I’ve seen a few links passed around to this Tom Siegfried post about science literacy, which is mostly a familiar story about how polls show most Americans giving incorrect answers to science questions. The sort of stuff you find in the NSF’s Science and Engineering Indicators report. What’s getting the social-media attention, though, is this paragraph near the end:

In fact, I’d contend (and have contended) that the problem with science education is not that it fails to inculcate enough facts, but that it tries to inculcate too many. Science classes in high school and intro classes in college seem to be taught as though everyone needed preparation to pursue a Ph.D. Seriously, calculating solubility constants in high school chemistry classes is about as useful as teaching drivers’ ed students how to maneuver an F-16 fighter jet. Important general principles that could (and should) be retained for a lifetime are diluted to the point of homeopathic impotence by a flood of excessive technical detail.

There’s a level on which I share the “yeah, screw the boring calculations!” reaction. Particularly when it comes to science classes in subjects I don’t teach, like high school chemistry. (Though, to be fair, my high school chemistry teacher was very good, and I enjoyed the class. The Regents exam in chemistry was a joke, though– I would’ve aced it, had it not been canceled over a cheating scandal. But I digress.)

Bring it around to physics, though– and we get lots of similar complaints in physics, mostly about deriving equations– and my reaction is a little different. And that’s the problem– it’s easy to deride stuff you don’t like or understand as “excessive technical detail,” but from the inside, that stuff often seems critically important. One person’s “excessive detail” is another’s minimum necessary factual content.

A lot of calls for sticking to “general principles” also completely miss the essential nature of science and education. That is, it’s all well and good to talk about grand overarching principles, but if you want to learn science, you have to do science. Which sometimes means getting into the tedious mathematical bits, because that’s what science is at some level. You wouldn’t teach driver’s ed students on a fighter jet, true, but you also wouldn’t give them a license based only on classroom lectures about the “general principles” of driving. They need to spend some time out on the road behind the wheel to know what driving is really like.

The other hidden assumption of these kinds of arguments is that perfect “tracking” is possible– that students at the high school level know what they really want, and that we can accurately identify those who really will be going on to get a Ph.D. in science from those who don’t really need to know the technical stuff. I’ll pause for those who actually deal with high school and college kids to stop laughing and catch their breath.

We regularly pick up new majors in their sophomore year, and have had students decide to add a physics major as late as the middle of their junior year. I’ve even signed paperwork to approve some interdisciplinary majors with a physics component for students who weren’t really on our radar until they were seniors. There are any number of “how I became a science writer” stories out there from highly respected journalists who only drifted into science after college, and you can even find people who made a late decision to become professional scientists. Hell, string theory demigod Ed Witten has an undergraduate degree in history with a minor in linguistics, and only switched to math and physics in graduate school.

It’s just not possible to perfectly sort students into groups who will definitely need the technical stuff and others who just need “general principles.” We can do it a little bit– I’ll be teaching a “Gen Ed” class on relativity in the Fall term, trying to cover the general principles with even less math than in my talking-to-the-dog book on the subject– and I’m fairly confident that none of the students taking that will end up missing the technical derivations. I’m not perfectly confident in that, though.

And, of course, there’s a resource issue. Yes, we absolutely could make introductory classes that stick to the inspiring general principles and leave out the technical details. If we add another year to the degree program.

Introductory college science classes include a lot of technical stuff because those courses have to do double duty, as both the token science class taken by students who won’t ever need it, and the entry point for students who are going to major in the subject, who really do need the technical detail at some point. And if the technical detail isn’t there in the introductory classes, lest it turn off the non-majors, it has to get put somewhere else. Which means either making the first “real” major class brutally unpleasant, as you try to pack in all the detail that they didn’t get in the introductory class, or you end up adding a whole extra course, and extending the time to complete the degree. Or you water the degree down, and push more stuff off to graduate schools, which brings a whole new set of problems.

This is not to say that there aren’t things we should be doing to make introductory courses more effective and engaging. We could certainly do better, and that’s largely why Physics Education Research programs exist in a lot of departments. Plenty of people take this very seriously, and are working to make science education better.

At the same time, though, the situation isn’t nearly as simple as the usual “too much technical stuff” arguments would suggest. The technical detail in intro classes isn’t just there because faculty are lazy traditionalists, but because a lot of other factors come into play that aren’t necessarily obvious to students grumbling about solubility calculations on their chemistry homework.

Comments

  1. #1 Margaret Harris
    May 7, 2014

    I agree with your arguments about resources and about high school/college kids not knowing yet what they will need later. However, I think that your arguments about what constitutes unnecessary technical detail are missing the point a little. Maybe what’s needed for those intro/preparatory courses isn’t *less* technical detail per se, but rather *different* technical detail that more accurately reflects what the majority of students will actually need.

    My favourite example from my own education is the inordinate amount of time I spent learning techniques for matrix manipulation in high school Algebra II. I didn’t need those techniques again until I took linear algebra and quantum mechanics in college 3-4 years later (by which time I’d pretty much forgotten them and had to re-learn), and since those courses are a minority pursuit even for science-minded, college-bound students, my guess is that most of my classmates never needed them at all. If we had done more (and earlier) calculus instead, it would have been a lot more useful for a lot more people, and certainly no less rigorous.

  2. #2 quasihumanist
    May 7, 2014

    Too many people seem to think that society only needs people to have true beliefs about subject X, when in fact society needs people to have justified true beliefs (i.e. knowledge) about subject X.

    The point of general education isn’t some vague canard about being well-rounded or acquiring soft skills. The point of general education is philosophical. College is supposed to teach you what knowledge is, and to understand what justified true belief is, you have to see a wide variety of the different forms justification can take. Technical derivations are part of justification in physics.

  3. #3 z
    May 7, 2014

    I think you should still stick to the old argument you had of why there aren’t any “English classes for Science Majors”.

    That or have him look at what is needed for PHD classes and the gap compared to what the intro classes teach. If intro classes are teaching F-16 piloting, then PHD students are learning how to manage highway traffic for an entire state.

  4. #4 Eric Lund
    May 7, 2014

    Both Margaret and quasihumanist are talking about things which perhaps could be taught better than they are.

    Matrix manipulation shows up in a bunch of contexts where it may not be immediately obvious that that’s what you’re doing. Coordinate rotations, for instance, are mathematically equivalent to a matrix multiplication, and I’m sure most science majors have to do something of the sort somewhere between high school algebra and quantum mechanics.

    Teaching students how we know something is true, rather than just that it’s true, is an important part of many college courses (not just in the sciences). It probably should be done more than it is, so that, for example, people who take Econ 101 realize that they are dealing with highly idealized systems which may or may not correspond to the real world (my freshman mechanics professor was quite up front on that point, and at least one could show that the stuff he was neglecting wouldn’t change the answers much, something which is not true of much of economics).

    But what ultimately happens, certainly in high school and far too often in college, is that teachers stick to a syllabus where the justification for learning most of the material is, “This will be on the exam.” This approach predictably leads to students forgetting the material until it shows up in a subsequent class where the professor should be able to (but can’t) assume that the students already know this stuff.

  5. #5 JGB
    May 7, 2014

    More to the above point about justified true belief. What students of all stripes truly need in modern society is the ability to form reasonably justified opinions about science and technology issues (both personal and political). Of course trying to predict what science and technology issues will arise within the lifetime of a 20 year old is impossible. The only sensible solution is a combination of breadth (to have a general idea of many things known to be important) and really deep technical explorations of certain science subjects, so the students understand how expert opinions are formed and they have a robust notion of what it means to have scientific evidence. Anything short of that and ultimately the student is forced to accept all scientific opinion on faith rather than a justified belief.

  6. #6 CCPhysicist
    May 7, 2014

    I’m curious what you think about emphasizing concepts (like those measured by the Force Concept Inventory and other similar diagnostics) over “excessive technical detail” like derivations or solving novel problems.

    Similarly, I’d love to see a wide ranging discussion (in a different thread) of what should be dropped from first year physics so it is less of a mini-PhD course and has more focus on deeper learning of skills and knowledge that they might have a chance to retain when then move on into their upper-division classes.

    At my college, we do not use an introductory physics class as a general education class suitable for english lit or business majors. That might work at a highly selective college like yours or a highly specialized college like Harvey Mudd, but not one where most students don’t even have an SAT score on record, let alone one above three digits. Similarly, we do not require a “general” class for students planning to major in science or engineering so they do not use valuable credits on something that will be an elective course for them.

  7. #7 Chad Orzel
    May 7, 2014

    I’m curious what you think about emphasizing concepts (like those measured by the Force Concept Inventory and other similar diagnostics) over “excessive technical detail” like derivations or solving novel problems.

    We use the FCI in our intro mechanics classes, largely because I hand-coded it into Blackboard lo these many years ago, and it was shorter than the other big mechanics test (the FMCE). We have periodic arguments about whether it actually measures what we actually teach using the Matter and Interactions curriculum, but thus far we’ve stuck with it out of a combination of inertia and the statistical power that’s been built up over many years of FCI use.

    I think it’s a useful tool for assessing things, largely because it steps around the technical details. It’s entirely possible, as the authors of the original test demonstrated, for students to do well on conventional problem-solving tests through memorization of algorithms without any real understanding of the underlying physics.

    Of course, you can do something similar with the FCI and similar tests– rote practice on the sorts of questions that show up on those will get you good scores without the ability to solve problems, or answer questions outside a particular range. So you need a mix of the two– some FCI-type conceptual stuff, and other textbook-type problems– to keep everybody honest.

    I was writing this hastily this morning, and so in places kind of munged together thoughts about intro college courses and high school stuff. We don’t use the intro courses as Gen Ed for English majors (though a fair number of economics majors take it, because they started out as engineers originally…), but they are largely service courses for the engineering departments, and most of the students in them take the view that they’ll never “really” need most of the technical details of physics. Most of our “gen ed” service load is in the introductory astronomy courses– we don’t actually offer a “physics for poets” conceptual survey class.

  8. #8 Bill Bradley
    May 8, 2014

    I think that the ideal way to teach is so that the lack of Mathematics does not prevent a student from learning the topic, but rather is an incentive for the student to learn the required Math. For high school Physics and Chemistry this barrier is typically Algebra, but at the University level, the entry barrier becomes much more serious.

  9. #9 rork
    May 8, 2014

    “Teaching students how we know something is true, rather than just that it’s true, is an important part of many college courses (not just in the sciences). It probably should be done more than it is”
    That was better for my ears than the article itself, and is what I always complain about, usually for high-school, or college intro courses, about math or statistics. That square root of 2 is irrational is worthless crap, but trying to show it’s true isn’t, and having it be obvious (for any n that is not the square of an integer) feels good. What scares me the most is that when teacher states a fact like that, rather than every student being trained to instantly think “wait a minute, why is that so?”, usually not a single one will ask that. Math proof-by-authority has taken over, and it strikes me as being anti-math.
    I’m also sick of folks saying they don’t need to know this or that, and are kinda proud about it even. Average folks may underestimate what the future holds – increasingly, if a task is worth doing, it’s worth getting analytic about it. I find people often fail to have intelligent debates about issues cause they have no clue what the decision theory looks like (they think “science” will tell us what to do – they seem to think science provides the loss function). I’m not saying every person needs to prove Savage’s or Finetti’s theorem though.
    Even if it is only a exercise for your brain, it might be worth it. You must learn to think hard. Feel the burn, be it math, physics, or Shakespeare.

  10. #10 CCPhysicist
    May 8, 2014

    Out of curiousity, particularly because your college is so different from the ones that provided the original data in the FCI paper, what “normalized gain” do you have in those years of statistics?

  11. #11 Young CC Prof
    May 8, 2014

    I teach a statistics class for general education, and this balance is always a tricky one. My goal is not to have students memorized 30 different formulas and use them by hand on large data sets, because that’s what computers are for. My goal is to teach them to do what computers can’t, which is to interpret a statistical analysis, and, more importantly, be able to understand what other people write about statistics.

    I want them to leave with the knowledge of what a p-value is, what it means and what it doesn’t mean, but you can’t understand what a p-value is unless you’ve calculated it yourself for at least one or two types of hypothesis tests. (Of course, you can calculate p-values until your processor overheats without comprehending a single one of them. Many people do.)

    Definitely, teachers should keep in mind the need to balance the ultimate big-picture educational goals with the technical detail necessary to comprehend the big picture in the first place. But big-picture without details is ultimately as vacuous as calculations and facts without context.

  12. #12 Jim Thomerson
    May 11, 2014

    I taught high school physics (poorly) one year, and had a student complain that I was teaching algebra. I explained that we learned algebra so we could do physics.

    I taught both university general education biology and various introductory biology courses. My goal in general education was to have the student, on finishing the course, thinking that biology is both important and interesting.

    A colleague characterized teaching biology as story telling. I have always thought it was important to understand how we came to know what we know, and why we thought it important to do so. So I tend to teach from a historical perspective when I can.

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.