5 ms·
"...making the case for human intervention in teaching, at least for those students who are not 100% self-motivated and self-assured (that is, those students wh
by glesica 12y ago
"...making the case for human intervention in teaching, at least for those students who are not 100% self-motivated and self-assured (that is, those students who actually need to be educated, rather than left to educate themselves). Two relevant notes of today."
This is really important. The discussion about education in this community (HN), and the a lot of the discussions I've seen related to the MOOC movement (or at least where it has seen success), tends to focus on self-motivated, usually high-aptitude students. But, almost by definition, those attributes do not describe most students.
- visarga 12y ago> MOOC movement (or at least where it has seen success), tends to focus on self-motivated, usually high-aptitude students. But, almost by definition, those attributes do not describe most students I think MOOC's could benefit from face-to-face coaching. Students need someone to see them, to acknowledge their efforts, to monitor their progress and give them encouragement or advice when needed. This is the job of a coach, or educational counselor, not a teacher. The teacher can very well be teaching the course on video, but the student needs to be observed by a real person in order to get that special feeling that motivates and makes the whole process efficient. A bunch of counselors distributed to major cities could serve a lot of students taking online courses, but they need to feel accountable to a person, not just a website. It's easy to do that and fits well with the online delivery model. MOOC's could be a great solution to life-long learning and keeping up to date with relevant skills. Especially in computer science, a 4 or 5 year college program is too long - by the time the students finish, many specific technologies they learned might be already obsolete. Instead, coming back for a course or two every year might work much better in the long run. MOOC's might still work out if they could be coupled with employment from the start - let Coursera or Udacity partner up with various companies and create relevant courses, and use this opportunity to teach and rank the participants and in the end, offer jobs to the ones that are up to par with their skills. This would make MOOC's much more valuable than what they are now.
- glesica 12y ago"Students need someone to see them, to acknowledge their efforts, to monitor their progress and give them encouragement or advice when needed." I agree with this wholeheartedly. It is always important to remember that students are people and that most people are extremely social creatures who want to interact with other people directly, in a bidirectional, real-time way.
- clavalle 12y agoBut does that interaction have to come in the form of a rigid and traditional teacher to student series of interactions? Or could study groups fill that need?
- RyanMiller 12y agoIn 100 or 200 years from now, I think we can be certain that teaching won't even resemble what we do today. Not to mention just how little we've applied scientific research to modern education. We have a long way to go before we see how usable and powerful our tools for information become.
- glesica 12y agoI am not certain of that. Not even a little bit. I don't think teaching will be identical, but I don't see how it can change that much. Human contact is important, that necessitates a "teacher" of some kind. One-on-one instruction isn't efficient, that necessitates classes. I mean, teaching is basically the same today as it was 200 or even 400 years ago despite truly massive increases in technology. I don't see how another hundred years or so will really change human nature enough to necessitate a paradigm shift in teaching.
- jonnathanson 12y agoTrue. I think the focus on self-direction is a tangent. The real potential for MOOCs is individually tailored pacing and material at scale. Self-direction is all well and good for the students who are capable of such. But even for those who aren't, there's great promise in the idea of "custom" curricula and even "custom" pacing of standard courses. Our current education system is based on standards and Gaussian distributions. Some aptitudes appear to be distributed that way. But a lot of things aren't, like learning styles, speeds, comprehension methods, imagination, specific interests and aptitudes, and so forth. We can't dream of assigning an individual tutor to every student in the country. But we can dream of a world in which technology enables every student in a standard course to learn in a personally optimized way.
- malexw 12y agoI agree with the main point - that MOOCs work best for students who are self-motivated - but I think you're not giving the average college student enough credit. From my experiences as a student in both environments, I think the motivation required to successfully complete a college course is about the same as completing a MOOC. I don't think that an increase in the amount of content delivered online at a college level would increase the failure rates, because the students who aren't self-motivated would fail out anyway.
- mgkimsal 12y agoThe fact that you're here discussing this matter indicates that you're not 'average'. You probably don't associate with many 'average' college students.
- malexw 12y agoThat may be true, but that's not what I was trying to get at. My observation is that the external motivating pressures are the same, whether the material is delivered via traditional instruction or online. In both environments, there are lectures to attend, assignments or essays to hand in, and a final exam to write. I suspect that anyone who has enough motivation to complete that work in a traditional setting would also be able to complete it online.
- glesica 12y agoI strongly disagree. I have three college degrees but my completion rate on MOOCs is hovering around 10% right now. Obviously I'm a sample of one, but I just wanted to give a personal account to match your own. The classroom setting is important because it sets up social expectations and accountability. I tend to blow off MOOCs partly (there are obviously other reasons as well, including the fact that they are free) because there is no scheduled time. There is less pressure and it is up to me to make time, it isn't blocked out for me. I also get bored more easily in MOOCs because the communication only flows in one direction, from the teacher to the student. I can't ask questions or challenge assumptions during lectures. To me, that isn't as interesting, so I'm more likely to drop out.
- bobcostas55 12y agoYou're implicitly assuming that education can actually do something for the group that needs "intervention". There is very little evidence that university education actually does anything other than act as a signal for students' already-present aptitude.
- gaetan 12y agoI agree with you. In my startup (marketplace for face-to-face teachers) we see clearly that most of the students are not self-motivated and self-assured when in difficulties. It is the rigor of the schedule appointments, the expertise of the face-to-face teacher to adjust the learning progress and his direct feedback that are helping efficiently our students. MOOC’s are still experimenting on the best way to massively transmit knowledge and it will take time for them before being as efficient as a human teacher.
- dragonwriter 12y agoMuch of the discussion I've seen of MOOCs is how the format enables more interactivity by enabling learning where lecture and associated self-check exercises are delivered purely online, and interaction time with other live humans -- whether co-students, individually-selected tutors, or formal group classrooms with faculty instructors -- is used to provide direct Q&A and support on substantive assignments (which may also be delivered and graded online) -- and a significant subset of that discussion has centered around use of existing MOOCs or custom online courses built on their open platforms as support for "flipped classrooms".
- glesica 12y agoBut the best time, for me at least, to ask questions is during the lectures. That's when my mind is working to understand the material, so that's when I am most likely to have questions. The same process happens during labs and the like. I also need to be able to ask questions during lectures because if I fail to understand something it may negate the entire value of the lecture.