Re: [Phys-L] Commentary on teaching in Physics Today

[John Denker <jsd@av8n.com>]

On 03/08/2017 08:00 AM, Bill Norwood via Phys-l wrote:

> where is the proof that persons educated in Feynman's manner, indeed
> did perform better in their careers?

There is little proof ... and (!) less and less as time goes by,
for the simple reason that success is increasingly /defined/
by scores on multiple-guess trivia tests.

OTOH you can take comfort that there is absolutely no evidence in
the other direction.  Almost any discussion you try to have in a PER
forum will attract assertions that being super-extra-quick at basic
arithmetic is a prerequisite for doing higher math ... in defiance of
all real evidence.  (You can find papers that quote other papers that
quote other papers, but the sequence never converges to real evidence,
and there are plenty of documented counterexamples.)

> - I was one of those students who could not memorize anything unless I
> fully understood it.
> [...] it did not serve me well grades-wise, and prevented career
> opportunities where more of my ability could have been used sooner.

That's a problem.  It's not a particularly new problem.  Consider the
mathematical tripos at Cambridge in 1837.  It was largely a test of
speed when doing simple arithmetical calculations.  George Green came
in fourth.  This is a guy who had already written a book in which, on
page 1, he introduced the idea of Green function.  My point is, if
you're evaluating mathematicians, and George Green comes in fourth,
you're doing something wrong.

Closer to home, to get a teaching certificate in Pennsylvania requires
a bachelor's degree and a 3.0 or better GPA.  That means that half of
the students graduating from Princeton are officially considered too
stupid to teach first grade in Yardley, PA -- 15 miles away.

On the day-to-day level, typical textbooks are chock full of end-of-chapter
exercises that can be solved in a single step and require little more
than hunting through the equations in that one chapter -- not even
previous chapters.

We need to push back against this at all levels, top to bottom.

Suppose there is somebody with a C- grade average and a 990 on the
physics GRE.  Or conversely an A+ GPA and appalling test scores.  Or
middling scores and grades plus a portfolio of real-world accomplishments.
Some of these folks flame out.  OTOH some of them, given a bit of
mentoring, turn into real superstars, which quite fun to watch.  The
point is:  make sure your admission and recruiting processes are fair
to such folks.  Seek them out;  don't filter them out -- not without
giving them a good hard look.

At a much lower level, here's a simple starting point:  Every time
you assign homework, take a minute to see if one of the problems can
be rejiggered so as to make it less likely to reward equation-hunting
and more likely to reward creativity and insight.  It doesn't have
to be fancy.  It doesn't have to be a Feynmanesque perpetual motion
machine.  Just something to make it less formulaic, less cut-and-dried,
more connected to the real world.

There are a lot of smart people on this list.  If each person cooked
up one improved question per week, and shared it with the list,
pretty soon it would be an awesome resource.

One more thing:  Make a little sign and hang it over your desk,
saying "Don't Teach Wrong Stuff."  The textbooks nowadays are full
of bugs.  In a 1000 page book it is not hard to find 1000 errors,
just counting physics errors, not counting approximations, not
counting typos, and not counting multiple occurrences of the same
error.  Forcing students to learn stuff that cannot possibly be
true is the opposite and the enemy of critical reasoning.

  It doesn't have to be that way.  Getting back to Feynman, who
  featured prominently in the Physics Today article:  You would
  be hard pressed to find 10 significant physics errors in the
  Feynman Lectures on Physics.