddy about this and jumps the gun — such as from Nandan Nilekani saying ‘the playing field is being leveled’ to ‘the world is flat’ — but I believe that in criticizing Friedman’s hyperbole, you’ve dissed online education more than necessary.

    PS: How about this topic for June’s NYCSkeptics meetup?

    ReplyDelete
    Replies
    1. liberalcynic,

      well, to been with, I said that I do lecture and I do teach online, so clearly I appreciate the value of both. I simply think that Friedman-type hype is misleading and dangerous because it downplays the problems of higher education.

      Second, the evidence is clear: motivated students do learn from online courses. But, frankly, motivated students will learn no matter what because, well, they are motivated! And unfortunately they represent a large minority of the people we want to reach.

      Third, of course few people can afford Stanford tuition. But I don't think *any* education is worth $40K, and the alternative is good, affordable public schools -- whose budgets are being cut and whose administrators often buy into Friedman-type rhetoric (we can do more! with less! and it's cool, because it uses computers!).

      Delete
  25. "But I don't think *any* education is worth $40K, and the alternative is good, affordable public schools -- whose budgets are being cut and whose administrators often buy into Friedman-type rhetoric (we can do more! with less! and it's cool, because it uses computers!)."

    Lets support public schools more overtly. Other than a few summer classes, all ~21 years of my formal education has been in public schools. Assuming that circumstances don't change too much it is likely that my (still very young) children will also be going to public schools through college. I agree that the cost of private schools is ridiculous, but I don't truly understand the nature of the problem.

    ReplyDelete

Note: Only a member of this blog may post a comment.