(I love to watch where a thread starts and where it ends up... A
subject for a study in psychology or complexity, perhaps?)
I'm going to stick my neck out here. From the suggestions from
contributors, the general consensus appears to be that if we provide
non-plug-and-chug problems, things would improve -- i.e., we need to
teach concepts. OK, how does one do that? From looking back at my
school days, I seem to think that I learned concepts by doing problems
(e.g. Halliday and Resnick). I had the distinct recollection that I did
not necessarily understand the concept when I started doing problems in
a particular topic, but that the concept came much later (in an "AHA"
moment). I don't remember doing anything consciously to get to this point.
Also, I think it was easier because it is my instinct that, when given a
problem, I tended to take it apart (literally and figuratively :)). I'm
not sure where I learned this, or whether some people are born with it
(and are therefore condemned to become Physicists and Engineers). This
curiosity (maybe?) about the real world appears, to me, to be essential
for learning Physics. If students want you to give them plug-and-chug
problems, does this mean that they have lost this sense of curiosity
somewhere along the line? (I think all humans have this at the
beginning of life.) Is this something they can reacquire, or is it
something that's gone for ever?
An aside: in the problem about the roller coaster by David Marx, the
concept being taught there concerns gravitational potential energy, and
that is why you don't need the mass (the assumption being that the
gravitational mass and inertial mass are the same). However, if instead
of climbing the second hill, the cart hit a spring and compressed it,
then you would need the mass.