Labs and Learning

Steve Gimbel at Philosopher’s Playground is calling for the abolition of lab classes:p>

As an undergrad I majored in both philosophy and physics and I have a confession my former physics profs will surely not like — everything I know about physics, I learned from my theory classes. You see, science classes come in two flavors. There are theory classes where a prof stands in front of the room and lectures and then there are lab classes where for many hours, students walk in ill-prepared and tried to figure out which one of these things we’ve never seen before is a potentiometer, fumble their way through procedures that yield results that are not even close to what they were led to expect, and then plug and chug their way through scientific and error calculations that frankly mean little to them. I will freely admit that all my experiences in lab classes were a waste of intellectual time and curricular space that could have much better utilized.

I found this link via Janet, who has her own thoughts on this subject.

As you might expect, I’m not wildly enthusiastic about this argument. In fact, I would say that you could make a functionally identical argument in favor of abolishing lecture classes. In my view, there are two sorts of edcuational experiences in the physical sciences: there are lab classes in which you directly engage with the principles of the physical world and then there are lecture classes in which students struggle to stay awake as a professor mumbles and fumbles his way through a series of proofs and arcane formulae that are only vaguely connected to any physical process.

“That’s not fair,” you protest, “Not all lecture classes are that incompetently done.” And that’s exactly the point– Gimbel’s argument isn’t an argument for getting rid of lab classes per se, it’s an argument for getting rid of badly designed lab classes.

Of course, this argtument then dissolves into a discussion of whether well-designed lab classes exist in the wild. Lots of people assert that they don’t, or that they’re so vanishingly rare that they might as well not exist. Personally, I can say that I probably wouldn’t be a physicist if it hadn’t been for labs– hands-on work with real physical phenomena is what I really enjoy, much more than theory. And the labs I had as an undergraduate were, in my opinion, very well done.

The problem is that doing labs well requires people to actually care about the labs– which means putting in the resources and the effort needed to make them work well. I had good experiences with my undergraduate labs because I was in a small department at a wealthy liberal arts college. We had equipment that worked, the labs were taught by the same professors doing the classes, and the labs were designed to really teach something about the physics and the equipment involved.

I didn’t TA any classes at Maryland when I was there, but based on the graduate laboratory class I had to take, I can easily imagine the experience being a good deal less pleasant. The graduate lab was stocked with equipment that nobody else wanted– at the start of the semester, the lab manager noted that they had recently upgraded all the computers to have 10 MB hard drives. This was in 1994, and even then, nobody was making 10 MB hard drives any more– they had scavenged them from other computers. That class was a miserable experience, because making it better wasn’t a high priority for anyone.

If you have to deal with lab sections taught by disgruntled grad students, using broken equipment and cookbook procedures that don’t engage with the material, then, yeah, you’re not going to enjoy lab classes. But that’s the same as saying that when you have to deal with professors with thick accents and bad handwriting working off thirty-year-old lecture notes, you’re not going to enjoy lecture classes. Both of those are arguments against bad implementation of good ideas.

Labs can be done well, and there’s a fair amount of pedagogical reasearch out there showing that students learn the essential concepts better through direct exploration than through lecture alone or through cookbook labs. There are people out there who have put a great deal of work into showing how to teach effective lab classes– look at things like the Workshop Physics program at Dickinson College, or some of the studio-style classes that are taught at other institutions. There are thriving Physics Education Research programs at lots of universities, and one of the things they study is how to make effective use of laboratory exercises.

Most undergraduate physics labs are done poorly for the same reason that most undergraduate physics classes are done poorly: educating undergraduates is not a high priority at most large research universities. Labs can be done well, but unless they’re funded and rewarded appropriately, they won’t be.

(There are a number of other points in Gimble’s post that I haven’t addressed here. I’ll take those up in another post.)

Comments

  1. #1 Moshe
    March 29, 2007

    I am a theorist, so have no credibility on the issue, but I did have wonderful labs in my senior year. I think though that first year labs are more problematic. I was surprised to hear Carl Weiman promoting the idea of replacing those by computer simulations. The argument I found most convincing is that in such simulations the student can be encouraged to explore and initiate more, whereas in labs one is more or less confined to following the script. If one deviates from the script some interesting things may happen, but those are unlikely to increase the physics education value…

  2. #2 Scott H.
    March 29, 2007

    MIT has a very grueling lab course taken in the junior year. I *almost* universally hear from the students that it’s the best course they take in the major, because it’s the first time that all the principles they’ve been learning about in lectures come together and turn into something real. (“Almost” because, well, people are different; not everyone gets their jollies from this kind of thing.) See a recent post from Amali to see the kind of joy that students get from this lab. The general reaction is that this is when students realize what doing physics is like, rather than just studying physics.

    What’s the secret? Dedicated technical staff who keep the equipment humming; dedicated faculty who teach this because they want to teach it; and fairly small sections so that the students get plenty of time to work with the faculty. Which of course means money: It’s well funded.

    There are some lab exercises associated with other classes (e.g., freshman E&M) which really are pretty lame — the kind of “show up clueless, get handed a bunch of equipment that you spend 30 minutes trying to figure out, then spend 10 minutes measuring 3 data points” lab that Gimbel is talking about. Those do suck.

    The punchline is that crappy pedagogy is crappy pedagogy, whether it be lab or lecture.

  3. #3 Adam
    March 29, 2007

    I agree that you can have good lab classes although lab was, hands-down, the most hated course by most physics students or former physics students I’ve spoken to (although Thermodynamics could make a strong case in terms of some ratio of ‘despised’ to ‘hours consumed’). Many complaints relate to crappy equipment, but poor supervision and design would be up there, too. It’s a shame; although theory is, perhaps, the sexy physics that grabs the most attention, most physicists aren’t theorists.

  4. #4 Stuart Coleman
    March 29, 2007

    I’ve enjoyed the one lab I’ve taken so far, and I think it’s for the reasons you listed. We only have two guys in charge of the entire lab course sequence, and they both care a lot. We also have enough money to make everything work right. I’m not so sure about how the remaining labs will go, but I don’t think they’re going to be abysmal.

  5. #5 Mike Bruce
    March 29, 2007

    Conclusion: education is lame.

  6. #6 Martin Madsen
    March 29, 2007

    Here are my two bits as a new (1st year) professor at a liberal arts college. It seems to me that introductory labs that ask students to follow a recipe to “cook” up some results using equipment that they don’t understand (and even worse connected to a computer they can’t use… sorry Mac people) is a pointless exercise.

    So, do we abandon labs? No. I am with those that believe we need to re-work the lab experience into something meaninful for the students. Unfortunately, I don’t have the answer as to what that “meaningful” thing is. Any thoughts as to how we can make an introductory mechanics lab “meaningful” for all students?

  7. #7 Clark
    March 29, 2007

    I spent about 5 semesters as a TA for physics labs, both pre-med level and for physics majors. I think they’re great. I don’t know of anyone who knows how to use an oscilliscope that didn’t learn by spending hours wiggling all the knobs trying to figure out why the darn thing isn’t working. Labs are necessary to introduce students to the ‘real world’ where things are broken and some measurements are hard to make.

    The labs my school had for pre-med level physics are a little trickier as students can’t be expected to have the same level of understanding. But they need to see that physics is not a list of equations in a book. It’s also a trial of patience for the TA in teaching the same people how to use a multimeter week after week. And replacing all those fuses as they try to measure current in parallel. But it still say that lab classes are very important.

  8. #8 Carl Brannen
    March 29, 2007

    My only complaint about labs was that I always got lousy grades in them. I am very practical at getting equipment to operate properly and am the fastest person I’ve ever met at debugging something that is broken. But my write-ups always got B at best. And after school I ended up working as an engineer for 15 years who was always comfortable in the lab.

  9. #9 Ponder Stibbons
    March 29, 2007

    It seems to me that introductory labs that ask students to follow a recipe to “cook” up some results using equipment that they don’t understand (and even worse connected to a computer they can’t use… sorry Mac people) is a pointless exercise.

    This has certainly been my experience. It gets even worse for advanced undergrad physics labs. For QM labs, all the experiments required far more theoretical knowledge than we had in order to understand what was going on. What usually happened was that the lab manual told us the equations that were supposed to describe our results. We would run the experiment hoping that those results did emerge. You could object that we should have tried to learn the theory ourselves. I submit that that was simply not possible given time constraints. Many of the experiments required advanced knowledge of solid state physics and/or electronics. The only way we could ahve carved out enough time to understand what was going on was if we had no other academic commitments besides the lab.

    As I commented on RPM’s blog, I suspect labs are much more poorly implemented in the physical sciences than in the biological sciences. Certainly my limited personal experience in one institution bears this out. But I get the general impression, from reading comments, that biologists seem to get a lot more out of their labs than physicists do.

    If there is a widespread problem implementing effective physics labs, then I would question if there is something endemic in the nature of the traditional physics labs that make them difficult to implement.

    I submit that it is better to have no labs at all than to have badly implemented labs. Negative experiences with physics labs put many prospective physics majors off physics. This is certainly the case at my college. I do not think this is trivial, because badly implemented physics labs are nothing like carrying out real research. It seems that there are enough badly implemented physics labs around that this should be a significant concern. Everyone knows that far more physics majors want to be theorists than to be experimentalists, even though far more experimentalists are demanded by the job market. I have little doubt that this is because of the imbalanced picture that labs paint: students never get to experience the frustrations of doing theory, but they associate experimental work with the frustrations of lab courses. Improving physics labs will, I suspect, help ameliorate the problem of over-supply of theorists to a significant extent.

  10. #10 Mike Kozlowski
    March 29, 2007

    I assume labs are useful and necessary for higher-level classes, and therefore necessary practice in intro level classes, but I honestly don’t think I ever learned a thing (except how tempting it is to forge results, and that gravity doesn’t actually work[1]) in a lab.

    [1] Mandatory link: http://www.cs.wisc.edu/~kovar/hall.html (“Electron Band Structure In Germanium, My Ass”)

  11. #11 Moshe
    March 29, 2007

    I am not sure the problem is only the implementation of labs. The lab is actually a very complex place, and quite a bit of knowledge is required to fully comprehend what is going on. So, one can follow the recipe and accept on faith the interpretation one is given, but again this is not a place to learn by getting your hands dirty. I really like the idea of simulations replacing this environment- some of them are very well made, and the crucial point is that the student can ask: what happens if I do this or that, and, well, experiment, and understand what just happened without any help, because it is much cleaner and easier to interpret. This is the point where it can become a game and not an unpleasant obligation.

    This is precisely why my upper level labs were so much fun for me, since at that point you know enough physics to understand things, and the labs therefore can involve quite a bit of problem solving. I am not sure how much of that can be demanded of non-major freshmen students.

  12. #12 CCPhysicist
    March 30, 2007

    It does not surprise me that a Natural Philosophyer, such as the author of the prompting article, finds labs pointless. That was the view of natural science until Galileo, and it is still much easier to do math or slam code than make a quality measurement.

    [*] Disclosure: At least that is how it looks to me, a theorist. My experimental colleagues would probably say that quality measurements are hard but quality theories are even harder to find.

    I’d say the problem is the mistaken belief that labs are supposed to work. I have a friend who spent 10 years getting an experiment to work, and is justifiably famous for doing so. Students need to be taught that they are in the lab to learn practical skills, and part of that process is learning from mistakes. A sequence of labs, where lessons learned from one can be applied to the next one, is as important in the lab as it is in homework.

    The other problem is that students will seldom get much out of any educational exercise when they come to class unprepared. No news there. My view is that they cheat themselves, and I use my alumni to teach that lesson. They rarely listen to me.