I have recently written a paper on Research led teaching: Definition and views. Check it out and feel free to send me comments.
As a graduate student at Tel Aviv University I taught laboratory courses to undergraduate classes and instructed classes of physics graduate students in statistical mechanics and electrodynamics. I also instructed undergraduate classes in general physics, classical mechanics, thermal physics and electromagnetic theory. The laboratory courses consisted of both frontal lecturing before experiments and supervision while the experiments were carried out. The instruction involved frontal teaching, expanding on rather condensed lectures, demonstrations of problem solving, homework assignment and marking. The undergraduate classes I instructed were of physics majors, as well as biology, physiology and engineering, while the graduate courses were all physics students. All the above involved grading the students at the end of the course. In the official surveys of teaching and instructing quality at Tel Aviv University I was fortunate to always rank very highly (among the top 10%).
Throughout my appointment at the Cavendish Laboratory, Cambridge, England, I supervised second and third year undergraduates in classical physics, thermodynamics and statistical mechanics. The system in Cambridge is that groups of three or four students from the same college meet with their supervisor in a particular subject several times a week. During the supervision the (rather intensive) lectures are clarified and problem solving practiced. I supervised students from Caius, Churchill and Pembroke Colleges, about three groups per trimester. I also gave a few lectures to graduate students and staff within a forum of lectures on special topics in physics. As a recognition of my presenting skills I was elected to organize the prestigeous Theory-of-Condensed-Matter seminar at the Cavendish Laboratory for two years.
At Princeton University I gave lectures to students on fractals and percolation theory in a course on heterogeneous materials. This subject proved very attractive and popular among the students.
At los Alamos National Laboratory there are no undergraduate students but graduate students who are affiliated with a university come to work at the lab for periods of a few months at a time. I have had two such students: Rudolph Held, a german student, worked with me on characterization of, and flow in, porous media. He is at present a graduate student at Princeton University in the engineering department. The other student, Yi Jiang, from Notre Dame University, worked with me on evolution of two dimensional interfaces and a coarse-graining procedure for heterogeneous media. Both of them did well in those projects and I enjoyed working with them very much.
From 1999 I have been a Specialist Subject Reviewer (Physics and Astronomy) for the UK Quality Assurance Agency for Higher Education. For the non-Brits, this is the agency that grades the teaching in academic departments and departments are putting an enormous amount (quite a few man-years) of work in preparation for a visit of a review team. These visits have contributed much to my ideas about effective teaching.
In general I like to teach and there is a great deal of satisfaction when the penny dropsthe and the light of understanding appears on a student's face. I believe that it is preferable to allow more time for the students to gain fundamental understanding sometimes (but not always) through practice rather than drill problem solving problems via memorizing formulae. I have found that teaching how to approach problems systematically, remembering as few formulae as possible, is in fact a most efficient way to produce successful students. By relying less on memory and more on understanding a student is more likely to gain confidence in and consequently is more likely to attempt applying the methodical 'tools' beyond the assigned homework. This approach should be beneficial to both undergraduate and graduate students, but to the latter I would regard it as essential. After all, the essence of their career will be to solve problems that no one solved before. Moreover, at the undergraduate level, the use and exploitation of students natural curiousity have received, in my opinion very little attention. Methods can, and should, be developed into incorporating forefront research into teaching because, in my experience, it is exactly this kind of exposure that fires the students' imagination. All this should be optimized against the idea that a university is not simply an extension of high school and students should be actively encouraged to independently obtain relevant material and solve problems. Achieving the student's enthusiasm is more than half the way toward this goal.
Most important is that students get a broad and as much as possible in-depth knowledge of the state of the art in science. In today's world there is a need to produce scientists at all levels who know how to apply their expertise flexibly and practically. At the same time one needs to give those who show the inclination the freedom to conduct a line of research that interests them. History shows that one can never predict where the next significant breakthrough in science may come from. This delicate balance between practicality and creativity is a relatively new skill that has to be developed in today's diminishing resources.
I am strongly in favour of open door policy to students who either wish clarifications on the lectures or come with crazy scientific ideas. Essentially, I would suggest to invest a graet deal of resurces (mostly staff time) in encouraging students to think about the ramifications of the subject addressed in class regardless how remote these might seem at first, second and third glance. To me one of the signatures of a potentially good student is the ability to ask questions whose answers may not be clear even to the teacher. And believe me, having spent most of my adult life in the frontier of theoretical physics, there are plenty of those around. In fact, I find that acknowledging the large gaps in current scientific knowledge rather tends to attract students and lecturers / instructors should be sufficiently self confident to admit that. In my opinion, 'embarrassing' questions should be not only acknowledged but actually rewarded. The worst practice is to quench such a question by refering to existing formulae, which are usually limited by their simplifying implicit assumptions. Note that this is somewhat opposite to the unfortunate practice in most highschool (or A-level) science teachings and I believe that universities have to expend much effort in years 1 and 2 to set straight some of the conceptual misrepresentation in pre-university science.
(Under construction)
My principal advisor throughout my graduate studies at Tel Aviv University was professor Amnon Aharony, whose approach to advisorship was probably not too common. In retrospect, it seems to me that his main goal was that a student finishing a Ph.D. with him would be in the best possible position to encounter the scientific arena. This means that he not only guided me in learning the necessary technical skills but also made an effort to expose me to all the top-ranking scientists in my, and related, fields so that I become connected. When a visitor would come to the department he would collect all his students in his not-too-large office and the visitor would present his work and discuss it with all of us. Amnon would usually lead the discussion but would later retreat and let the students take over. He sent me to as many conferences as possible, selecting (under budget constraints) the conferences that maximised exposure and name-making. These were usually the most prestigious ones. I remember one instance when, while I was in his office, he got a call from Mandelbrot who apparently needed someone to work with him for a few months. He turned to me and asked "would you like to go to Benoit for a couple of months?" "Yes", I said, "When?". "Next week" replies he, leaving me dumbfounded. I did go for much longer than a couple of months, but slightly more than a week later. Amnon also put a lot of effort in teaching me to write clear manuscripts. A draft manuscript would come back to me completely covered with red ink at least 15-20 times before he was satisfied. It may be unfortunate that he did not extend this help beyond the Ph.D., possibly feeling that the student needs to find his own place under the sun.
When Amnon went abroad for a sabatical I approached Professor David Bergman who agreed to be my second advisor. (Interestingly, he took a sabatical the following semester or so. The administration, suspecting that it may have been connected with me, warned me in advance that I cannot take a third advisor, even if I tried, lest he takes off too!). David's style was completely different from Amnon's, which provided me with a very broad perspective. He tried very hard to educate me both in the practicalities of scientific life as well as in the right ethics. At the time when my publications list just started to grow he maintained that one GOOD paper a year is worth more than five half-baked ones. He was also one of the most critical persons I had ever met, which is probably why we had relatively few publications together (less than 10 I think) but some of these were real gems from my point of view.
All this, however, is on the surface or just below it. A lower-lying theme, of which I was aware already as a first-year graduate student, was the emotional undercurrent. In retrospect, I see advisorship as not too different from parenthood (In fact, I could easily take Bill Cosby's book on the subject and interpret it in terms of this situation). The advisor receives a raw student and moulds him into a scientist in whatever way that he perceives a scientist should be. Isn't this what parents do with children? The student first looks up to his advisor with admiring eyes, thirstily drinking everything that is poured down on him, much like toddlers do. Then the student tries hard to prove himself to the advisor. Gradually, however, as the student becomes more involved in his field, he starts to notice that there are some things that he knows better than the advisor. This is where the 'teen-age'-like period sets in and the budding feeling that he can eventually surpass the advisor. In due course the student leaves the nest to make do on his own, and, looking back, starts to appreciate the values and education that he received. All this process is not dissociated from emotions that resemble those between children and parents: Devotion, love, disappointment and feeling of being abandonned at times. Talking to quite a few colleagues on this subject, I found that I am not alone in this view and quite a few felt a child-parent sort of relationship with their advisors.
I absorbed much from my advisors and the standards that they set are with me to stay. Whatever I have been given in this respect I regard as a debt that I need to repay to other students, inasmuch as my debt to my parents I repay to my children. Hopefully, I will be able to follow and continue the line of good mentorship that my advisors exemplified.