🚀 go-pugleaf

In article <1992Jul29.212844.15222@uvm.edu> cavrak@emba-news.uvm.edu.UUCP (Steve Cavrak) writes:
>Herman Rubin  comments :

>	Unfortunately, at this time the great bulk of material
>	presented is exactly stimulus-response.  It is trying to make
>	humans act like computers.  What is needed in education is
>	concepts.  There is a big amount of talk of transfer of
>	training; what I have seen on this is either negative, or it
>	is application.  

>There is no doubt a large amount of the "stimulus-response" type of
>thing going on out there -- but from my experience there is as much if
>not more high quality "though provoking" material.  Some, like
>Tarski's world has a "game like" approach, but at a very high level.

>One of the key points to remember is that "concepts" can only be
>taught in the context of "experience."  The "stimulus-response" model
>is a very one dimensional experience; a program like MathMaze,
>however, and Tarski's world, provide fairly rich experiences -- always
>under the control of the learner who can modify twist bend flip the
>rules to suit her curiosity.  

As one teaching concepts in mathematics and statistics, I would argue
that it is more important NOT to have too much experience so that the
concepts can be developed without being swamped by the details.  Not
that examples should not be given, but that they should be chosen 
carefully to be so varied that the student does not carry irrelevant
garbage to confuse the concepts.

>Yes, this does take a lot of developmental work -- probably as much as
>writing a new textbook.  But imagine a math textbook with Mathematica
>built in; a statistics textbook with JMP and several years worth of
>opinion polls and economic data; an astronomy text with tons of images
>of the planets (gis type); an environmental studies text with 
>stella, etc.

I can see something like Mathematica being included in a good textbook
teaching mathematical concepts.  But the stuff you have mentioned has
no place in a statistics book until the concepts, none of which are
even illustrated by the methods and data suggested above, are clearly
understood.  (Well, the opinion polls might be all right for examples,
but not as usually given.)  It is very difficult to get a person who
has learned statistical methods to discuss the statistical problem 
with a statistician; the ingredients are not known at all.  This 
statement definitely includes the great bulk of those who misuse
statistics in psychology and education.
-- 
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
Phone: (317)494-6054
hrubin@pop.stat.purdue.edu (Internet, bitnet)  
{purdue,pur-ee}!pop.stat!hrubin(UUCP)