Date: Thu, 28 Feb 2002

To: k-12sd@sysdyn.mit.edu

From: Tim Joy <tjoy@jps.net> (by way of Nan Lux)

Subject: Holy SyM*FEST, StellaMan!

13 March @ University of Portland

At Tubman Middle School, students work on health and salmon models, draw behavior over time graphs of stories and populations, while some of the older students build and study the Mono Lake model described in Professor Andy Ford's work Environmental Modeling.  At Franklin High School, students there construct a theoretical model based on the long wave shifts of human ideas.  Cross town at Wilson High School, students study the Ross Island Lagoon and construct a model to test theories of its condition.  And at Boeckman Creek Elementary, fourth graders put stock and flow diagrams together based on the garden outside their classroom.

William Blake said that we ought "to see the world in a grain of sand."  From what I've heard over this last month, our students are evidently getting the idea.

                      OHSU/SimHealth

                       503-494-0183 (FAX)

                       bongj@ohsu.edu

Xxxxxxx

Subject: Re: Chemistry is also part of the SD family (In response to Scott Guthrie's comments)
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: "Rafael Soto/RS2/CC01/INEEL/US" <RS2@inel.gov>
Date: Thu, 28 Feb 2002

In response to Scott Guthrie:
Most of the world we live in is "full of chemistry", yet we don't even try
to understand it, we simply accept the observable results at face value.
The reality is that a good many chemical reactions have two arrows
(implying the reaction can go in either direction), depending on conditions
(temperature, pressure, pH, ...). Even single-direction reactions have a
reaction rate, which may vary with the concentration of reagents on either
side of the arrow, seeking a balance. All of this suggests "feedback
loops" and "stocks of reagents" - dynamic systems. Reaction kinetics
studies reaction rates (among other things) and we typically see
differential equations describing the behavior of the system; again,
feedback loops.
Chemistry is a "classic" subject to which apply system dynamics concepts
and let the students simulate the potential results of any reaction.
Granted, it is not the same as a real lab, but if the students modeled the
reaction before going to the lab it would make the concept (and experiment)
much more durable.

Rafael Soto
208-526-4250

Xxxxxxxx

From: "Scott Guthrie" <sguthrie@pps.k12.or.us>
To: "k-12sd" <k-12sd@sysdyn.mit.edu>
Subject: RE: Soto: Chemistry is also part of the SD family
Date: Tue, 5 Mar 2002

Yes, Chemistry is very much a dynamic system. Perhaps the implied
"tongue-in-cheek" emotion needed to be made explicit (it was edited out by
me, since I felt everyone would know that it was implied!).
However, gathering reaction rate data is difficult as the probes and other
equipment are beyond the price range of most high schools. Models of these
[chemical reaction rate] systems at this level are good examples of
hand-waving "We know it's this way, but we fudged the data." The model is
"correct," the *shape* of the graphical output is "correct," but the actual
numbers and time scales are meaningless.

Scott

Xxxxxxxx

From: "Ron Zaraza" <rzaraza@jps.net>
To: "k-12sd" <k-12sd@sysdyn.mit.edu>
Subject:
Date: Sun, 3 Mar 2002

Ron Zaraza

On Monday, March 4, 2002, at 11:10 AM, k-12sd wrote:
Most of the world we live in is "full of chemistry", yet we don't even try
to understand it, we simply accept the observable results at face value.
The reality is that a good many chemical reactions have two arrows [...] All of this suggests "feedback loops" and "stocks of reagents" - dynamic systems. Reaction kinetics
studies reaction rates (among other things) and we typically see differential equations describing the behavior of the system; again, feedback loops.

The problem here has always seemed to me to be that high school chemistry tends to deal in (static) outcomes, not the transient dynamics of reactions. Can we overcome that problem, so we can do some seriously interesting dynamic modeling (or dynamic systems thinking) in chemistry at the high school level?

...George

In one of the messages posted, someone said that more needs to be heard
from K-12 teachers. Here is a high school teacher who is beginning to
think "systems". Barry Richmond and Tim Joy presented Systems Thinking
to fifteen faculty members from my high school.
The scenario:
Setting: Today's class
Topic: Genetic Engineering
Event: Transgenic salmon are eleven times heavier than normal salmon
and have a faster reproductive rate than normal salmon.
Question posed to students after a cursory examination of other
transgenic organisms: Do you see any unintended consequences to creating
transgenic salmon?
Response: Lots of answers. Many of which I was surprised to hear from my
students.
Those of you who are experts in this field, how would you use the tools
of ST to develop this scenario more fully?
This really did happen today.

Thanks,

Della Robertson
Norwalk High School

Xxxxxxxxxxx

In my previous reactions to Jay and Ed's comments about the importance
of computer simulation to the practice of Systems Thinking/System
Dynamics, I was critical of what I perceived to be the "us and them"
tone of their comments. I do not wish to diminish my previous
reactions. However, in the ongoing conciliatory thread of the
discussion, I think there is an important distinction that is being
lost. The distinction is that between "us, and them," and "good, and
bad, practice." I would like to clarify this distinction because I
think it is a very important one to keep in mind as we move forward in
the field.
My response to Jay and Ed's comments was triggered by statements that
implied (or explicitly stated) that people who were not doing computer
simulation were harvesting only a small percentage of the learning
that's available, or that their efforts were "superficial," or that they
were not "one of us" ("us" being the real system dynamicists). Such
implications and statements do damage to the cause of promoting Systems
Thinking/System Dynamics because they create an "us" and "them"
environment. I believe one of the lessons that result from
internalizing ST/SD is that it is often more useful to view the world
continuously, rather than as discrete chunks. I feel we'd be much
better off by acting on this lesson, and avoid creating "us/them"
distinctions. Let's instead just embrace the fact that we are all
trying to improve the average quality of thinking in the world, and that
we believe ST/SD has something important to offer in this regard.
However, avoiding "us and them," does not mean abdicating responsibility
for advocating "good practice," or for constructively criticizing "bad
practice." ST/SD offers a "disciplined" way of thinking. It is not
"anything goes!" I have seen many things that I consider to be "bad
practice." And when I do, I say something about them-but I always try
to do so in a respectful way, and to offer a rational and experiential
basis for my opinion. For example, I think it is "bad practice" to
create very high "wire and converter density" stock/flow maps with
resulting unreadable variable names and undecipherable logic. In this
case, I think the rationale for my position is straightforward-it's not
possible to see what the thinking is!
I would like to develop a second example in a bit more detail, because
the rationale appears (in my experience) to be less straightforward-and
therefore pushes the defining edge of what I am saying. The example has
to do with "causal loop diagrams."
I think it is "bad practice" to use causal loop diagrams to represent
the "structure" of a system in anything but the very late "communication
stages" of a Systems Thinking effort-if I do use them (which is
extremely rare), I always do so very cautiously! In making such a
statement, I am not saying that people who do make more use than I of
this tool are doing "superficial work," or even that they are not
inspiring significant learning. I have seen skilled teachers wring gobs
of learning out of a particular arena via skillful causal loop diagram
analysis! Bravo! However, after offering congratulations, I will
always make the point that the causal loop diagramming language does
violence to one of the fundamental "good practices" of ST/SD. That is
the practice of seeking to represent systems in the way they actually
work (i.e., capturing "the physics")!
I have labeled the thinking skills involved in this practice,
"Operational Thinking." Operational Thinking is the first of the
Systems Thinking representational skills. It is based on the assumption
that the world is best represented as consisting of stocks and flows
(i.e., that it is essential to distinguish between states and rates of
change in representing reality, and that you cannot really construct a
functional feedback loop without them!). Operational Thinking brings
with it a significant structuring discipline in the form of unit
consistency and conservation laws. Applying the discipline enables one
to make accurate inferences about dynamics. Stocks and flows exist even
if one chooses not to represent any associated feedback loops!
Stock/flow infrastructures can generate interesting dynamics even in the
absence of feedback loops.
The causal loop diagramming language, by contrast, carries no real
discipline. Anything can be connected to anything else; there is no
distinction between stocks and flows; there is no unit consistency, and
conservation laws are not respected. As a result, it is a much less
reliable language for making inferences about dynamics. Strike one. In
addition, causal loop diagrams are ready vehicles for implementing
closed-loop Laundry List Thinking (or Factors Thinking)-the very
antithesis of ST/SD. At the root of this type of thinking is
correlation, not causation! Causal Loop diagrams do support Closed-loop
Thinking-the second of the Systems Thinking representational skills-but
they do so at the expense of violating the premises of Operational
Thinking! Strike two. Finally, in my teaching experience, if someone
first internalizes the causal loop diagramming language as a
representational tool, they will have significantly more difficulty in
learning to use the stock/flow language (later on) than someone who has
not done so. Strike three.
Three strikes (sometimes fewer) garner a "bad practice" characterization
from me. This is especially true given that there is an excellent
alternative (the stock/flow language) that avoids all of the
difficulties associated with causal loop diagramming. It is highly
disciplined, and based on a view that distinguishing between states and
rates is fundamental. Yet, at the same time, it supports representing
feedback loops (Closed-loop thinking), while also supporting Non-linear
Thinking (the third of the representational Systems Thinking skills) by
enabling graphical functions to be depicted.
The "knock" on the stock/flow language is that it is "more difficult" to
learn than causal loops. Although I think this is true for most people,
in my experience, it is not very much more difficult (and for some, it's
easier!). The perception of difficulty, I believe, springs from the
fact that people associate the stock/flow language with numbers,
equations and computer simulation-and these things are certainly more
difficult for many people than causal loop diagrams. Jay and Ed's
comments reinforced this perception.
However, it is very important to recognize that the stock/flow language
stands on its own-independent of numbers, equations, and computer
simulation! That is, and this is important, it is "good practice" to
think in the rigorous way required to construct a good stock/flow map,
even if you do not decide to continue with the process of transforming
that map into a computer simulatable model! I am not saying there are
no learning benefits to be realized from continuing with such a
transformation. What I am saying is that it is not "bad practice" to
not continue! Someone who does not continue, in my view, has executed
"good practice." Could they learn more by continuing along the
continuum of rigor? Absolutely! But they have executed "good practice"
as far as they've gone. They are operating at a legitimate position
along the continuum. That position, as does all others, has costs and
benefits. This contrasts, for example, with someone who has constructed
a causal loop diagram as a primary representational map. Such a person,
in my view, is not executing good practice because they are violating
thinking principles that are fundamental to the definition of ST/SD.
They are therefore not at a position along the ST/SD continuum of rigor.
This doesn't make them "bad people." And, I am not saying they are
"wrong!" But I am saying they are not executing "good ST/SD practice,"
and hence they should be encouraged to modify their practice if they
wish to develop as Systems Thinkers. The person who "stops at" the
stock/flow map should also be encouraged-but not to modify their
practice. Rather they should be encouraged to extend it. I hope this
important distinction is clear. I think it is an important one.
I have taken some time developing the previous example because I feel it
makes concrete the points I am trying to make. We should abstain from
either explicitly or implicitly demeaning people who are working hard to
achieve improvements in learning and in seeing the world in new, more
productive ways. But we should not abdicate responsibility for
encouraging "good practice" and for "arguing against" bad practice in
the field of ST/SD. If we hope to improve our craft, we must press for
the highest possible standards in the application of our disciplined way
of thinking. Those standards must be based on a rational and empirical
foundation, and when that foundation is shared with developing ST/SD
practitioners, it should always be done in a respectful manner.
If you've made it this far, thanks for listening!

Barry Richmond
High Performance Systems, Inc
45 Lyme Road, Suite 300, Hanover, NH 03755-1221
Tel. 603-643-9636 - Fax. 603-643-9502

Xxxxxxxx

From: John Heinbokel <heinbokel@vtcommonsschool.org>
To: "'k-12sd'" <k-12sd@sysdyn.mit.edu>
Subject: RE: Robertson: A scenario
Date: Fri, 8 Mar 2002

Hi Della (and anyone else lurking out there),

My immediate reaction to your actual question, based largely on my personal
comfort zone with the various systems tools available, would be to try to
use stock/flow mapping approaches to help the students to refine and present
for comparison/constructive critique their mental models of how those
unintended consequences might develop.
On further reflection, though, I'd like to challenge the idea of moving
immediately or exclusively to the area of "unintended" consequences, which
has the flavor of focusing the kids on the negative consequences of this
development. My guess is that there are all sorts of rational and positive
reasons for actually doing this; getting the kids to recognize that those
exist as well and that this development is being driven by benefits acruing
to someone will be absolutely necessary if the kids are truly to understand
all the push/pull pros and cons in this justifiably controversial idea.

Thanks for your attention.

john heinbokel
vermont commons school
heinbokel@vtcommonsschoool.org
(802) 865-8084

xxxxxxxx

Date: Thu, 07 Mar 2002
From: "RICHARD TURNOCK" <Richard_Turnock@pgn.com>
To: <k-12sd@SYSDYN.MIT.edu>
Subject: Teaching and Learning
 
"Handbook of research on science teaching and learning"
Edited by Dorothy L. Gabel. Q181 .A1 H35 1994
Excellent resource.
With only personal anecdotal evidence of ST/SD results, we don't have
the research that meets the requirements of the reference listed
above. The education system does not respond to opinions.
"Tinkering with Utopia" is a book about all the education reform
efforts that have failed in schools.
Where are the published research results to show improved student
learning using systems thinking, system dynamics and integrating the
tools?

Richard Turnock
Adjunct Instructor, Portland State University
Educational Services,Portland General Electric
121 SW Salmon 1WTC-3-3
Portland OR 97204
Phone: 503-464-8503

Xxxxxxxxxx

Date: Fri, 8 Mar 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: Ed Gallaher <gallaher@ohsu.edu>
Subject: Re: Richmond: Additional reactions

Wow! This has been a very challenging and rewarding discussion. Many
of these issues have been floating around at or near the surface for
many years, but rarely has this much thought been applied to really
trying to listen, understand, and expand our horizons. I really
appreciate the many thoughtful responses and comments!
I agree that it is very important to distinguish between "us and
them" and "good practice vs. bad practice". Our ranks are growing,
but there are still too few of us. We need to work together, and
learn from each other.
Barry's latest message was very well thought out, and I agree with it
almost entirely. We share our opinions of causal loop diagrams. I
have seen several -very- effective presentations using CLDs, from Jim
Hines and George Richardson. They moved seamlessly (and it goes
without saying, correctly) between CLDs, S&F diagrams, and
simulations. We must acknowledge, however, that they are masters at
this game, and CLDs are fraught with peril. George gave another
presentation in which he described a number of ways one can go very
wrong with CLDs, and believe me, it's easy!
Barry's careful example advocates the careful development of S&F
diagrams as an integral part of the ST process, whether or not the
final result is a running simulation. I agree with this viewpoint. In
fact, I think the S&F "language", in and of itself, is highly
under-rated as an intellectual tool.
In support of this notion, I have developed a S&F diagram of human
thermoregulation that I use in a graduate course to illustrate
homeostatic systems. After an introductory lecture I hand out this
diagram and assign the students the task of finding an "appropriate"
physiology text (this is part of the exercise). My diagram is used as
a skeleton on which to organize their reading. Each icon (N ~ 30) is
numbered, and they are asked to write 2-4 sentences describing its
role in thermoregulation. The next lecture is an interactive
discussion; I lead the way, and the students contribute most of the
substance. This is a -very- productive exercise!
Nevertheless, I am still concerned about two important issues.
ISSUE ONE:
I suspect I fall between Barry and Jay on this one; I will
paraphrase what I believe are their positions, and they can correct
me if I'm too far off.
Jay seems to feel that all (or almost all) inquiries should be taken
to the computer simulation stage, or else the end result falls far
short of its potential. Model development is essential, but the real
learning just -begins- during the simulation process itself. In fact,
it is more likely than not that initial simulations will fail because
of faulty "insights", suggesting that without simulation, the work is
likely to be "bad practice". The danger comes from patting ourselves
on the back too early in the process, because it feels so good to
think we our discussions have been "at a higher level".
Barry makes a strong case for the careful development of disciplined
stock-and-flow diagrams, even if the end result does not include
explicit simulations.
Barry says, "It is "good practice" to think in the rigorous way
required to construct a good stock/flow map, even if you do not
decide to continue with the process of transforming that map into a
computer simulatable model!"
Note his use of the words "that map". Jay might disagree, but I agree
here with Barry. I have not developed "that" thermoregulation map
into a computer simulatable model. I found the exercise of creating
the map to be quite useful; I'm also positive (as are Jay and Barry),
that the development of a realistic computer simulation would require
considerably more work, and would provide many deeper insights.
On the other hand, I HAVE built a number of other models, up to and
through the computer simulation stage. I am not overly confident in
my thermorgulation model, but as a result of my other work I am
reasonably confident that it is not utter nonsense.
However, if someone has NEVER built a computer model and run
simulations, it is -highly likely- that their S&F diagrams will -not-
be examples of good practice. The S&F diagrams will contain
structural flaws, unrecognized dimensional inconsistencies, and other
serious problems.
Barry also wrote:
"The causal loop diagramming language, by contrast, carries no real
discipline. Anything can be connected to anything else; there is no
distinction between stocks and flows; there is no unit consistency,
and conservation laws are not respected. As a result, it is a much
less reliable language for making inferences about dynamics."
Undisciplined stock-and-flow diagrams carry no real discipline
either! Anything can be connected to anything else; there is often
considerable confusion between stocks and flows; there is no unit
consistency; and conservation laws are not respected.
CLOSELY RELATED -- ISSUE # TWO:
I am concerned that this ongoing intense discussion, however useful,
also sends the message to newcomers: "Boy, these guys are really
serious about this stuff. Those computer simulations must really be
horrendous. Whoa! I better be careful!" Newcomers have every right to
be petrified!
I realize I am a left brained, computer geek, ivory tower college
professor. And yes, I like the discipline of computer simulations.
But I KNOW from extensive personal experience, with guidance, and
with patience, INTRODUCTORY MODELING AND COMPUTER SIMULATION IS JUST
NOT THAT DIFFICULT!
As part of the CC-STADUS I have taught the simple first-order delay
Rain Barrel model to several hundred high school teachers. Some were
math and science teachers. Many (by design) were social science
teachers. Many had very limited math skills, and in the early years
many had NO previous computer experience.
Ron, Diana, Scott, and Tim can comment further, but the vast majority
of these students were right on track. I received many compliments
for this workshop, and I appreciate the strokes, but the reason this
worked is not because I am a genius teacher. It is because, taught
correctly, it's just not that difficult!
- The model has one stock, one input, one output, and one auxiliary
variable. Build the structure in 5-15 minutes, depending on whether
you know how to use a mouse or not.
- Define the initial value of the stock (zero), the input (2 {gallons
per minute}), the fraction removed (.05/minute), and the output
(Volume * fraction removed). We'll show you how.
- Create a graph and set the range specs. We'll show you how.
- Set the simulation time and dt. We'll show you how, and talk about
some of the issues you should watch out for.
- Predict what is going to happen. Yes, you'll be sticking your neck
out, but we won't chop it off. Take a guess anyway.
- Now, don't be afraid. Go on - push the 'run' button!
- There, that wasn't so bad, was it?
This process takes about an hour, including a preliminary discussion
and a hands-on computer lab session.
Additional simulations use STEP functions to turn the water input up
or down (or the intravenous infusion), and the PULSE function to pour
in buckets (or drug injections).
By the end of the day virtually everyone is reasonably comfortable
with single and multiple drug doses, intravenous infusions, and the
effect of kidney or liver damage on the removal of drug. Yes, it's
new and unfamiliar. Yes, they need much more practice, and time to
absorb the new concepts. But they did it!
In every workshop, one or more participants approached me during a
break and asked surprisingly insightful questions about personal or
family medical issues. "Where can I find the halftime of
phenobarbital? My daughter has epilepsy, and when the dose is too
high she is drowsy; when it's too low she has seizures. This makes a
lot more sense now." I was incredulous when this happened the first
year, but then it happened -every time-.
I feel we are doing more harm than good by tiptoeing so cautiously
around this issue.
We're not asking someone who is afraid of the water to swim across
the lake. We're simply encouraging them to walk in up to their
waist, and we'll stand next to them the whole time.
If we spend too much time wringing our hands we're going to make them
even more nervous. Let's all relax, lower the barriers, be casual and
supportive, turn on the computer, and get on with it.
Don't feel obligated to turn every discussion into a simulation. But
don't be afraid to get your feet wet either.

Ed Gallaher

Xxxxxxxx

Date: Fri, 8 Mar 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: Ed Gallaher <gallaher@ohsu.edu>
Subject: Della; More info?

Della -
The initial description of your classroom experience was intriguing.
However, could you provide a few examples of the issues that were
raised by your students? And perhaps some the interactive comments
that evolved during the discussion?
I realize you are new at this, and I want to make someting very
clear; this is not a trick question!
I am not looking for "evidence" that newcomers are going to screw
everything up because they are not yet expert computer modelers.
You have made a number of helpful comments recently, and your early
(and continuing) experiences will be enlightening to all of us.
You started with the best with Barry and Tim. Welcome to the club!
It's great fun! :-)

Ed Gallaher

Xxxxxxx

There is something missing as I think, from the teaching and using of GIS,
STELLA, html and many systems that are said to "foster thinking." These real
world learning tools usually are the focus instead of the learning. That is
often what happens when technology is used to expose students to real world
tools.
The first concentration is on the technology to try to learn the program.
Too many times so called real world tools are flung at students and then
let's see what happens. However the best way to get the most out of the tool,
is to focus more on the stages of education that the student must learn to
learn to apply the tool, not learn how to use the tool.
While there are many worthy projects utilizing real world tool, the key
should rely more then just gathering data and analyzing data. Is that what
system thinking is? Should it not be visualzing the processes. I really feel
that there to much reliance on connecting the parts (stocks, flows etc) and
not visualizing the whole concept and breaking it down to the components.
Maybe the reliance on this mode is the it is easier to check and tell whether
someone is showing the process in the right way instead of developing the
ideas to show process and the stocks, flows etc help visualize this whole
process.
I think there should be more visualizing in systems thinking and
concentrating on that. Sometimes you have to think backwards instead of
frontwards.

Dr. Eric Flescher (KCStarguy@aol.com)
Project S.I.M. (Simulations, interdisciplinary Internet and Metacognitive
activities)

Xxxxxxxxx

Date: Mon, 11 Mar 2002
From: Della Robertson <frobchen@earthlink.net>
To: k-12sd <k-12sd@sysdyn.mit.edu>
Subject: Re: Gallaher: Della; More info?

Ed,

Thank you for your comments; they are encouraging. Today, as I lectured on the
topic of "Community Ecology" it became more and more apparent to me
how Behavior
Over Time Graphs can be very useful. We were examining predator-prey
relationships. I need to build that strategy into my teaching.
I will go back to the original question on the unintended consequences of
transgenic salmon. Another respondent suggested that I use other language which
would open students up to the positive aspects of this technology. I
suppose the
traditional "advantages" and "disadvantages" would suffice, but I don't think
that kind of language moves students to look beyond the obvious.
I am hoping to spend some time devoted to having students identify stocks and
flows (as has been suggested) within the context of the state
standards, which I
believe to be helpful.

Thank you for taking the time to respond.

Della Robertson
Norwalk High School

Xxxxxxxx

From: "John Gunkler" <jgunkler@sprintmail.com>
To: "'k-12sd'" <k-12sd@sysdyn.mit.edu>
Subject: Robertson: A scenario
Date: Mon, 11 Mar 2002

In response to suggestions from John Heinbokel, I'd simply ask students
to create stock-and-flow diagrams that included various "consequences"
(outcomes) of the development, use, release (etc.) of transgenic salmon
-- without first classifying these consequences as "intended" or
"unintended" (very slippery notions, anyway, since they presume
knowledge of someone else's intentions), nor even as "positive" or
"negative." Then by seeing how these consequences affect other outcomes
you could foster a genuinely interesting debate about whether
Consequence X is positive (i.e., a reason to develop transgenic salmon)
or negative (i.e., a reason not to do so.) Ideally you would simulate
the development and use of transgenic salmon to determine how
consequences affect other consequences. I think simulation is almost
essential here, since I suspect the really interesting dynamics of such
a model would include shifting loop dominance over time, with early
outcomes being "positive" and later outcomes either adding to or
subtracting from the early results.
From the language in the original message ("unintended consequences") I
presume that the teacher is assuming that later outcomes will be
negative -- or is at least asking the students to create arguments that
this is the case. While it's never a bad guess that initial outcomes
will differ from later outcomes in any complex system, it's also
probably not a good idea to presume to know exactly what kind of changes
will occur without actually simulating the behavior.
Using (stock-and-flow) model building as a way to foster the kind of
discussion you want is very powerful; more powerful than merely asking
students to come up with laundry lists of initial reactions. I think
you'll find that it is in the process of trying to put together a model
with some face validity that truly enlightening discussions arise.

John W. Gunkler
jgunkler@sprintmail.com

Xxxxxx

Date: Tue, 12 Mar 2002
From: Della Robertson <frobchen@earthlink.net>
To: k-12sd <k-12sd@sysdyn.mit.edu>
CC: Frank Zepeda <zepeda_frank@nlmusd.k12.ca.us>,
Dina Leslie <leslie_dina@nlmusd.k12.ca.us>
Subject: Re: Gunkler: reply to A scenario

John,
Thank you for your detailed response. I want to encourage those of you who are
experienced with ST to respond in such a detailed way. I have an appreciation
for ST and what it can facilitate in terms of thinking, but I need all the help
I can get with developing the tasks which encase and unfold the learning.
A component of our specialized curriculum is called "Computer Applications in
Math and Science", CAMS, for short. We have been having problems with
the server
crashing multiple times per day, but I am going to take what you suggested and
have the students work with pencil and paper and discuss their results. From
there, we can move into some simple STELLA tools. Barry and Tim emphasized the
importance of the "story" first.
Thanks for your response.

Della Robertson
Norwalk High School

Xxxxxxxxxxxx

Barry Richmond sets the right tone when he puts ST and SD on the same
continuum instead of viewing them as separate "camps". One other thing worth
saying about this continuum is that students and teachers can move along it
in both directions. They can start with a problem, develop BOTG's, causal
loop and/or stock and flow diagrams, write equations and finally do
simulations to better understand the system they are dealing with. This
direction will work for many, but may prove daunting for students and
especially teachers who have limited experience in developing models.
The alternative would be to start with a simulation model, one that someone
else has developed, and learn a lot about modeling by "deconstructing" that
model. A model that shows some intriguing behavior could quickly get them
excited about simulation. Students could then do all sorts of sensitivity
tests, add variables, add loops and cut loops, put delays in and take them
out, change the shapes of curves, and try different functional relationships.
This is similar to the way kids used to learn how radios worked by taking
them apart (before everything was put on one chip). Students would get a
better understanding of the systems they are studying, of systems in general,
and of how one models systems. They could finish at the other end of the
continuum by drawing causal loop and/or stock and flow diagrams that display
the feedback loops they discovered and help "tell the story" of how the
system works.
Teachers experienced in ST/DM have probably learned how to do this with
existing models. Teachers new to the field could use some help and might
find this a more comfortable way to get into simulation modeling than
building models "from scratch". After studying a few models in this manner,
they would probably have the confidence to develop their own and also have a
sense of "good practice" that will guide their model building.

Gary Hirsch
GBHirsch@aol.com

Xxxxxxxxxx

Date: Tue, 12 Mar 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: Ed Gallaher <gallaher@ohsu.edu>
Subject: Re: Robertson: with more information

Della -

You are making great progress, and obviously taking care to
incorporate these new ideas into the current educational climate.
Your observations will be useful to many others on this list serve,
so don't be bashful about keeping us posted. Successes and rough
spots or confusion will be equally valuable.
Re "unintended consequences". I saw that comment as well. I believe
the intent was to avoid focusing primarily on using SD/ST to avoid
disasters. In contrast, we can use ST/SD to gain insights into the
real underlying issues, thereby making policy decisions that minimize
the unintended consequences.
Both viewpoints are important, but one is considerably more
optimistic and uplifting than the other.
There is no need to change the wording. The term "unintended
consequences" is familiar to the SD/ST community, and communicates
just what it should. A major reason we are faced with unintended
consequences is because they are also unexpected consequences.
A good SD/ST effort can begin to predict the long-term influences of
a given policy decision, and indeed, provide "unexpected" insights.
Together with thoughtful value judgements, economic and political
constraints, etc., it should be possible to minimize the "unintended
consequences".
And you are quite right. S&F diagrams, along with BOT graphs, enable
a new level of inquiry and discussion.

Keep up the good work!

Ed
--
Edward J. Gallaher, Ph.D.
VA Research Pharmacologist
Research Service R&D17
Veterans Affairs Medical Center
Portland, OR 97201
(503) 220-8262 x56677
Associate Professor
Depts. of Behavioral Neuroscience
and Physiology-Pharmacology
Oregon Health Sciences University

Xxxxxxxxx

Date: Fri, 15 Mar 2002
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>
To: K-12 discussion group <k-12sd@sysdyn.mit.edu>
Subject: Detailed ideas

Della wrote:
Barry and Tim emphasized the importance of the "story" first.

Ed Gallaher wrote:
Your observations will be useful to many others on this list serve

I can second that wholeheartedly. Living here in Germany in relative
SD-isolation, I'm always glad to see "broad-brush" descriptions of
teaching ideas accompanied by crucial details like Della's comment
above. These details give me what I need to try out the ideas for
myself.

Amongst other things I teach software engineering, where a modern trend
is to emphasise the narrative of the user's interaction with the
software system being built, so thinking about the importance of the
"story" sends my mind spinning off into ways of mapping the idea of
narrative into my teaching of maths (which is where I most use SD).

Thanks Della.

Best wishes,
Niall.
 
Xxxxxxxx
 
End of March, 2002