Labs and Naivete

While I do agree with your general point, I must say that the humanities classes at RPI were very much "Poetry for Physicists". I remember taking Shakespeare and being stunned by the elementary nature of the class. I think my total workload for the whole semester were two 3 page papers and a 8 page paper! The same essays that my high school British Lit. teacher would have savaged got an A! Other humanities classes were similar, some better than others, but the workloads were almost always trivial compared to the 'real' classes.

I would point out that I constantly meet PhD's who proudly proclaim that they don't know a thing about music. Science/math illiteracy (innumeracy?) is not alone in this. And we do have to offer classes that don't require the students to read music notation, which is as big a problem as nonmath-based science.

While I disagree with you quite strongly on the random admissions process (or even partly random where you lottery select after a cut-off, so don't take my silence in that area as a notion that you've convinced me...) I agree strongly with you here.

Physics, and the hard sciences and their cousins, the engineering disciplines, are physical disciplines, and moreover, physical sciences. (Not that I'm impressed with non-physical sciences, per se....) Without labs and experiemntation, we take a giant step back to the Greek notion of, "Let's think about it really hard and decide we understand it," or at best to the Medieval idea of, "Our betters in the past have said that this is so."

I have a very dim view of, say, engineers who practice at a high level in disciplines where they have no prior actual physical experience. The physical experience lends a strong and useful sense of intuition about the process in a way that I'm not convinced book-learnin' alone can impart. And that process should begin in school, not at the bench on the job.

How many people actually teach labs? At Stanford there are a total of three faculty who take care of labs, one for the introductory ones (which are kind of boring and useless), and two more for the upper-level ones, plus TAs. It seems to me that should be about standard, and three faculty out of an entire department just isn't that much.

I totally agree with you that what we'd really need to clean up are the introductory courses, which seem to be awful. I had to choose between one that was soul-crushingly hard or mind-numbingly easy, and the hard one did a good job of crushing my soul. I bet that series turns a lot of people off of Physics.

But the stuff afterward has been great. I think it's kind of like Bio (at least how Bio's run here) that you have to take the year of Chem, then the year of introductory Bio and labs, but after that you get into really interesting and fun material. But there are like five classes available to non-majors, which is unfortunate.

The real problem is that it's become acceptable for people who are ostensibly highly educated to be largely ignorant of science and math.

I wouldn't really encourage science majors to take a typical introductory physics course, either, because typical introductory physics courses are really badly done, for the most part.We need to do a better job teaching everyone, regardless of their major. At the same time, though, we need to make it clear that ignorance of basic science is not something you can shrug off with a laugh.

I'm very much in agreement with both of these statements. And I would like to see much more effort spent by subject specialists in understanding the work that has been done by education researchers. There's already a lot of knowledge out there on how to do it better, it's just not being done. It isn't valued in a place where the changes will be made. As long as faculty are judged solely on their research and fundraising ability, with no regard to their teaching skills, it won't change. And that's a damned shame. We're all paying the price for it.

Well, if you only have 7 faculty in the department (like mine), then 3 out of 7 is a lot. But Stanford isn't the only model. In my department, 7 out of 7 faculty members teach labs.

But I disagree with Chad about this:

It is true that if you eliminated labs, you'd see a bunch of new classes taught in the sciences, just to pick up the slack in the teaching load.

If labs were eliminated, the slack in the teaching load would be picked up by eliminating faculty, not by having faculty teach other courses.

if a science student were to say "Oh, I just can't deal with literature," that would be a major crisis.

While I agree with you regarding the importance of labs to science education, and taking a dim view of supposedly-well-educated people who are comfortable with innumeracy and lack of scientific understanding, I should point out that science & engineering departments at research-oriented universities are *full* of students who slack off and don't care about humanities and social science disciplines.

Granted, the attitude among science and engineering types is not usually, "Oh, gosh and gee, reading is *hard*" as "Bah, that's trivial stuff and I could ace it if I wanted to, but I just don't care to be bothered." But IMO both expressions come from the same sentiment--my area of study is *super-important* and yours is so worthless that I can't be bothered to learn anything about it.

Two points, if you please:

First, I think you are falling into exactly the false dilemma here that I was trying to point out. It is not simply a question of "physics for poets" or "physics for geeks" where the first is dumbed down and has no math and the second one is a legitimate course. The sorts of classes that get squeezed out are courses I've taught on the philosophy of quantum mechanics, the philosophy of chemistry, and the history and philosophy of geometry. Students in the class were expected to do (in some mainstream sense of "do") real science and math. These were not baby physics, chem or math classes, but interdiciplinary discussions that required some sophisticated understanding of the science before we asked questions about it -- like what would count as an answer to the measurement problem and what are the strengths and weaknesses of the historically important attampts to provide an answer.

The key to my argument is that there is a difference between training and teaching. These classes were not about training to teach students how to solve Schrodinger's equation for the free particle, particle in an infinite well, the harmonic oscillator, or the usual fare for the QM class every phys dept offers. Students in the class we taught thought about QM, but were not trained in it. Unfortunately this is a distinction that often is not made in scientific pedagogy -- if we teach you how to set up and solve this class of equations that is considered to be identical with understanding the physics in this area. No it isn't. One can teach science to both the major and non-major populations without necessarily dumbing it down, but courses that do this are innovative and the room in science departments for such innovations -- especially in small schools with small departments -- is sorely limited if that department believes that every theory class needs to be coupled with a mandatory lab class. that lack of innovation, that lack of resulting classes is a loss for the students.

Second, I also am not arguing that lab courses are unnecessary or problematic in general. Rather, it is the supposition that each or most theory classes must have corresponding labs. When this is the case, they are frequently taught in a way that does not give the curricular bang for the buck that could be gotten from other sorts of pedagogical work by scientists (this, of course, is not true everywhere, labs are taught innovatively and well some places).

Finally, as a former physics major who tried to get dates in college, I know a brush off when I see one and the "I can't do it because I need to teach the lab" is not just a way of saying "Thanks, but no" politely. Actually, the last one who said it is a physical chemist with a really good personality and we now have a standing date every MWF next semester at 1:10 -- and his chair doesn't mind if he teaches with other people...even someone like me...

I have a number of problems with this, starting with the fact that any class will necessarily knock out a large range of other possible classes. It might not be 25% of all possible courses (though given that a four-course load is fairly standard, you could argue that any class accounts for a quarter of a student's possible courses), but it will be a substantial number, particularly if the class is offered beween 10 and 2, which are the most convenient hours for both students and faculty.

This also ignores the fact that time spent in lab is just another example of the different distribution of work between science and humanities classes. Yes, science classes will force you to spend an afternoon a week in lab, but on the other hand, science classes won't expect you to read 200 pages of Nietzsche between one lecture and the next. There's more in-class time, but less out-of-class time, and those more or less balance out.

I don't think the argument that labs knock out time for other classes is a good one, but I think that they do knock out substantially more classes than the average non-lab course. The labs I've had in college were either 4 hours a week or 2 x 4 hours a week, in addition to the normal lecture time. So that's 4-8 hours extra time knocked out. And besides time spent in lab there are problem sets. In my experience it takes more time to do a typical problem set than it takes to do my weekly reading for a philosophy class. While I do well in both philosophy and physics classes, I spend about five times more hours on a typical physics class than on a typical philosophy class. I also feel that many of those hours were unproductive --- I learnt almost nothing from some labs and some courses (as Steve says, many physics classes teach only procedural methods of calculation, not new physical concepts), but I have yet to emerge from a philosophy course learning nothing. But perhaps that is due to poor implementation of a theoretically good system at my university. I don't know how it is implemented elsewhere, so I can't say for sure.

Oops, the second paragraph above was also part of the quote from Chad's post.

I should clarify:

While I think lab courses do knock out many more time slots than non-lab courses, I don't think this fact is a good justification for abolishing lab courses.

Pam: While I agree with you regarding the importance of labs to science education, and taking a dim view of supposedly-well-educated people who are comfortable with innumeracy and lack of scientific understanding, I should point out that science & engineering departments at research-oriented universities are *full* of students who slack off and don't care about humanities and social science disciplines.

Absolutely.
I think the difference is that if a student said "I don't do well with reading," even the science and engineering faculty would view that as a major problem with the student's preparation for college. If a student says "I don't do well with math," though, faculty in the humanities don't see that as a big issue, because Math Is Hard.

Steve: The key to my argument is that there is a difference between training and teaching.

And I think that you're trying to make a sharper distinction between those than can really be made, at least when it comes to lab classes. While there's some "training" element to a lab, a properly designed lab is also about "teaching"-- making the physical phenomena more concretely visible for the students.

There's a whole pile of literature on the subject, all of it showing that students learn and retain physics concepts better when they experience them in a real, hands-on way (either through labs or "interactive demonstrations," which are basically in-class mini-labs). It is not possible to cleanly separate classes into "teaching" lectures and "training" labs-- when they're done correctly, labs are very much a "teaching" experience, and an essential counterpart to lecture teaching.

Pam:
Granted, the attitude among science and engineering types is not usually, "Oh, gosh and gee, reading is *hard*" as "Bah, that's trivial stuff and I could ace it if I wanted to, but I just don't care to be bothered." But IMO both expressions come from the same sentiment--my area of study is *super-important* and yours is so worthless that I can't be bothered to learn anything about it.

There's something to that, but when I was an undergrad, I and my cohort regularly took advanced undergrad non-science humanities classes (at the 300/400 level) and either passed or legitimately did well according to our efforts and natures. Sometimes we argued our way into classes for which we had no prerequisites, because the alternatives looked dreadfully boring.

I do not ever recall anyone from the non-science humanities taking the corresponding level of math classes, and the math requirements for non-science and non-engineering majors were truly laughable. (And without those level of math classes, the possibility of taking an advanced undergrad science or engineering class is zero.)

Anecdotally, at least, the situation is not symmetric.

Cutting back on lab classes will hurt science education rather than help. I think labs are an integral part of the science curriculum. Without labs science classes boil down to "believe this because I say so and it is in the text book".

In middle school I became interested in science because of the labs. Not only did the teacher make so much sense in explaining how things work, but you could also prove it to yourself in lab.

You learn the scientific method by doing the labs and writing up lab reports. You learn to look at data and try to think critically. You learn to trust science over things like religion, superstition or woo and also realize some of the pitfalls so you don't become disillusioned or confused when scientific reports seem to contradict one another.

My freshman QM class was taught by a physics prof with a PhD in physics and a PhD in philosophy. Further, all of my QM classes were generally structured to teach the theorists and train the experimentalists. Lab classes were the other way around, which might explain Steve G's "theoretical" view of labs as a training exercise.

Doesn't explain his straw man, because no one has ever argued for a 1:1 match between lab and lecture classes. (Typical majors curricula are maybe 1:3 at the most.) The real argument is whether one can have an informed discussion of epistemology if you don't have a concrete grasp of the standard deviation in a real world context. I think not.

I'll agree that we probably fail to convey a full understanding of sigma in a semester or two, but that is why you need more than one class to become a physicist or an engineer. It is also why a more productive discussion would be about how to teach those ideas, and by whom (grad students in lecture, faculty in lab?), than to argue that there should be less critical thinking about objective reality in physics education just because it is hard work.

Part of the reason I didn't major in physics as an undergraduate is because of the tedious lab classes. They burn a lot of time. And I hated spending the time writing the results up. If I wanted some grad student to comment on my ability to write up a lab report on a sliding block, no thank you, I'll have the English department do it.

In a lecture classs I could frequently ignore the professor and grind out the homework for some other class. Labs, you had to attend. What a bummer. So I got my math degree in 3 years and earned a physics degree as a grad student.

It's not that the lab classes were difficult for me. I was very proficient at getting equipment to work, at least partly due to the hobbies I had before college. After grad school I ended up designing electronics for a couple of decades and in that I invariably had the assignment of getting new systems debugged because I was very good at using a 128-channel logic analyzer, soldering little things, and finding and fixing bugs.

By the way, Dr. Pion linked this in.

> And we do have to offer classes that don't require the students to read music notation, which is as big a problem as nonmath-based science.

That's a ludicrous comparison, because one can understand music and appreciate it without knowing how to write it down. Music works on a much more primitive level and existed long before there was a notational system for defining it permanently.

Math and True Science are inextricably intertwined. Most math has historically developed as an attempt to explain or understand a complex natural problem. There are aspects of math itself -- "metamathematical entities", which are used to define and understand the nature and limitations of what we know about math -- but for the most part, math exists to support science. And to understand how and why science works, one needs to understand how the math works, because it is generally through the math that you identify and make predictions.

The Swartzberg Test:
The validity of a science is its ability to predict.

Given the choice between a dozen geniuses in the lab or a couple of idiots who can do field work, take the idiots. It is more important to observe the facts accurately than to be able to interpret them, because if you observe enough of them, they'll explain themselves.
- David Gerrold, "A Matter For Men"

Science is Experimentation. All the rest is philosophy.