random snippets

The derision expressed by the way you say "the phys ed types" says all you need to say. It's like when some creationist says "oh, the science-types will claim to have an explanation for this" and you just know there's no point in engaging with this person.

No, it is not like that at all.
But, the internet is a big place, feel free to hang out somewhere else - I'm very short on patience for humourless pedants this week.

Let's hear the theory. Although I don't think this is physics education versus whatever. It's my impression that the physics education types are concentrating mostly on how to teach intro physics - but attrition as described in the article, for science majors, occurs at all stages, even among students who get through the intro class without difficulty.

FWIW, I have thought the physics curriculum was broken ever since I started my first year of grad school and realized we were using 20 to 40 year old textbooks (Jackson, Baym, Goldstein). It's one thing to say classical mechanics hasn't changed in the last 30 years, it's another to say that research emphases haven't changed or that there's nothing different to teach. This is one reason I moved into astrophysics.

One major problem I see is that real working scientists (and people in industry for that matter) spend a lot of time writing, and doing open ended projects, and the curriculum teaches neither. The traditional physics curriculum is an endless march of problem sets, and if some of those could be replaced with work that gave students a background in writing a research proposal, everyone would benefit.

The traditional physics curriculum is an endless march of problem sets, and if some of those could be replaced with work that gave students a background in writing a research proposal, everyone would benefit.

The traditional Japanese language curriculum is an endless march of memorizing rules of grammar, and if some of those could be replaced with work that gave students a background in driving Japanese cars, everyone would benefit.

I agree with Ben #3, let's hear your hypothesis Steinn ;)
Actual real-world astro. research is complex, messy & involves being able to contort your brain to think about things in a way that other people haven't. We don't teach this at undergrad. & tend not to at grad. I'd like to see a complete overhaul of science education here & around the world.
We teach bunches of facts or rules or laws (& sometimes theories)- that's not science as she is done. Science is process. A never-ending process. It is hard, but people can do it. We lose too many people who don't feel creatively challenged. We need to start them researching way earlier than we do & support them in it.

Michael, don't be snide.

Perhaps I didn't phrase myself clearly. I don't mean that students should be assigned to write HST proposals. Rather that they should get projects that allow them to practice the skills they'll eventually use: designing a project, writing up how they intend to address the question, carrying it out, and writing up the results.

I was an English major for a while in college and one of the things I had to learn was how to spew out a coherent 2-3 page paper in a couple of days, to be used as a basis for class discussion. This was incredibly helpful because it actually is a closer parallel to what I do for work than solving algebraically complex problems from Jackson is.

Sorry, I guess that was a bit snarky.

I teach calculus-based physics to first-year and second-year students at a university. Most of them are engineering majors, with the remainder mostly physics, chemistry and math majors. We are under pressure to prepare these students for future courses in their majors; thus, we are told, "be sure to cover X, Y and Z". You can figure out what X, Y and Z are by examining any of the giant university physics textbooks on the market these days.

A fair fraction of my students need a good deal of time to absorb concepts which are used over and over again in this first-year curriculum. For example, how can one break an extended object into pieces, apply some rule (for example, the calculation of moment of inertia) to each piece, and then integrate over all the pieces? Even after devoting 2 or 3 or 4 hours to this concept, scattered over several weeks, some students just don't get it.

Another fraction of the students need time and repetition to learn how to do some basic operations: solving 2 or 3 equations for 2 or 3 unknowns, taking derivatives or integrating a function like A*exp(1 - x/T).

It's not possible to cover all the required topics in a single year and also provide enough time to help a good fraction of the students to grasp some of these concepts. Our department has over the past 15 years discarded a number of topics from the first-year physics course, and we've also tested the idea of increasing the number of class hours per week from 6 to 8. We remain under pressure because the number of students receiving less than desired outcomes is far from negligible.

When I read comments like yours, which advocate spending less time doing basic math and physics, and more time doing other things, I grow frustrated. How can I add additional topics and exercises to the course when a good number of students are struggling with the basics as it is?

In a vague way, it reminds me of the long-ago conversations about the New Math, and the relative importance of understanding sets and theory versus the ability to multiply two numbers (I'm sure I'm oversimplifying, but that's what little I remember ...). Or the discussions over writing: is it better for a student to write essays about his life and his feelings, or to learn proper grammar?

I don't think that there are any silver bullets, and I'm bothered by the academic community's persistent search for those silver bullets.

My comments about fewer problem sets and more open ended assignments were intended to apply to upper division classes, especially for majors, not to intro classes. There is a lot of attrition out of the sciences in upper division as well as at the intro level and it's not just because it's hard, but because the traditional upper division classes are unrewarding.

I think the intro class needs a rethink as well and that is what the Physics Education people are trying to do, while Steinn may be expressing skepticism of it. Nevertheless I think their ideas about more interactive exercises are good.

I taught intro physics to engineers/premeds as well when in grad school and it is a frustrating experience. Something like 10% of the class can figure it out and doesn't need you that much, 30% of the class is capable of learning it if you have enough time to bring them up to speed on all the stuff they should have been taught before getting to you, 30% can just scrape by but will never master the concepts, and 30% are not going to get it - some of them may pass, but you can't make up for their previous lack of preparation. Which 30% is the instructor there for?

Anyway, it's a real problem that quality instruction, interactive exercises, and teaching math that they should have learned already, take time and that means covering less material. My feeling is that we should have told the engineers that our goal is to teach the students some physics well, not all the physics the engineers demand less well.

The other problem is that we're being required to make up for all the crap they didn't learn in high school. But if you talk to someone who teaches, say, English or writing, they have the same problem.