I've stayed out of this discusion so far, because I didn't want to toot
my own horn, but here goes anyway. I agree that canned models _can_
interfere with understanding underlying processes. That can be avoided,
but it takes considerable effort on the teacher's part.
  But another approach is to have students build the models. Then (one
hopes, at least) they not only learn how the particular models work, and
the assumptions that went into them, but also something about the
process of modeling in general and its values and limitations.
  Therese Donovan and I wrote a couple of books showing how to build a
variety of ecological models from scratch in Excel. You could do the
same in any spreadsheet, or in Matlab, as Roughgarden does, or in a
variety of other programming environments. The point is to get the
students to think through the process, rather than handing them the
finished product. 
  I have had the greatest success when I don't give students explicit,
step-by-step instructions, but general guidance and let them really
struggle with the problem. I find I must exercise great patience and
forbearance to let them make mistakes and correct themselves. They get
frustrated, I have to stop myself from fixing their mistakes, and
everything takes far, far longer than it would if I would just give them
the models (or tell them how to make them). I console myself with the
thought that maybe, just maybe, they've really learned something when
it's all over. Why do I think that? Only because I know that as an
undergraduate never really understood models, and only began to
understand them to the limited degree I now do when I began to try to
teach them, and had to struggle through them on my own so I could
explain them to my students. Doubtless understanding comes differently,
and more easily, to those of you who are not so mathematically
challenged as I am, but on the other hand perhaps I am closer to
understanding the mindset of many of our students.
  Finally, I assure you I am not trying to sell books (check them out of
a library), but just in case anyone is interested, here are the
references:
Roughgarden, J. 1998. Primer of Ecological Theory. Prentice Hall.
Donovan, T. and Welden, C. 2002. Spreasheet Exercises in Conservation
Biology and Landscape Ecology. Sinauer Assoc.
Donovan, T. and Welden, C. 2002. Spreasheet Exercises in Ecology and
Evolution. Sinauer Assoc.
Charles


Charles W. Welden
Department of Biology
Southern Oregon University
[EMAIL PROTECTED]
(541) 552-6868 (voice)
(541) 552-6415 (fax)

>>> Dan Fiscus <[EMAIL PROTECTED]> 01/20/06 12:40 PM >>>
John,

I don't have specific references you seek, and am not
really against computer modeling, but I have myself
been questioning the role of computers in education
in general. I think there are studies related to the
negative effects of certain kinds of computer learning,
likely for younger kids like elementary through high
school. My brother teaches elementary and uses
computers a lot and sometimes he says kids that can
do well on the computers can't do well in other arenas.
This suggests to me that the context of learning is
important and that if the computer does too much to
make the context rare, special, glitzy, "inter-active" in
bad ways (providing graphics and multiple choice,
walking kids through by the hand, spoon feeding, like
TV almost, taking away the creative and struggle
and internal work parts) - then the learning may be
dependent on that specialized computer context and
may not be transferable or "embodied" via true
mastery.

Most of this would be moot in a typical undergrad
class where computers and/or computer modeling
are only one part, since the other parts would create
a more general and rich context. But I thought I
would mention it.

Beyond that I think it important to examine that
computers themselves are not sustainable. Thus I
am reluctant to invest lots of blood, sweat and tears
into them for any purpose, as I don't think they will
be here, in this same form, in 100 or 1000 years.
Without an open-ended evolutionary future they
seem a dead-end to me, and I want to contribute to
aspects of science and culture and education that
can "live long and prosper". This is a highly biased
view and also minority view. I think it interesting to
imagine modeling systems that *are* sustainable
over the very long term (i.e., run on renewable
energy and recycling materials, like all life systems).
Such self-examination could perhaps add a neat,
complex, self-reflexive element to computer modeling
- could we model our own modeling process? Where
would computer ecological modeling fit in to an
ecosystem model of a town or university in terms of
energy, materials, information stocks and fluxes and
transformations? Can computer modeling pay its
way, justify its own existence, more than
compensate for what it consumes and degrades?

Some thoughts...

Dan Fiscus

John Petersen wrote:

>Back in September of '05 I sent out an announcement about a conference 
>at Oberlin College that would focus on the role of computation and 
>modeling in the undergraduate curriculum. I was very interested when 
>several colleagues responded on this list-serve expressing a rather 
>negative view regarding the value of teaching modeling to 
>undergraduates. To summarize, the arguments seemed to focus on the 
>notion that the development of specialized and technical computer
skills 
>involved in modeling represents a counterproductive distraction. Has 
>anyone seen this argument made anywhere in any literature? I would 
>greatly appreciate references to papers or book chapters that adopt
this 
>view or otherwise criticize the value of modeling education for 
>undergraduates. Beyond that I would appreciate suggestions for 
>literature that takes any position on the pedagogical role of modeling 
>in the undergraduate curriculum.
>
>Thanks!
>John Petersen
>Associate Professor of Environmental Studies and Biology
>Oberlin College
>
>
>  
>

Reply via email to