In my last post I mentioned the possibility of university-level course lectures being delivered online, possibly reducing the number of jobs for college professors. But Marshall Kirkpatrick writes for ReadWriteWeb about a new high-tech approach to learning that dispenses with lectures and focuses on group discussions:
Lectures made sense before the invention of the printing press, argues Harvard physics professor Eric Mazur, but at this point in history they are far from the best way to transmit large amounts of information or to make use of face-to-face time in the classroom.
Over nearly 20 years, Mazur has developed an innovative teaching methodology and is now testing software to support its application in any classroom. The basic idea is that the bulk of information consumption should be done outside the classroom and in-class time should be spent doing guided, measured, optimized peer-to-peer discussion in order to maximize retention of knowledge. Mazur’s National Science Foundation-backed startup Learning Catalytics looks like a very cool way to facilitate that class time using web and mobile devices.
The course will be taught by Sebastian Thrun and Peter Norvig. The course will include online lectures by the two, and according to the course website both professors will be available for online discussions. And according to the video embedded below, students in the online class will be graded on a curve just like regular Stanford students and receive a certificate of completion with their grade.
One of the interesting things here is that you can, for the most part, get the full education of the course. You just don’t get the course credit. But maybe students at other universities could take the class and then test out of their own school’s AI course? What impact would it have on professors if universities accepted certificates like this to count towards credit toward a degree at their school?
[This post refers to a study with a large sample, but which has not been replicated]
One of the arguments in favor of college is that, regardless of whether you learn practical content in college or whether it helps you find a better job, at least it teaches you to “learn how to learn” and enriches you with knowledge. Unfortunately, according to research presented in the book Academically Adrift, colleges aren’t doing a very good job of this. The authors used the Collegiate Learning Assessment, which measures skills like critical thinking and analytic reasoning to assess 2,300 students enrolled in several different colleges.
According to Inside Higher Education, the authors found:
45 percent of students “did not demonstrate any significant improvement in learning” during the first two years of college.
36 percent of students “did not demonstrate any significant improvement in learning” over four years of college.
Those students who do show improvements tend to show only modest improvements. Students improved on average only 0.18 standard deviations over the first two years of college and 0.47 over four years. What this means is that a student who entered college in the 50th percentile of students in his or her cohort would move up to the 68th percentile four years later — but that’s the 68th percentile of a new group of freshmen who haven’t experienced any college learning.
There are various causes for this lack of learning, but low expectations from professors is allegedly the biggest cause. Students are getting by with less and less work and being rewarded for it. But what’s causing the grade inflation? The demand to keep enrollment up may be one cause, exacerbated by professors desire to reduce the amount of work they must grade. This avoidance assigning more work may itself be caused by budget cutting at universities, and by professors’ focus on their own research instead of teaching.
Another interesting finding is that business, education and social work majors learn less than those in other majors. This lengthy New York Times piece looks at the sorry state of undergraduate business education, and notes that my own major (communications) was among the worst as well.
The Louisiana libertarian think-tank The Pelican Institute rounds-up the evidence that we’re experiencing a higher education bubble:
Concurrently, students are defaulting at an alarming rate: 25 percent of all government loans default, 30 percent of community college loans default, 40 percent of two-year college loans default, and for-profit schools have a 43 percent default rate.
Although student loans are defaulting faster than home loans at the height of the housing crisis, a 2005 decree from the Bush Administration stated that student loan debt could not be dissolved through bankruptcy proceedings. The only other scenario where this “no-escape” clause exists is debt from criminal acts and debt from fraud.
Lukeprog at Less Wrong talks about what he learned about interpersonal communication in a Scientology class, and what it taught him about learning:
Building small skills in the right order is an excellent way to create and maintain success spirals.
Trying to master a large skill set like salesmanship is a daunting task that will likely involve many demotivating failures before you ever taste success. The same goes for public speaking, writing research papers, and lots of other large skill sets involving a complex interaction of many small skills.
Anna Salamon uses math to explain this concept. You could tackle calculus immediately after Algebra I, and you might eventually pick it up after many frustrating failures if you read the calculus textbook enough times, but why would you do this? It’s much easier and more satisfying to learn more algebra piece by piece until the jump to calculus is not so great. That way, you can experience the pleasure and confidence-boost of mastering new concepts all along the way to calculus.
Mr. Jones, 30, and his wife, Alicia, 27, are among an emerging group of people in their 20s and 30s who have chosen farming as a career. Many shun industrial, mechanized farming and list punk rock, Karl Marx and the food journalist Michael Pollan as their influences. The Joneses say they and their peers are succeeding because of Oregon’s farmer-foodie culture, which demands grass-fed and pasture-raised meats. […]
The problem, the young farmers say, is access to land and money to buy equipment. Many new to farming also struggle with the basics.
In Eugene, Ore., Kasey White and Jeff Broadie of Lonesome Whistle Farm are finishing their third season of cultivating heirloom beans with names like Calypso, Jacob’s Cattle and Dutch Ballet.
They have been lauded — and even consulted — by older farmers nearby for figuring out how to grow beans in a valley dominated by grass seed farmers.
But finding mentors has been difficult. There is a knowledge gap that has been referred to as “the lost generation” — people their parents’ age may farm but do not know how to grow food. The grandparent generation is no longer around to teach them.
Another interesting thread for autodidacts on Less Wrong, this one dedicated to compiling a list of the best text books on particular subjects.
There have been other pages of recommended reading on Less Wrong before and elsewhere, but this post is unique. Here are the rules:
1.Post the title of your favorite textbook on a given subject.
2.You must have read at least two other textbooks on that same subject.
3.You must briefly name the other books youve read on the subject and explain why you think your chosen textbook is superior to them.
If you don’t know who Fadereu is, I’m not sure I can explain him quickly or accurately. For simplicity sake, he’s an Indian artist and mathematician – and he’s running an online workshop of the study of symmetry:
he KNK101 workshop introduces the elementary study of symmetry ( known as ‘group theory’) to an audience with no background in mathematics. This field of mathematics has very little to do with numbers, instead – it studies transformation and movement of abstract structures. The applications of group theory range from simple permutation puzzles and military or monetary cryptography to particle physics and general relativity, making it the central conceptual framework of our age.
The workshop is spread over 6 weeks ( or three fortnights ) and the details for registration are here. [ tldr: just drop me a mail at fadebox/gmail. The fee is $50 (international) and the equivalent Rs 2400 for India. Please hurry!]
I don’t agree with PayPal co-founder and early Facebook investor Peter Thiel’s worldview, but I agree with much of what he has to say in this interview in National Review – particularly the section on education:
Education is a bubble in a classic sense. To call something a bubble, it must be overpriced and there must be an intense belief in it. Housing was a classic bubble, as were tech stocks in the ’90s, because they were both very overvalued, but there was an incredibly widespread belief that almost could not be questioned — you had to own a house in 2005, and you had to be in an equity-market index fund in 1999.
Probably the only candidate left for a bubble — at least in the developed world (maybe emerging markets are a bubble) — is education. It’s basically extremely overpriced. People are not getting their money’s worth, objectively, when you do the math. And at the same time it is something that is incredibly intensively believed; there’s this sort of psycho-social component to people taking on these enormous debts when they go to college simply because that’s what everybody’s doing.
It is, to my mind, in some ways worse than the housing bubble. There are a few things that make it worse. One is that when people make a mistake in taking on an education loan, they’re legally much more difficult to get out of than housing loans. With housing, typically they’re non-recourse — you can just walk out of the house. With education, they’re recourse, and they typically survive bankruptcy. If you borrowed money and went to a college where the education didn’t create any value, that is potentially a really big mistake.
There have been a lot of critiques of the finance industry’s having possibly foisted subprime mortgages on unknowing buyers, and a lot of those kinds of arguments are even more powerful when used against college administrators who are probably in some ways engaged in equally misleading advertising. Like housing was, college is advertised as an investment for the future. But in most cases it’s really just consumption, where college is just a four-year party, in the same way that buying a large house with a really big swimming pool, etc., is probably not an investment decision but a consumption decision. It was something about combining the investment decision and the consumption decision that made the housing thing so tricky to get a handle on — and I think that’s also true of the college bubble.
One important difference between the housing bubble and the education bubble is that there was sort of a class aspect to the housing bubble: upper-middle-class people in the U.S. tend to be invested in equities, and middle-class people tend to be invested in housing, so there was a way in which the housing bubble was a way of making fun of the middle class for various sophisticated elites that ran all the way through the housing bubble. It was sort of like, “Look at those dumb people and beatniks in suburban America who are doing this crazy housing thing.” So even though it was a crazy bubble, there was at least a kind of counter-narrative; you had a bit of a dissenting narrative. Education is an upper-middle-class thing, and so something that is not questioned by elites at all, and that’s why the education market is more likely to be distorted.
You know, we’ve looked at the math on this, and I estimate that 70 to 80 percent of the colleges in the U.S. are not generating a positive return on investment. Even at the top universities, it may be positive in some sense — but the counterfactual question is, how well would their students have done had they not gone to college? Are they really just selecting for talented people who would have done well anyway? Or are you actually educating them? That’s the kind of question that isn’t analyzed very carefully. My suspicion is that they’re just good at identifying talented people rather than adding value. So there are a lot of things about it that are very strange.
The Great Recession of 2008 to the present is helping to bring the education bubble to a head. When parents have invested enormous amounts of money in their kids’ education, to find their kids coming back to live with them — well, that was not what they bargained for. So the crazy bubble in education is at a point where it is very close to unraveling.
In early 2009, there was a question of why the stimulus money was not going to infrastructure, and a very large amount was going to subsidizing college loans and encouraging people to go back to school. The argument was that we get a higher return on human capital than on infrastructure. While that’s certainly possible, and I agree that human capital is extremely important, I think we’re not actually measuring the return we’re getting on the human capital. It is, in fact, considered in some ways inappropriate to even ask the question of what the return is. We are given bromides to the effect of, “Well, you know college education is good, but it’s good precisely because it doesn’t teach you anything specific; you become a more well-rounded person, a better citizen, you learn how to learn.” There tends to be an evasion of specificity of what exactly it is that is learned. And so these human-capital intuitions may be very far off in a lot of colleges.
I thought his position on seasteading was interesting as well:
Seasteading was thought up by acolytes of Milton Friedman. The idea is that we need to create competition between governments. If it’s very hard to reform existing ones, we need to create new sovereign states — in the oceans or elsewhere. There’s a technological question about how far away we are from these kinds of things. It’s probably not around the corner. But these technological projects are worth pursuing.
It’s one of the ways in which I see things in the U.S. as having declined from the 1950s, when people had a real sense of the future, and the future was an important subject for public discussion. We thought about being on the moon, or living underwater, and what we were going to do about farmlands and forests and so on. Different ideas about how technology would change in the future played an important role in our society. That sort of collapsed with everything else in the late ’60s and into the ’70s. I want to go back to the future and back to a time when people were thinking about how to use technology to make the world a dramatically better place — not like the present, where technology is largely seen as irrelevant and specifically as bad.
Some people have criticized my advocacy of a student loan jubilee by saying that college kids who made bad decisions don’t deserve to be bailed out. Well, as Clint Eastwood said, “deserved” hasn’t got anything to do with it. We need a student loan jubilee to keep the angry, unemployed hordes from storming the Bastille and dragging the royalty to the guillotines. It’s not about what THEY deserve, it’s about YOUR survival. But hey, just keep repeating those mantras of “self-reliance” and “pulling yourself up by your bootstraps” – I hear the grave’s a fine and private place.