Slashdot Mirror
When Getting Rid of College Lectures Makes Sense
timothy writes "NPR reports that Harvard physicist and professor Eric Mazur has largely gotten rid of the lecture in his classes, after finding that in lecture-based classes, students tend to commit to memory formulae and heuristics, but fail to develop deep understanding of concepts. Mazur has tried — and seemingly succeeded — to cultivate deeper learning with a combination of small group peer-instruction and a tight feedback loop based on in-class polling about particular problems. Joe Redish also teaches physics, at the University of Maryland, and says, 'With modern technology, if all there is is lectures, we don't need faculty to do it. ... Get 'em to do it once, put it on the Web, and fire the faculty.'"
The slashot summary isn't terribly accurate, and even if you violate the social norms of /. and click through to read the article, the article is pretty sketchy as well. We're already getting comments from people who think this is about substituting video lectures for live lectures, and that's totally inaccurate.
This method is not new. I teach physics at a community college (not at Hahvahd like Mazur, alas), and I've been using methods similar to his for about 15 years. I learned about them from Mazur's book, which was published in 1996.
It's also not just some guy's opinion about how to teach. It's solidly backed up by research.
Let's start from the evidence. There is very strong evidence that lecturing is a terrible way to teach physics. The classic studies work like this. You give students a multiple-choice test at the beginning of the semester on very simple, basic concepts of physics. What hits the ground first, a larger rock or a smaller rock? What forces act on a book that's lying on a table? They do badly, but you expect that, because most of them haven't had high school physics. Then you teach a semester's worth of physics to them and give them the test again to measure how much they've improved. The usual statistic used to measure their improvement is the gain, G, defined as G=(final score-initial score)/(100%-initial score). In other words, if they haven't improved at all, G=0, and if they've improved as much as it was possible for them to improve, G=1. With classes that use traditional lecturing -- even by experienced, award-winning teachers who get glowing reviews from their students, are enthusiastic, and put a great deal of effort into their lectures -- you get about G=0.25. In other words, the students have developed very little conceptual understanding beyond what they came in with. On the other hand, if you use interactive teaching techniques that force students to participate actively and talk about concepts, you can usually get much higher G's.
The evidence is that it doesn't really matter very much what specific interactive technique you use, as long as it's interactive and deals with concepts. Mazur pioneered a technique called peer instruction. Just to be concrete, I'll describe his specific technique. You require the students to read the book *before* they come to class. You enforce this with reading quizzes given when they walk into lecture. The class consists basically of a bunch of multiple-choice conceptual questions. You pop up one of the questions on the screen and ask students to show you their initial opinion about which answer is right. This can be done with expensive electornic "clickers" or with cheap pieces of cardboard marked A, B, C, and D. If you see that almost everyone got it right, you briefly confirm that, and then move on. If they didn't, you have them break up into small groups and discuss the question. You walk around and listen a lot without saying much. Then you have them vote again again. The theory is that the right answer is supposed to win out over the wrong answers in the discussion. When it's time to give a test, you make sure that the test includes some purely conceptual questions, because otherwise students will tend to resist dropping the "plug and chug" approach they're used to and switching to focusing on concepts.
Mazur's book shows data where he got G~0.5 with this method. Nobody has ever gotten a G that high with traditional lecturing. Over the years since 1996, many of us who use interactive techniques have refined what we do, and it's not uncommon to significantly higher G's. The average for three of us who teach freshman calc-based physics at my school last semester was 0.7.
A common concern is that if the teacher d
Find free books.
If Indian instructors are nearly as hard to understand at the tech phone supports I've had from "Bob" lately....well, it will surely degrade the already failing US education system. Hard to learn if you can't understand a damned thing the instructor is trying to say...
I'm going to have to strongly disagree with the stereotyping going on all over this topic, having recently experienced the reverse of this phenomenon. In my computer science degree we covered advanced AI + Intelligent systems in the second and final years and I found the local based lecturer difficult to understand. Not his accent I hasten to add, he spoke very clearly but he taught in an overly complicated manner. In the end I was saved from panic/ruin and ultimately failing the course when I found a massive set of free lectures on YouTube by some Indian professor who explained it really clearly in spite of his strong Indian accent.
I guess the Einstein quote rings true in my experience, "If you can't explain it simply, you don't understand it well enough."
Interestingly the lecturer is part of some government funded Indian University that do free e-courses on a wide range of science + Engineering topics (all in English it seems) with all the lectures online and handouts / coursework up for free digital download. Not looked through the site in depth but seems to be genuinely free beer learning on degree level e-courses. Again I stress that I've not looked through the site in detail yet, but it seems like they take some of the strongest lecturers from Indian universities and basically record their lectures and upload all their hand outs etc.
Here's the address if anyone is interested: http://nptel.iitm.ac.in/
Please don't perpetuate stereotypes, yes sometimes they do hold some truth but that's no reason to write off so many other really cool things out there.