Xref: utzoo comp.ai:1456 comp.edu:1016 comp.cog-eng:511
Path: utzoo!mnetor!uunet!husc6!ncar!ames!ucsd!sdcsvax!ucsdhub!hp-sdd!andrea
From: andrea@hp-sdd.HP.COM (Andrea K. Frankel)
Newsgroups: comp.ai,comp.edu,comp.cog-eng
Subject: Re: Becoming CAI literate
Message-ID: <1219@hp-sdd.HP.COM>
Date: 16 Mar 88 00:09:38 GMT
References: <2960@orstcs.CS.ORST.EDU> <1988Mar2.125247.28809@lsuc.uucp> <934@aucs.UUCP> <707@l.cc.purdue.edu>
Reply-To: andrea@hp-sdd.UUCP (Andrea K. Frankel)
Organization: Hewlett-Packard, San Diego Division
Lines: 48

In article <707@l.cc.purdue.edu> cik@l.cc.purdue.edu (Herman Rubin) writes:
>Another problem is that the books are almost universal in slighting the 
>concepts in favor of manipulation.  One can question whether this is not
>unavoidable.  For example, the calculus course has three concepts, and
>each of them can be stated, and well illustrated, in a few pages.  I think
>that it should be considered criminal to pass a student who does not have
>a reasonable understanding of those concepts.  I would be surprised if
>20% of the students taking calculus meet these standards.  I am not even
>sure how many of the teachers understand the concepts!  The bulk of 
>the course is devoted to those manipulations which can easily be done by
>computers.  The ability to do these manipulations is useful, but without
>the concepts is even dangerous.
>
>A problem with teaching concepts is that one student learns them in minutes
>while another takes months.  The solution clearly requires that we abandon
>some of the structure of our educational system, and our evaluations should
>denote only knowledge and ability, and time and effort should be ignored.

The only course I ever took which had a fairly satisfactory solution to
this problem was the frosh physics course at CalTech.  We used three
different textbooks:  the Feynman Lectures (big red books) for pure
concepts, the L/V problem set books that went along with them, and
"Holiday and Grundgenik" (I forget the real name) to learn the mechanics
of manipulation of formulae and such.  Lectures were based on Feynman
and were 100% concepts; recitation sections were based on the other
books and were devoted to teaching us how to do the problems.

If you slacked off on reading Holiday and Grundgenik, doing the
homework and attending recitation sections, and spent all your time
grokking the great Feynman prose, it was possible to learn all the
concepts beautifully and not be able to solve a single problem on the
tests.  However, if you managed to scribble convincingly on your test
in a way that demonstrated mastery of the concepts involved in the
question (but without making any headway on getting the answer), it was
possible to pass (but just barely).

I understand that they have since discontinued the three-textbook
approach.  Pity.  Feynman was the only thing that made physics
remotely palatable!

Andrea Frankel, Hewlett-Packard (San Diego Division) (619) 592-4664
                "...like a song that's born to soar the sky"
______________________________________________________________________________
UUCP     : {ihnp4|decwrl|sun|tektronix|<other hub>}!hplabs!hp-sdd!andrea 
                                  or {nosc|hpfcla|ucsd}!hp-sdd!andrea 
Internet : andrea%hp-sdd@ {nosc.mil | sdcsvax.ucsd.edu | hplabs.HP.com}
CSNET    : andrea%hp-sdd@hplabs.csnet
USnail   : 16399 W. Bernardo Drive, San Diego CA 92127-1899 USA