Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

[Phys-L] playing for keeps



On 06/28/2013 07:29 PM, Bruce Sherwood wrote:

Apparently it is well established by experiment that in general forgetting
follows a fairly universal power law.

Only if you let it!

My understanding of this issue has been tremendously improved
by the recent discussions.
-- Until about an hour ago, I would have said something like
this: There is always a tendency to teach to the test. That's
either a good thing or a bad thing, depending on the test.

-- Until about an hour ago, I would have said that when training
private pilots, I teach to the test, and am happy to do so,
because it's a good test.

Now, however, I would like to say something better: I teach
to the /standard/. I'm playing for keeps. That is, I want
the student to meet the standard on the day of the test, and
on the day after, and two years down the road, and twenty
years down the road. This is a much taller order! Playing
for keeps includes teaching to the test, plus a lot more.

I claim I've always played for keeps. I've always explained
this to students. Recently I've gotten better at explaining
it to third parties.

Interestingly enough, the key document is called
_Private Pilot Practical Test Standards_
http://www.faa.gov/training_testing/testing/test_standards/media/FAA-S-8081-14B.pdf
so we really do have a standard, not just a test. The standard
controls the test, and not vice versa. The standard is routinely
used for a lot of other things, not just the test. It's a good
standard.

So, how is it possible to evade the allegedly "universal law"
of forgetting? Well, I confront the thing head-on. I tell
students my job is to put myself out of job. That is, my job
is to get students to the point where they can teach themselves.
I make a very big deal out of this.

Before students can take the practical test, I have to certify
that they are ready. At that time, I tell them that I may
not see them again for six months or a year ... but when I do
see them, I expect them to fly /better/ than they do now. I
expect them to read, I expect them to practice, and above all
I expect them to hold themselves to high standards. This is
not new to them on the day of the sign-off; we've been talking
about it more-or-less every day since the first lesson.

If I thought a guy could meet the standards on the day of the
test and not thereafter, I would not sign him off.

===============

The same ideas apply to academic subjects such as math and
physics, except that people don't take such things nearly
as seriously.

All too often, each chapter in the book prepares students to
answer the end-of-chapter problems and not much more. All too
often, the course as a whole prepares students to pass the
almighty end-of-year test and not much more.

My point is simple: If you decide that long-term retention
is one of the essential goals, then it dramatically changes
how you go about teaching. For starters, you find yourself
spending a huge amount of time and effort on mnemonics. It
doesn't matter how true or important or elegant something is,
if it will not be remembered.

The textbooks are, by and large, terrible at this. A good
teacher can help a lot, by passing on the mnemonics, the lore,
the rules of thumb, et cetera. However, it doesn't have to
be that way. There's no reason why the textbooks couldn't
include a ton of mnemonics.

I insist that things can be learned at more than one level,
namely at the rote level and at deeper levels. If you learn
a certain rule by rote, using a five-word mnemonic, it
absolutely does not prevent you from learning where the
rule came from, including the deep physical principles
behind the rule, the provisos, the exceptions, et cetera.
The rote learning and the intellectual learning support
each other, if they're done right.

At the end of the year, if you ask the third-grade teacher
whether the students learned everything they were supposed
to, so that they are fully prepared to go on to fourth grade,
the teach invariably says yes, absolutely yes. Then at the
beginning of the next year, if you ask the fourth-grade
teacher if the students arrive well-prepared for fourth
grade, she says no, absolutely not.

That's /partially/ unfair, because it is partially explained
by human nature: everybody thinks the grass is greener on
the other side of the fence, and everybody wishes the other
guy would do a bigger share of the work. However, there is
another part that is fair and real. Imagine how nice it
would be if the third-, fourth-, and fifth-grade teachers
could all sit down and coordinate with each other: What
can we realistically accomplish? What do we really need
from each other? More importantly, what do the students
really need, long-term?

I say all these teachers should play for keeps. That means
not just teaching the kids stuff in a form that will let
them pass the almighty end-of-year test, but rather in a
form that will stick with them, a form that will be useful
during the next year and forever after.

The same applies to high-school physics and college physics.
We should play for keeps. If we can't find a way to teach
stuff that will actually be remembered, we should give up
go home. We should stop wasting everybody's time.

================

One more thing: Feynman said that knowledge is like a
grand tapestry. A forgotten fact is like a hole in the
tapestry. You can repair the damage by re-weaving up
from the bottom, down from the top, and/or in from the
sides. This is called /figuring stuff out/.

Similarly, William James (1898) emphasized the role of
connections as the basis of a good, useful memory.

Physics is so richly connected that it is hard to forget
stuff. It shouldn't stay forgotten for long, because you
can figure it out.

As an illustration of what I'm talking about: If you
remember Ohm's law by remembering three different equations
V = I R
I = V / R
R = V / I
then you're doing it wrong. The smart approach is to learn
one of those equations, and then figure out the others as
needed, using simple algebra. Bruce Sherwood makes the same
point using a different example: There is no need to memorize
the formula for escape velocity, because you should be able to
figure it out in less time than it takes to ask the question.

I have a terrible memory. I compensate for it by figuring
stuff out. For example, I can never remember the mass of
the proton. I can, however, figure it out, with better than
1% accuracy, in less time than it takes to ask the question.

Bottom line: There is no point in teaching people physics
factoids that won't be remembered. Instead we need to teach
people how to learn, how to remember, and how to think. If
we get students to form the habit of looking for connections
and figuring stuff out, we can stop worrying about how to
measure memory loss over time. The students will get smarter
over time, on their own.