Summerhill

Photo from Summerhill Photographers

Summerhill, founded in 1921 by A.S. Neill, is an unusual school that everyone should know about. It provides complete freedom to the children in what, how, and when they learn. It has been a great inspiration to the small number of similar schools around the world.
Quoting from its website:

Today, all over the world, education is moving towards more and more testing, more examinations and more qualifications. It seems to be a modern trend that assessment and qualification define education. 

If society were to treat any other group of people the way it treats its children, it would be considered a violation of human rights. But for most of the world's children this is the normal expectation from parents, school and the society in which we live. 

Today many educationalists and families are becoming uneasy with this restrictive environment. They are beginning to look for alternative answers to mainstream schooling. 

One of these answers is democratic or ‘free' schooling. There are many models of democratic schools in all corners of the globe, from Israel to Japan, from New Zealand and Thailand to the United States. 

The oldest and most famous of these schools is Summerhill, on the east coast of England.

Here are a few videos about this school:

• Imagine a school... 
• Summerhill 
• Who is scared of Summerhill 

It really is still ahead of its time. 

Motivation and the Ph.D.

Recently, a couple of my colleagues sent around the following three articles to the faculty mailing list:

• Reform the PhD system or close it down (in Nature)
• Education: The PhD factory (in Nature)
• The disposable academic: why doing a PhD is often a waste of time (in The Economist)

These are all criticisms of the modern Ph.D. system. The authors make two key points in common:

• Ph.D.'s take too long.
• There are few jobs for Ph.D.'s that relate to their research training.

On the face of it, it is hard to disagree with these articles, but there's more to the Ph.D. than these points would suggest. I'll limit my thoughts to the Ph.D. in engineering, which is what I'm most familiar with. The situation is quite different from field to field. While the duration of the Ph.D. is very long in the humanities (routinely taking 10-12 years), it is much shorter in many domains of engineering (4-6 years is the norm).

Nearly all engineering Ph.D.'s do get a job after graduation, but the quality and research-orientation of those jobs shows high variance. I recognize that a good number of the students who did their Ph.D. under my guidance are now in jobs they were sufficiently qualified to do after their masters (perhaps even undergrad!) Not only is there a great paucity of tenure-track academic positions, there are also relatively few industrial research jobs these days, as many companies have laid off their internal groups doing basic "blue skies" research. In my own area of communication networks, aside from a few examples like Microsoft Research and Qualcomm, it's hard to find a large industrial research group in the U.S. (There are more opportunities in the rest of the world, though, for students willing to move to places like China or India). 

Nevertheless, I think there is something deeply valuable about the Ph.D. process, that is not accounted for in the discussion in these articles. It is (or, at least, should be) a far more autonomous learning process than the undergrad or masters degrees because of the focus on independent research. Students are challenged to develop their critical thinking abilities and gain confidence in their ability to reason from first principles. By going in depth, they reach the very boundaries of knowledge, and in many cases for the first time in their lives, help contribute something new to human knowledge. And some students even find a way to balance this depth with  breadth acquired by attending a wide range of talks and courses in fields outside their main focus area. Despite its length, I think a Ph.D. in a high quality engineering program is not a waste of time. 

With respect to jobs, I would like to think that even those of my students who are now in engineering jobs that didn't absolutely need a Ph.D., find that they are better equipped in some ways because of their Ph.D. experience. As for the paucity of academic and research positions, there is a different perspective on this: that the resulting competition may be a good thing. One of the comments posted online by Igor Litvinyuk (who I found to be a faculty member in Physics at Kansas State) on the Nature article "Reform the PhD system" is worth re-posting here in full, as it offers a sharp rebuttal (I should note, however, that I don't share this market-based view entirely): 

Fierce competition for academic positions is the only way to maintain excellence in academia. That's why academia needs more qualified Ph.D.-holding candidates than there are vacancies. Otherwise every Ph.D. graduate would be guaranteed a tenure-track position. It seems like that is what most of the authors and commentators here advocate. But society's interest in having its investments in research to be put into most efficient use by the most qualified people surely makes some extra spending to maintain competition and sustain excellence worthwhile. The populist goal of matching every Ph.D. candidate with a well-compensated satisfying permanent job in academia may seem humane and desirable, but it is not necessarily in the best interest of society as a whole. There is an obvious contradiction in most of the articles here: they claim that Ph.D. overproduction as an established fact, while at the same time stating that unemployment level among Ph.D. holders is lowest of all educational levels. 

So earning a Ph.D. clearly improves everyone's employment prospects. How does it square with the argument that many of those positions do not require a degree and resources are wasted on training people for them? Even if that argument had some merit, those extra costs have to be compared with advantages the society gains from having an excellent merit-based academic research system. Some measure of frustration and disappointment among the less successful contestants who chose to participate in this competitive system is unavoidable and is not unreasonable price to pay. Is it really all that different from other walks of life where competition is the norm, i.e. sports, literature or show business? After all nobody forces people to enter those Ph.D. programs, do they?

So if nobody forces people to enter Ph.D. programs, and many of them don't get research-oriented jobs, why do they enter them? And what influences where they end up after the Ph.D.? Over the years, having observed and interacted with many Ph.D. applicants and students in Electrical Engineering and Computer Science at USC, I have made the following observations:

• For many students from international universities (at USC, we get many applicants from Asian countries such as China, India, Iran, South Korea), doing a Ph.D. in the U.S. is primarily a ticket to the U.S. job market in high-tech areas. For most of these students, there is no motivation to stay in academia. They also see relatively little significance in the distinction between a creative research-oriented position, and a regular development/programming position in industry. They are often quite happy at the end of their Ph.D. to land a salaried industry job, regardless of its research content, because it pays well, is relatively stress-free, and guarantees a happy and stable  future in the U.S.
  
  
• Many students who come straight from undergrad or an MS with little prior exposure to a non-academic setting, might be academically strong, but are still unmotivated and unclear about why they are doing a Ph.D. It's where the conveyor-belt of life has brought them, precisely because they were good at school: having been good at school, they specialized in science and mathematics in high school; having been good at high school, they went to engineering school; having been good at engineering school they applied to grad school, so here they are. All their life the schooling system has controlled their destiny and pointed them towards a Ph.D. They get here, and go through the Ph.D. trying to do their best at it. At the end of it though, feeling at least a little burned out by having faithfully followed the system all these years,  all they want to do is finally, finally, enter the "real world" in the form of any decent job that they can land (research-oriented or not).
  
  
• Students who have had prior industry experience and then come for a Ph.D., particularly from within the U.S., are generally more motivated to seek research positions after the Ph.D., than students in the above category. They appreciate and value the intellectual autonomy afforded by a research position as compared to a non-research position.
  
  
• The bar for academic positions is extremely high. Students need to demonstrate not only a rich publication record, but also should have worked on cutting edge topics, should be outspoken, and should have good networking skills to have interacted with faculty and researchers at other places who can refer/recommend them. For this reason, generally only those Ph.D. students that are extremely motivated to go after a tenure-track position from the very beginning figure out how to get them. (Paradoxically, though, given my previous two observations, a good number of those that go on to tenure track faculty positions have never been in industry.)
  
  
• In engineering, another interesting, albeit relatively smaller, category of Ph.D. students are those that go on to found or work at small startup companies. Here again, with their focus on stability and their concerns about visa issues, I rarely see international students venture in this direction. In some cases, this category consists of students following their entrepreneurial faculty advisors; in many others, it's an expression of the great self-confidence and independence needed to be an entrepreneur. I would argue the Ph.D. is helpful for these students by exposing them to the possible opportunities for a high tech start-up.

My observations suggest that the kinds of jobs that Ph.D. students in engineering seek and find after their graduation (and to some extent, even the richness of their experience in graduate school) are very much a function of their individual motivation when they enter the Ph.D. program. This, in turn, depends on several factors, including (1) whether they have had prior exposure to an industry job before coming for the Ph.D., (2) whether they are international students, and (3) their risk-orientation. It would be nice to see these hypotheses investigated  more systematically through quantitative surveys. 

Preserving the Innovator's DNA

An article in the Wall Street Journal by Peter Sims, titled "The Montessori Mafia," provides further evidence of the benefits of early childhood education that focuses on freedom by giving more names of creatively successful Montessori-educated people (See also my earlier post "How to Make it Big"):

...the Montessori educational approach might be the surest route to joining the creative elite, which are so overrepresented by the school’s alumni that one might suspect a Montessori Mafia: Google’s founders Larry Page and Sergei Brin, Amazon’s Jeff Bezos, videogame pioneer Will Wright, and Wikipedia founder Jimmy Wales, not to mention Julia Child and rapper Sean “P.Diddy” Combs.

Sims points to a study which suggests that these are not isolated examples:

The Montessori Mafia showed up in an extensive, six-year study about the way creative business executives think. Professors Jeffrey Dyer of Brigham Young University and Hal Gregersen of globe-spanning business school INSEAD surveyed over 3,000 executives and interviewed 500 people who had either started innovative companies or invented new products. 

“A number of the innovative entrepreneurs also went to Montessori schools, where they learned to follow their curiosity,” Mr. Gregersen said. “To paraphrase the famous Apple ad campaign, innovators not only learned early on to think different, they act different (and even talk different).” 

I dug around for more information on this study. It was in fact conducted by three researchers, Jeffrey Dyer, Hal Gregersen, and Clayton Christensen from Harvard, and published under the title "The Innovator's DNA" in Harvard Business Review, December 2009. Their study concludes that there are five key "discovery skills" characteristic of creative executives:

• Associating: Making connections between different concepts
• Observing: Paying attention to learn what works, and how
• Experimenting: Trying out new things, dealing with failure
• Questioning: Expressing curiosity
• Networking: Seeking out interesting people to learn from

These are very the much the traits I seek in my graduate students, and I have learned from experience that they are extremely rare! This is surprising because these are things, in fact, that all small children do naturally (until they are "schooled" out of these behaviors by a lifetime of highly structured education). Gregersen says exactly this in an article about their study posted online at INSEAD Knowledge: 

Gregersen says the five discovery skills may seem ‘intuitive’ but when it comes to the actual practice, “doing them is counterintuitive.” That’s because the adult world in which we live “does not value these actions.”

Gregersen’s advice? Start acting like a child again: “Not 100 per cent of the time, that would be absurd. We’re adults and we have to run businesses. But 20 per cent, 25 per cent of our time, act like a four-year-old again,” Gregersen told INSEAD Knowledge. “Because all these skills are what four-year-olds do. They ask thousands of questions: ‘Why?’ ‘Why not?’ This and that. They’re always asking those questions … They observe intensely and they’ll talk to just about anybody.” 

“These are the things that we all did as four-year-olds. We all did this stuff. And if we happened to attend a Montessori-type school like many innovative entrepreneurs did, then we still might be doing this stuff. But most school and corporate systems consistently say: ‘Don’t do it, stop doing it’ … and we lose our innate creative capacity.”

The connection is really quite simple. If we expect our workforce to be creative and innovative, their education has to allow them a lot of opportunities and freedom to exercise these skills.

The problem with standards

Alfie Kohn has written widely about freedom in education and unconditional parenting.

Here's an article by him on the danger of standardizing education: "Beware of the Standards, Not Just the Tests".

Kohn argues that in all the concern about testing in schools, we may lose sight of the point that the very existence of uniformized standards is a threat to learning.

He writes:

We have heard the phrase "standardized testing" so often that we may have become inured to the significance of that first word. To what extent do we really want our students to receive a standardized education? At a national conference last fall, a consultant announced with apparent satisfaction that now, thanks to standards-based reform, "for the first time in my experience, people on a grade level, in a subject area, or teaching a course at a high school are [feeling] a responsibility to all have the same destination." That she did not even feel it necessary to defend this goal says something about the current acceptance of a one-size-fits-all model of education. 

Once again, the problem is not just with the construction of the tests, but with the uniformity of the standards. Wanting to make sure that students in low-income communities don't receive a second-rate education is a laudable objective. Wanting to make sure that all students in your state receive the same education, such that they are treated as interchangeable recipients of knowledge, is a very different matter. Even more troubling are grade-by-grade standards. Here, the prescribers are not just saying, "We expect students to know the following stuff by the time they're in 8th grade," but "We expect them to learn all the items on this list in 5th grade, all the items on that list in 6th grade," and so on. Apart from the negative effects on learning, this rigidity about both the timing of the instruction and its content creates failures unnecessarily by trying to force all children to learn at the same pace.

Reading this leads me to some introspection. Each time I give a mid-term or final exam in any of my classes, am I not treating all students in my class similarly as "interchangeable recipients of knowledge"? Is it not what I do whenever I "follow" a standard syllabus or adopt a textbook? The moment we package a set of facts or tools or approaches into the rigid curriculum for a class, we enter a danger zone, that should be navigated with a great deal of care.

We should be leery of all attempts to ensure that everyone "knows" the same things about the subject, because they generally imply a naive view of knowledge as an objective collection of facts and methods. But the veneer of objectiveness obtained, for instance, by the adoption of a standard textbook, masks the sheer subjectivity of the process of putting together a class.

Consider the new course I have been developing at USC on wireless networks. It is  informed entirely by my own readings and research, and is not how the course is taught anywhere else. Say I were to write a textbook based on my course notes, and it were to be adopted as the textbook for a class in another university some day. The view of the students on the subject is then shaped in a peculiar fashion (perhaps irreparably) after my tastes. Somewhere an employer will glance at a student's transcript, and noticing that he/she has taken a course on "wireless networks" and gotten an 'A' on the subject, will assume that he/she "knows" something useful about the field. Yet this knowledge, as *all* knowledge that can be acquired in or outside a classroom, is fundamentally subjective because its very roots were subjective, shaped and colored as they were by my experiences.

While my arguments about the subjectivity of knowledge may seem particularly compelling in the context of emerging new subjects, I argue that the situation is fundamentally no different even for "ossified" old subjects. Here decades of textbook writing may seem to have lead to a consensus about the "core" body of knowledge in the field, but the converged point of view is very much a function of how the original texts were written and organized.

The only saving grace about the college environment, in contrast to the more rigid standardized curricula in schools, is the far greater freedom students have in selecting their program of study.

How to make it big...

A friend pointed me to a nice quote in an article by Steven Levy in Wired magazine: 

“You can’t understand Google,” vice president Marissa Mayer says, “unless you know that both Larry and Sergey were Montessori kids.” She’s referring to schools based on the educational philosophy of Maria Montessori, an Italian physician born in 1870 who believed that children should be allowed the freedom to pursue their interests. “In a Montessori school, you go paint because you have something to express or you just want to do it that afternoon, not because the teacher said so,” she says. “This is baked into how Larry and Sergey approach problems. They’re always asking, why should it be like that? It’s the way their brains were programmed early on.”

It reminded me a little of this essay by John Taylor Gatto, where he provides an impressive list of people who also didn't like to do things because the teacher said so, and exercised the freedom to pursue their interests: Bill Gates, William Faulkner, Steve Jobs, Steve Wozniak, Ted Turner, Ray Kroc, Richard Branson. Every one of these was a school/college dropout.

Searching for a school

This article on the school admissions season in New York is simply hilarious.

Katie Roiphe writes with biting humor about something that many urban parents can relate to. I particularly like this passage:

The interesting element of this obsession is that each of these unique and excellent schools seems to be conferring some ineffable quality, not just on its students, but on the parents of these students. In the 10 minutes they spend dropping their children off in its hallowed hallways, they are seeing some flattering image of themselves reflected back: progressive, enlightened, intellectually engaged. 

The most sought-after school in my neighbourhood, a famously open-minded and progressive and arty yet very exclusive private school, is conferring a kind of creativity on the parents, so that even if they are bankers or hedge-fund guys, as many of them frankly are, they can tell themselves in the dark of night that they are creative people, because their children attend this impeccably creative school. And if they are creative people – that is, people who have somehow made enough money to send their children to this school, but work in film or music or advertising – they can congratulate themselves on their creativity, even if they are not, although in a creative profession, exactly creating anything themselves. The secret suspicion that you might be a hack, a glorified hack, making a rather nice living doing something fun (but not truly living out your fantasy of creating art the way you honestly thought you would be in college), well, the cheque you make out to that fancy, creative, open place you are sending your child to is proving otherwise. They are putting on operas when they are three years old, after all. They are illustrating Wallace Stevens poems by the time they are six. How could anyone accuse you of just being a banker, or a music executive, or an internet guy with good glasses? I have a friend whose five-year-old attends this school. She and her husband were pleased that when their daughter had an assignment to write down what she wanted to be when she grew up she wrote "artist". But when they arrived at the class presentation the next day they saw that all 22 children had put down "artist": there were no "veterinarians", no "circus acrobats", no "doctors", no "hair cutters". Twenty two artists, and one kindergarten class: the school, you see, does not play around.

Teaching limits exploration

Alison Gopnik, a professor of Psychology at UC Berkeley, has written an article on Slate, titled "Why Preschool shouldn't be like School", about two recent research studies that show how teaching inhibits creativity in children.

The research studies she points to are to be published soon in the journal Cognition. One, co-authored by Gopnik, is titled "Children's imitation of causal action sequences is influenced by statistical and pedagogical evidence," which shows that children are particularly inclined to repeat what they see a teacher do.  The other is titled "The double-edged sword of pedagogy: Instruction limits spontaneous exploration and discovery", and shows that when instructed by a teacher, children are likely to learn more efficiently, but are also less likely to discover new facts.


She writes:

As so often happens in science, two studies from different labs, using different techniques, have simultaneously produced strikingly similar results. They provide scientific support for the intuitions many teachers have had all along: Direct instruction really can limit young children's learning. Teaching is a very effective way to get children to learn something specific—this tube squeaks, say, or a squish then a press then a pull causes the music to play. But it also makes children less likely to discover unexpected information and to draw unexpected conclusions.

... 

Knowing what to expect from a teacher is a really good thing, of course: It lets you get the right answers more quickly than you would otherwise. Indeed, these studies show that 4-year-olds understand how teaching works and can learn from teachers. But there is an intrinsic trade-off between that kind of learning and the more wide-ranging learning that is so natural for young children. Knowing this, it's more important than ever to give children's remarkable, spontaneous learning abilities free rein. That means a rich, stable, and safe world, with affectionate and supportive grown-ups, and lots of opportunities for exploration and play. Not school for babies.

This is very much along the lines of what I've been thinking and writing about. I would argue that we need to apply these ideas well beyond preschool. The title of Gopnik's article could be generalized. School (and college!) should be less like school as well...

8-10-year-olds write a real science article

I wrote a little while ago about the importance of giving school and college students "real" work to do.

What better a way to have kids learn about science than letting them be real scientists?

A delightful paper: "Blackawton Bees", published in the journal Biology Letters in December 2010,  is authored by an unusual string of scientists. They are 8-10-year-olds at Blackawton primary school in UK.

This group of kids not only conducted the experiments described in the paper, they also wrote the article in their own words (apparently all of it except the abstract).

The study shows the ability of bees to "solve" a puzzle in the process of searching for food (sugar water). The kids describe the experiment in their own endearing words:

This experiment is important, because, as far as we know, no one in history (including adults) has done this experiment before. It tells us that bees can learn to solve puzzles (and if we are lucky we will be able to get them to do Sudoku in a couple of years’ time). In this experiment, we trained bees to solve a particular puzzle. The puzzle was go to blue if surrounded by yellow, but yellow if surrounded by blue.

Here's part of their description of the methodology, discussing how the bees were marked:

We let the foragers into the arena and turned the lights off, which made the bees stop flying (because they do not want to fly into anything). We picked the bees up with bee tweezers and put them into a pot with a lid. We then put the tube with the bees in it into the school’s fridge (and made bee pie :-)). The bees fell asleep. Once they fell asleep, we took the bees out, one at a time, and painted little dots on them (yellow, blue, orange, blue-orange, blue-yellow,
etc.). We put them into the tube and warmed them up and then let them into the arena. No bees were harmed during this procedure.

The grown-ups involved in formulating this study and planning it were David Strudwick, a teacher, and Beau Lotto, a neuroscientist, whose son was in the class.


For more on the story of this amazing study, see:

http://blogs.discovermagazine.com/notrocketscience/2010/12/21/eight-year-old-children-publish-bee-study-in-royal-society-journal/

Math for kids

Prof. Elwyn Berlekamp is a famous coding theorist from UC Berkeley. He gave the annual Viterbi lecture at USC yesterday. While much of his talk focused on some puzzles/games and their fascinating relationship to coding theory, he also made a point about  mathematics education for kids that I think is worth repeating.

Echoing Lockhart's Mathematician's Lament, he made the point that it is valuable to give kids problems that cannot be solved in a single sitting. That these are, in some sense, the problems that are most valuable, yet never encountered in typical schooling.

He gave the example of the following classic problem he encountered as a ten year old, to which he credits his lifelong love of mathematics.

You are given twelve coins, at most of one of which is defective (either heavier or lighter than others). You are given a weighing balance. Can you determine with exactly three weighings, which, if any, of the coins is defective?

Prof. Berlekamp also talked about his involvement with the Berkeley Math Circle for kids. There are apparently similar math circles for kids in many cities/places across the country. (Here, for instance, is one in Los Angeles: http://www.math.ucla.edu/~radko/circles/ ).

Making work count

Sigh. I just realized it's been more than a month since my last post...

I was having a conversation this afternoon about effective educational methods with a bright young colleague who is very dedicated to teaching (it is sad how rare this kind of conversation is at a University!) During the course of the conversation, I came across a little nugget of wisdom that I thought worth posting. He was telling me about a discussion he had with some one from Red Hat, on incorporating open-source projects into programming classes.

I think this is a great idea that generalizes nicely to other kinds of classes, at least in engineering. A wonderful way to provide students with a source of self-motivation for the work they do in the context of a class (home/in-class assignments/projects) is to make their work available and useful to others. This makes the work meaningful in itself, not merely something done to get the grade.

As mentioned above, if it's a software class, they could contribute to an open-source effort. If they do some experiments or projects that may be of interest to others (in academia, industry, or even hobbyists), they could be asked to post and showcase the results online. This will also motivate students to do a good job of documenting and communicating the work they do. They could also do assignments that are directly documentary in nature, e.g., write a tutorial, create or contribute entries for a wiki on a topic related to the course.

The assignment/project could also be a real-world problem that is of direct interest to some external party. Two years back, I got freshmen students to look into and propose energy-efficient alternatives to lighting for buildings on the USC campus based on real data on the current lighting fixtures that I obtained from USC Facilities and Management Services (FMS). Many of the students were excited about the project and came up with great ideas for upgrades, taking into account the lighting needs of the buildings, potential energy savings, and the material and labor cost involved in upgrades.  I then forwarded the student reports and recommendations back to FMS for consideration in their future plans for upgrading the lighting. The project was not just "make-work" but "real" in the sense that folks at FMS were actually interested in what the students came up with.