February, 2002
Date: Wed, 30 Jan 2002
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>
To: k-12sd <k-12sd@sysdyn.mit.edu>
Subject: Re: What curricula do we need?
Lees Stuntz schrieb:
What pieces of curricula would you
1. Love to have for your classroom?
2. Think are the highest leverage curricula which we need for our current
state of knowledge and expertise?
I am currently constructing a first-year undergraduate maths course
around Stella, and I'm reaping benefits all over the place in terms of
making cross-connections between different course components. Last week,
for example, I was able to make very clear the idea of decoupling a
system of coupled differential equations into n independent DE's, just
by rubbing out links from a SFD on the board. In general I like seeing
standalone pieces of curriculum in which a single idea, which is very
abstract and difficult for students to understand, is graphically
portrayed as a series of stock and flow relationships.
An example which students (and I) find very difficult to represent
visually is the issue of wave phenomena in physics - waves (by contrast
with vibrations) act in both time and space, which makes them hard to
visualise. I've been racking my brains to find a way of illustrating the
following situation in Stella without any success. I'd be very grateful
to anyone who can offer a neat little diagram to represent it clearly:
---------------------------------
A microphone is positioned on the line between two identical
loudspeakers, at an equal distance from each one. The loudspeakers are
fed from the same source and they therefore emit exactly the same sound,
which is a continuous tone of frequency 220 Hz. What is the minimum
distance (to the nearest cm) that the microphone must be moved towards
one of the loudspeakers for the sound to reach an intensity minimum?
---------------------------------
The answer is 1/4 of a wavelength, but I'd love to really give students
a visual understanding of this fact.
Niall Palfreyman.
----------------------------------
To: k-12sd@sysdyn.mit.edu
From: "Frank Duffy" <frank.duffy@gallaudet.edu>
Subject: Re: Managing nonlinear change....
Date: Tue, 29 Jan 2002
Hello,
I am replying to a note from George Richardson where he
said,
„There have been a lot of recent messages to this
listserve talking about Œlinear‚ and Œnonlinear‚
notions applied to management, school change, cause and
effect, and so on. I'm worried that we don't all mean
the same things by these terms, and that some good
ideas are getting jumbled with jargon.‰
George, your concern is well-grounded. For me, the
confusion in language comes from the difference between
the technical, mathematical definitions of linear,
nonlinear, etc. and the metaphors that are derived from
those concepts.
For people interested in managing change in social
systems, the mathematical definitions are viewed as
irrelevant, abstract, and confusing (no offense
intended :o) ). Given these characteristics, others
are attempting to translate the technical terms into
metaphors for understanding how to manage change in
social systems.
I am one of those who reads the theory and tries to
convert it into plain English metaphors to help
teachers and school administrators understand the
practical applications of that theory. But I may be
going to go too far afield with my interpretations and
I don't want to do this.
So, I for one, would appreciate a plain-English
translation of the mathematical constructs that
comprise systems theory.
What would really help me learn is to read how you (or
others on the list) would transform those mathematical
constructs into practical management methods and tools.
For example, in your note you said,
„To a systems thinker, the notion of nonlinearity is
hugely important, but has a narrow mathematical
definition: a linear model is one in which every rate
(flow) is a linear combination of the stocks in the
model (rate = a*stock1 + b* stock2 + ...); a nonlinear
model is anything else.‰
Would you please describe for us a practical
application of those two definitions (i.e., a linear
vs. nonlinear models) for managing change in a K-12
sociotechnical system (i.e., a school district)?
Thanks,
Frank
--------------------
Date: Tue, 29 Jan 2002
From: "Philip S. Abode" <pxabode@fresno.k12.ca.us>
To: k-12sd <k-12sd@SYSDYN.MIT.EDU>, Patzito1@aol.com
Subject: Re: Personal systems
Patzito1@aol.com wrote:
In order to change the educational system, I believe we have to engage
teachers in exploring and discovering their own personal systems.
My question is why educational change depends solely on the teacher? Based on
available data, teachers do not perceive themselves as the nuclear force in
education, no more than factory workers consider themselves the core
of business.
This is to say that teachers are really not the strategic nerve center of
educational organizations, they are workers on the instructional line
supervised
by principals who report to a hierarchy of middle-level administrators who are
supposed to work under direction of the district's core management
team (cabinet,
if you like) headed by a chief executive, the superintendent. Why is this model
not productive? Anyone?
Philip Abode
-------------------------
Date: Wed, 30 Jan 2002
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>
To: k-12sd <k-12sd@sysdyn.mit.edu>
Subject: Re: Managing nonlinear change....
George Richardson schrieb:
linear model is one in
which every rate (flow) is a linear combination of the stocks in the
model (rate = a*stock1 + b* stock2 + ...); a nonlinear model is
anything else.
This discussion of linearity was of tremendous use to me - thank you
George. I'm used to defining these terms mathematically, but since I'm
currently trying to construct a first-year university maths course
around Stella, this formulation is crucial. I would love to see more
definitions of terms like this on the list, which make (or cut!) the
connection to non-rigorous uses of the terms with which we are often
confronted.
Niall Palfreyman.
-----------------------
Date: Wed, 30 Jan
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>
To: k-12sd <k-12sd@sysdyn.mit.edu>
Subject: Re: What curricula do we need?
Lees Stuntz schrieb:
What pieces of curricula would you
1. Love to have for your classroom?
2. Think are the highest leverage curricula which we need for our current
state of knowledge and expertise?
I am currently constructing a first-year undergraduate maths course
around Stella, and I'm reaping benefits all over the place in terms of
making cross-connections between different course components. Last week,
for example, I was able to make very clear the idea of decoupling a
system of coupled differential equations into n independent DE's, just
by rubbing out links from a SFD on the board. In general I like seeing
standalone pieces of curriculum in which a single idea, which is very
abstract and difficult for students to understand, is graphically
portrayed as a series of stock and flow relationships.
An example which students (and I) find very difficult to represent
visually is the issue of wave phenomena in physics - waves (by contrast
with vibrations) act in both time and space, which makes them hard to
visualise. I've been racking my brains to find a way of illustrating the
following situation in Stella without any success. I'd be very grateful
to anyone who can offer a neat little diagram to represent it clearly:
---------------------------------
A microphone is positioned on the line between two identical
loudspeakers, at an equal distance from each one. The loudspeakers are
fed from the same source and they therefore emit exactly the same sound,
which is a continuous tone of frequency 220 Hz. What is the minimum
distance (to the nearest cm) that the microphone must be moved towards
one of the loudspeakers for the sound to reach an intensity minimum?
---------------------------------
The answer is 1/4 of a wavelength, but I'd love to really give students
a visual understanding of this fact.
Niall Palfreyman.
----------------------
Date: Mon, 4 Feb 2002
To: k-12sd@sysdyn.mit.edu
From: "Jim Hines" <jhines@MIT.EDU> (by way of Nan Lux)
Subject: ANNOUNCE MIT adds new distance course (SD3557)
We're adding Environmental Dynamics to MIT's distance course offerings.
The course is taught by Forrester Award winning author Andy Ford.
MIT's distance program now offers seven for-credit courses in system
dynamics via the web. We're currently accepting applications for the
term that starts in late February.
To apply or for more information please visit
http://caes.mit.edu/asp/off_campus/system_dynamics/. For additional
info about the content of the new course, you can write Andy at
FordA@mail.wsu.edu. For questions about other courses in the program
feel free to contact me.
Regards,
Jim Hines
MIT
jhines@mit.edu
-----------------------------
Date: Fri, 1 Feb 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: "Jay W. Forrester" <jforestr@MIT.EDU>
Subject: System dynamics and mathematics
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>
I'm
currently trying to construct a first-year university maths course
around Stella, this formulation is crucial. I would love to see more
definitions of terms like this on the list, which make (or cut!) the
connection to non-rigorous uses of the terms with which we are often
confronted.
There is now a text book on teaching mathematics using system dynamics:
Fisher, Diana. M. (2001). Lessons in High School Mathematics: A Dynamic Approach. Hanover, NH, High Performance Systems, Inc.
--
---------------------------------------------------------
Jay W. Forrester
Professor of Management
Sloan School
Massachusetts Institute of Technology
Room E60-389
Cambridge, MA 02139
tel: 617-253-1571
fax: 617-258-9405
Home office:
tel: 978-369-9372
fax: 978-369-9077
----------------------------
From: "John Gunkler" <jgunkler@sprintmail.com>
To: "'k-12sd'" <k-12sd@sysdyn.mit.edu>
Subject: RE: Personal systems
Date: Fri, 1 Feb 2002
Philip Abode asks why teachers are so important to educational change.
In particular he says, "Based on available data, teachers do not
perceive themselves as the nuclear force in education, no more than
factory workers consider themselves the core of business."
I'm sure there will be some who take exception to this comparison, but I
draw a partial answer to his question from it. In my 15+ years of
helping organizations change, I have certainly learned at least this:
No change will occur if the people on the front line (teachers, factory
workers, customer service employees, etc.) oppose it. Teachers have the
ultimate power to block change -- and they've often shown themselves
willing to do so! The history of educational reform is rife with
examples of (pedagogically) well-designed changes that, in their
implementation, don't work because of the actions of teachers. If you
read John Dewey (not his interpreters, but what he personally wrote),
you'll hardly recognize the reform that was supposedly based on his
theories. What teachers (and others -- I'm not picking on teachers) did
was implement what they liked, what was easiest to do, what caused the
fewest ripples ... and simply discarded the rest. Well, in my opinion,
his genius was mostly in the discarded "rest." And this is just one
example. Look at any other extensive attempt to reform education (even
the long-overdue mathematics reforms being tried now) and you'll see a
similar pattern -- partial implementation, primarily controlled by
teachers and local school administrators, no matter where the reform
originated.
So, I must agree, from my experience, that teachers are perhaps the most
important "cog" to deal with when trying to turn the wheel of education
reform. Are teachers supposed to be the only source of reform? Are
teachers supposed to be the only ones who must sacrifice in order to
make reform happen? Are teachers the only ones who are to be held
responsible for reform? No, no, and no -- of course not. But do
teachers, ultimately, control whether reforms will be implemented as
intended? Yes!
That's why in my work on organizational change I always use a
simultaneous "top-down, bottom-up" approach. Top-down because change is
more difficult without support (time, money, encouragement, leadership,
etc.) from those with control of the purse strings. Bottom-up because,
ultimately, those whose behavior is most changed must engage in the
change process for their own compelling reasons. And if I have to do
without one or the other, I'll do without the "top," because revolutions
can be made by those on the ground in spite of opposition from those
above -- but I have yet to see the opposite be true.
John W. Gunkler
jgunkler@sprintmail.com
-----------------------
Date: Sun, 3 Feb 2002
From: "A.S.Munhoz" <munhoz_br@yahoo.com.br>
Subject: skill model
To: k-12sd@sysdyn.mit.edu
Hi!
I'm working in a simple model to get better
description for the truth: if you want to learn, you
have to work.
I think it contributes to some feeling about dynamics
of human skills development.
As we will see, although there's a common belief
that human skill increases in a exponential way with
the quantity of solved problems and tried
problems, the intensity of that may be so low that it
looks like to have a linear behavior. But
if we vary some factor that behavior can change to a
stronger exponential behavior.
The factor that we will concern is the efect of
quantity of tried problems as a step function in the
beging of the learning process.
The model that we develop here was simulated with
Powersim Constructor, Version 2.51.
You can find that in
http://br.briefcase.yahoo.com/bc/munhoz_br/lst?.dir=/System+Dynamics&.view=l
So, the model works with 3 Stocks:
Quantitity_of_Tried_Problems,
Solved_Problems(Quantitity), Capacity.
And with 3 fluxes:
Trasference_Rate, Tried_Problems_Increasing,
Capacity_Increasing.
The Trasference_Rate gives the tax of transference
between the first two Stocks. We
suppose the proportionality
Trasference_Rate=
0.01*Capacity*Quantitity_of_Tried_Problems
which has, I believe, an acceptable form, since the
the success is impossible without fails.
Also we supposed a proportional dependence for the
rate of Tried Problems, since
as we are working with problems, more problems rise
and as we are solving problems,
we obtain more confidence to attack more problems. So
Tried_Problems_Increasing=0.01*Quantitity_of_Tried_Problems*Solved_Problems
.
For the Capacity_Increasing, the common sense is
described with a proportional
dependence with the quantity of Solved Problems, and
we write
Capacity_Increasing=1*Solved_Problems .
The Skill1.sim, Skill10.sim, Skill100.sim, found in
the above
link, differ only in the initial value for the
Quantitity_of_Tried_Problems .
With, respectively, 1, 10, 90 Tried Problems in the
initial time .
The effect of this change is strongly.
For the case 1, the
increasing is slow and in the time 100 months,
we get a approximated value of 200 for Capacity. For
the case 10, the increasing is stronger
but no much and we obtain 1000 for the Capacity after
100 months.
But when we tried a initial value of 90 for that
Quantity, the effect is much more stronger
and the capacity is greater than 5000 after 20 months
!
This reveals that a concentrated work in the begining
of learning process can imply
consequences of much great magnitude rapidly with
intense exponential behavior.
What do you think?
Antonio Sergio Munhoz
M.Sc. Applied Math
--------------------------
Date: Tue, 5 Feb 2002
To: k-12sd@sysdyn.mit.edu
From: Nan Lux <nlux@MIT.EDU>
Subject: Road Maps with Vensim now...
Dear List Members,
Our small group, the System Dynamics in Education Project, has just
completed a new version of the Road Maps series (available to
download from http://sysdyn.mit.edu). We have kept all the original
exercises for Road Maps in STELLA and added pages that describe how
to use Vensim software for each computer exercise.
Please check it out and let us know how we did!
Best regards,
Nan Lux
Ms. Nan S. Lux, Program Manager
MIT System Dynamics Group
E60-375, 30 Memorial Drive, Cambridge, MA 02139
Phone: (617) 253-1574 Fax: (617) 258-9405 Email: nlux@mit.edu
Web sites: http://sysdyn.mit.edu or http://web.mit.edu/sdg/www
-----------------------
Date: Mon, 04 Feb 2002
From: Steve Kipp <skipp@glynn.k12.ga.us>
To: k-12sd <k-12sd@sysdyn.mit.edu>
Subject: Re: Managing nonlinear change....
I am joining this excellent discussion late, but have any of you worked much
with "dialogue" (in the technical, Martin Buber/ Bill Isaacs/Daniel Yankelovich
sense) in your work with leading systemic change? I have found
dialogue to be an
extremely powerful tool in my initial attempts to use it to tap into the
collective wisdom of practitioners in the trenches of k-12 public education.
User-friendly and succinct references include Yankelovich's book "The Magic of
Dialogue", a summary of which may be found at
www.nonprofitquarterly.org/collaboration/yankelovich.php
and Glenna Gerard and Linda Ellinor's booklet "Dialogue at Work: Skills for
Leveraging Collective Understanding", from Pegasus Communications.
Steve Kipp
--
"I believe that really profound change can't be imposed; it has to be nurtured.
We must unleash the forces of innovation and the passion of
individuals, and top
down solutions won't do that."
-Peter Senge
----------------------
Date: Mon, 4 Feb 2002
From: "A.S.Munhoz" <munhoz_br@yahoo.com.br>
Subject: skill model
To: k-12sd@sysdyn.mit.edu
Hi!
Well, I saw a problem with the Skill Model.
That structure implies increasing of level Capacity
although the Solved_Problem level
maintain inalterable.
Then, I think that a possible solution is substitute
that model with a model when we continue
admitting the same dependence between Capacity and
Solved_Problems, but
now there's a flux between those. This avoid
contribution of old solved problems for
Capacity Increasing. Another change is that the level
Capacity also affect the
Tried_Problems_Increasing and the equation for this
rate is substitute by
Tried_Problems_Increasing=0.01*Capacity*Quantitity_of_Tried_Problems*Solved_Problems
You can find the model in Powersim Constructor,
Version 2.51 in
http://br.briefcase.yahoo.com/bc/munhoz_br/lst?&.dir=/System+Dynamics_Review&.src=bc&.view=l&.last=1
whose names are skillr1.sim, skillr10.sim,
skillr90.sim .
Now we note a similar behavior. If we try for initial
Quantity_of_Tried_Problems
1 or 10, we observe that the capacity maintain
constant after some time. But with the value
90, we recall the strong exponential behavior.
I think that the conclusions of the first Skill Model
continues correct.
With the relevant difference that with the first two
values, the model preview a stagnation
behavior.
Antonio Sergio Munhoz
M.Sc. Applied Math
----------------------------------
Date: Thu, 7 Feb 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: "Jay W. Forrester" <jforestr@MIT.EDU>
Subject: Threats to innovation
From: "John Gunkler" <jgunkler@sprintmail.com
Date: Fri, 1 Feb 2002 11:09:43 -0600
The history of educational reform is rife with
examples of (pedagogically) well-designed changes that, in their
implementation, don't work because of the actions of teachers. If you
read John Dewey (not his interpreters, but what he personally wrote),
you'll hardly recognize the reform that was supposedly based on his
theories. What teachers (and others -- I'm not picking on teachers) did
was implement what they liked, what was easiest to do, what caused the
fewest ripples ... and simply discarded the rest. Well, in my opinion,
his genius was mostly in the discarded "rest."
Perhaps, on this list, it is time to discuss whether or not the
threat attributed above to Dewey's ideas is also likely to befall
system dynamics in K-12 education.
1. How often are teachers getting a glimpse of system dynamics and
then falling back into "systems thinking" without engaging in serious
computer simulation? By "systems thinkers" I mean the large group
of people who think about systems, talk about systems, believe that
systems are important, but have no solid understanding of the dynamic
behavior of systems. "Systems thinking" reveals less than five
percent of the deep insights that can be developed through actually
working with computer simulation.
2. Is there a tendency for teachers to get an introduction to system
dynamics, find that it is useful, and then fail to go further and
continue to build their skills?
3. Is K-12 education settling for an introductory glimpse of systems
taught separately at every grade level without working to build a
cumulative education that progresses throughout the full twelve years?
4. What would teachers like to have available in system dynamics
training to extend competence? What suggestions are there for moving
forward? If a clear need can be defined, it may be possible to
provide what is required.
--
---------------------------------------------------------
Jay W. Forrester
Professor of Management
Sloan School
Massachusetts Institute of Technology
Room E60-389
Cambridge, MA 02139
tel: 617-253-1571
fax: 617-258-9405
Home office:
tel: 978-369-9372
fax: 978-369-9077
------------------------------
From: paulnewton@attglobal.net
Date: Thu, 07 Feb 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
Subject: Managing nonlinear change....
In response to Steve Kipp's note about use of dialogue, as part of a
Door County Wisconsin initiative a few years ago I wrote a brief 2 page
note about the relationship between system dynamics and dialogue.
It is downloadable from:
http://www.stewardshipmodeling.com/paul_newton.htm
The paper is entitled, "Notes on the Relationship between Dialogue and
System Dynamics" and is the last paper in the "Working Papers" group at
the bottom of the page.
Paul Newton
-------------------------
Date: Mon, 11 Feb 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: Ed Gallaher <gallaher@ohsu.edu>
Subject: Threats to innovation
Jay W. Forrester wrote:
Perhaps, on this list, it is time to discuss whether or not the
threat attributed above to Dewey's ideas is also likely to befall
system dynamics in K-12 education.
1. How often are teachers getting a glimpse of system dynamics and
then falling back into "systems thinking" without engaging in serious
computer simulation? By "systems thinkers" I mean the large group
of people who think about systems, talk about systems, believe that
systems are important, but have no solid understanding of the dynamic
behavior of systems. "Systems thinking" reveals less than five
percent of the deep insights that can be developed through actually
working with computer simulation.
2. Is there a tendency for teachers to get an introduction to system
dynamics, find that it is useful, and then fail to go further and
continue to build their skills?
3. Is K-12 education settling for an introductory glimpse of systems
taught separately at every grade level without working to build a
cumulative education that progresses throughout the full twelve years?
4. What would teachers like to have available in system dynamics
training to extend competence? What suggestions are there for moving
forward? If a clear need can be defined, it may be possible to
provide what is required.
I have just developed a two-compartment dilution problem based on a
differential equaitions solution in a text on Mathematical biology.
For those of you familiar with the "Rain Barrel" models, this model
is only slightly more complicated.
- Compartment A has an input (pulses or steps) and a first-order
(exponential) output.
- Compartment B is identical.
- A and B are connected by a membrane that allows diffusion of drugs
between them.
Example 1:
- Drug is introduced into A at a constant rate.
- As Conc_A rises, drug diffuses into B, while at the same time it is
being removed from A.
- Conc_B will always lag behind Conc_A.
When the input ceases, Conc_A begins to decline, but it is
replenished by drug diffusing back from Conc_B.
There are literally -hundreds- of interesting variations with this model.
- If diffusion is very small, the model simplifies to a single compartment (A).
- If diffusion is very large, the model simplifies to a single
compartment (A+B).
- If there is no output from B, Conc_B will always rise to equal
Conc_A, following a lag.
- If B also has an independent output, the level of B will stabilize
below the level of A. How much lower? Over what time course?
- What happens if the volume of B = 2 x Volume A? What about 4x? 10x?
- What happens if input A is a series of pulses instead of a constant input?
- What happens if both A and B have independent inputs?
Which of these scenarios have relevance in the real world? Drug
regimens? Duration of action of drugs? Accumulation of lipid-soluble
drugs in fat, followed by slow release after drug administration
ceases?
Long-term release of marijuana metabolites makes drug testing
possible long after the initial exposure.
I completely agree with Jay's comments above. The model I describe
has only two stocks, and yet opens up many hours of modeling and
simulation, with significant insights to the real world.
The insights, however, come from SIMULATION, not merely via looking
at the structure, or discussing the possible (?!) dynamic behaviors.
George's keynote address at Skamania (July 2001) was an elegant
example of "Insights from Simple Models".
I believe there should be more emphasis on -really- studying the
behaviors of four or five basic 1- and 2-stock models.
Rain Barrel (first-order delay)
Oral drug model (second-order delay)
S-shaped population growth
Basic epidemic model
Two (or three) interacting S-shaped growth models
These models could be used in support of hundreds of curricular
topics, across all disciplines. Taken together they would provide a
very solid foundation in System Dynamics for both teachers and their
students, from grades 5-12.
Ed Gallaher
--
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
--------------------------
From: GBHirsch@aol.com
Date: Wed, 13 Feb 2002
Subject: Re: Threats to innovation
To: k-12sd@sysdyn.mit.edu
I think that Jay Forrester is right in implying that many teachers are
getting a glimpse of System Dynamics, but may be pursuing it in a superficial
manner rather than developing skills in modeling and simulation that would
lead to real understanding. Perhaps they aren't motivated to develop those
skills because they don't really understand how SD can help them be better
teachers. Furthermore, the majority of teachers aren't even getting a
glimpse. SD is totally unknown to them. The question to me is how can more
teachers become aware of SD and then be motivated to pursue it seriously.
Materials that help teachers develop the simulation modeling skills will
certainly help. However, an important part of the answer lies in the
teachers' ability to see SD's relevance to what they already have to teach.
I think that most teachers already have enough demands placed on them and are
not interested in learning a new set of skills unless they can see how it
will help them do better at what they already have to do.
The link to establish this relevance is in curricula that apply SD to
particular topics they are already teaching. These might be topics that are
not being taught well because they are inherently about dynamic phenomena,
but are being taught with static methods. An example might be a lesson about
a period in a country's history when fundamental changes are taking place,
but the teaching is in terms of facts rather than the underlying causal
structure, its behavioral consequences, and alternative outcomes if different
policies had been followed. Or a science topic where the emphasis is on
calculating the right answer with a formula rather than understanding the
underlying structure and dynamics and how behavior changes under different
conditions.
There don't need to be pieces of SD curricula for every topic, but enough so
that teachers can see the relevance. Once teachers are "hooked" and become
interested in SD, Jay is right that we need curricula, training
opportunities, and other pathways for teachers to become proficient in all
aspects of SD including simulation modeling.
The CLE is an excellent resource for teachers who become interested in SD and
want examples of how to apply it. It has many examples in different subject
areas. But what about the average teacher who doesn't know anything about
SD? Are they finding their way to these examples? Are the materials
packaged in a way that teachers who are not familiar with SD will be
comfortable picking them up and using them with their students? We need to
be asking these questions of the "average" teachers rather than just the more
motivated ones who populate this listserv.
Gary Hirsch
GBHirsch@aol.com
-----------------------=
From: Barry Richmond <brichmond@hps-inc.com>
To: k-12sd@sysdyn.mit.edu
Subject: Reactions to Jay & Ed's Comments
Date: Mon, 18 Feb 2002
I recently caught wind of a discussion here regarding the importance of computer simulation in the learning process, and more generally dealing with the worth of "systems thinking" versus "system dynamics." Not being a regular contributor to, or listener of, this discussion Forum, I don't know how many people may have opined on these topics. I will focus my reactions on comments made by Jay Forrester, and supportive follow-up comments contributed by Ed Gallaher. I apologize for slighting anyone else who may have made contributions to the discussion, and I hope my response is not covering ground already adequately covered by others.
Both Jay's and Ed's comments were troublesome to me because although I know both mean well, I feel comments like the ones they've made are contrary to the advancement of "systems" into K-12 education. Their comments serve to divide the community of K-12 practitioners, creating an "us" (who rely heavily on computer simulation) and a "them" (who don't). In fact, there is only a "we," people who are trying to effect a change for the better in the thinking skills students carry forward into the world.
But my reaction to their position is not simply a "Hey, let's work together on this" message. I'd like to go on record as saying that they are factually wrong in saying, in effect, that "important learning" can come only from doing computer simulation. Jay says, "Systems thinking reveals less than five percent of the deep insights that can be developed through actually working with computer simulation." I will not embarrass Jay by asking how much evidence he has to support the 5% number, but I will take him to task for the essence of the assertion. Ed reinforces Jay's point of view, saying: "I completely agree with Jay's comments above…The insights…come from SIMULATION, not merely via looking at the structure, or discussing the possible (?!) dynamic behaviors." Ed, of course, meant "computer simulation," as opposed to the mental simulation that everyone does.
I agree that it's possible to learn new things from doing computer simulation. In my experience (both direct and in observing others), the amount of learning that results has ranged from a little to a lot. Sometimes one merely confirms what one has already intuited from constructing the map, discussing the issue, or staring at a behavior over time graph. Other times one sees something they didn't see prior to computer simulation, but it doesn't completely "blow socks off." And sometimes, one is truly surprised (and thrilled!) by an amazing insight, with truly counterintuitive behaviors arising out of the computer simulations. I will not hazard a guess at the distribution among these outcomes. However, the "amazing insights" outcome is certainly not anywhere close to a 100% of the time phenomenon.
Perhaps more importantly, though, both Jay and Ed fail to recognize and acknowledge a place in the learning cycle where hugely rich learning outcomes occur. This "place" comes prior to computer simulation, though computer simulation can later add to the learning gleaned in this phase. The "place" (actually an activity) I am talking about is construction of the stock/flow map. If students are equipped with the appropriate system-as cause, dynamic, and 10,000 meter, thinking skills, they will do a good job of filtering the reality they are seeking to model (i.e., setting the extensive and intensive model boundaries). They will make appropriate decisions about what to include within the model boundary, and at what level of aggregation to include it. Then, if they are equipped with the appropriate operational, closed-loop, and non-linear, thinking skills, they will do a good job of representing (using stocks, flows and wires) what they have decided to include.
The process of thinking hard about what to include, at what level of aggregation, and how to represent what's been included, generates a vast amount of learning-all of which precedes computer simulation! Again, computer simulation can serve as an excellent "sanity check" on this thinking, but before it does, mental simulation should always precede it! Mental simulation offers a second major learning opportunity in this phase of the learning cycle. If students only computer-simulate models that others have constructed, they will not develop the critical thinking skills needed to construct their own mental models. I would argue it is precisely these critical thinking skills-i.e., the ability to think for oneself-that are the most important skills we can help students to develop! And, because the map-construction process is best executed as a team activity, it affords an opportunity for developing another set of skills that are vital to students entering our ever more interdependent world: empathy and appreciation for diversity. When many voices are heard and viewpoints shared in the process of rendering a collective mental model as a stock/flow map, not only is it likely to result in a better collective mental model, but students will move closer to becoming "systems citizens" (the true purpose of "systems" activities!).
It is possible that Professors Forrester and Gallaher agree with what I am saying, but are only arguing that computer simulation of simple models ought to "come first"-i.e., as a vehicle for developing the thinking skills needed for constructing stock/flow maps. I agree that computer simulation can play a role in building the requisite thinking skills. However, this would not lead me to make statements like the ones the Jay and Ed have made. I would urge them to stop making such statements, and to begin appreciating and acknowledging the full spectrum of "systems" activities in which K-12 teachers are engaging. I only wish Jay and Ed could have been present at the workshop that Tim Joy and I co-facilitated in Norwalk California last December. Tim held the group spellbound while he illustrated how he used a behavior over time graph to facilitate a rich discussion of Lord of the Flies. I would defy anyone to characterize such a discussion as "superficial," or in any way "learning impoverished," because he did not invoke computer simulation! Let's get on with the work and forget the "us" and "them."
--
High Performance Systems, Inc
45 Lyme Road, Suite 300, Hanover, NH 03755-1221
Tel. 603-643-9636 - Fax. 603-643-9502
----------------------------------
From: SCGteach@aol.com
Date: Fri, 15 Feb 2002
Subject: Re: Hirsch: Threats to innovation
To: <k-12sd@sysdyn.mit.edu>
I think the problem is not that teachers are unmotivated to use
system dynamics in their classrooms, but they do not know how to use
SD effectively. The other factor is the technology that each teacher
has access to in their own room. I am currently in a Graduate class
and learning about SD. I teach First grade in a Catholic school and
have not yet obtained any information on how I can incorporate SD
into my classroom. I am not unmotivated, just not sure on how to use
it effectively. Any suggestions would be great!
Suzanne C.
--------------------------
From: "Joe Rimback" <rimback@erols.com>
To: "'k-12sd'" <k-12sd@sysdyn.mit.edu>
Subject: RE: Reactions to Jay & Ed's Comments
Date: Thu, 21 Feb 2002
In response to Barry Richmond's "Reactions to Jay & Ed's Comments", I would like to suggest that Jim Lyneis' 1999 paper on "System dynamics for business strategy - a phased approach" (System Dynamics Review Vol. 15, No. 1) offers some additional insight and relevant experience. One possibility seems to be that the "place" (reply author's quotes) where learning occurs most extensively (or rapidly or sufficiently) depends on the complexity of the problem and the ultimate purpose of the research. Quoting from the last paragraph on page 42:
"While Phase 1 and the systems thinking that is a key part of it (Phase 1) are a necessary start, it should not be the end point. Two problems limit Phase 1's effectiveness in supporting business strategy. First, simple causal diagrams represented by system archetypes, while useful pedagogically, take a very narrow view of the situation (typically, one or two feedback loops). In reality, more factors are likely to affect performance, and it is therefore dangerous to draw policy conclusions from such a limited view of the system. A more complete representation of the problem considers more feedback effects and distinguishes levels from rates, but introduces the second problem: research has shown that the human mind is incapable of drawing the correct dynamic insights from mental simulations on a system with more than two or three feedback loops (Sterman 1989, Paich and Sterman 1993). In fact, without the rigor and check of a formal simulation model, a complex causal diagram might be used to argue any number of different conclusions. In addition to overcoming these limitations, as discussed below, formal modeling adds significant value to the development and implementation of effective business strategies."
And about small insight models (Phase 2) on page 45: "These models can help managers improve their intuition (mental models), and thereby make better decisions, but are generally insufficient for critical strategic decisions."
And on page 50: "Calibration (Phase 3) often reveals incorrect or incomplete mental models. The initial definition of a problem and the formulation of the structure of a simulation model are based on managers "mental models". These models are rarely complete or one-hundred percent correct. Homer notes that "Hard data can materially affect the final structure and key parameter values of a model and, consequently, its predictions and even its policy results" (Homer 1997, p.297). Winch discusses how a large, quantitative model is used to correct two common types of problems encountered in consensus-building in management teams: (1) different views (structure or resultant behavior) held by different managers; and (2) the collective view is incomplete or unrefined (Winch 1993, pp.289-290).
And, finally, on page 55: "Given the significant increase in cost as one moves through the phases, and particularly between Phases 2 (insight model) and 3 (calibrated model), the question of value for money often arises. While I believe that the client obtains value, regardless of when they stop (i.e. when simulation development stops), strategy consulting is one case where the "80/20 rule" does not apply - the client does not get 80% of the value for 20% of the cost (which would be essentially at the end of Phase 1). . . . In this situation, the "value" is back-end loaded"
Having said (quoted) all that, I have no problem believing that a lot of learning still occurs within system thinking and initial (team) model-building activities. But, if value can be at least loosely equated to learning, Lyneis makes a strong case for essential learning late in the "systems thinking-model building-simulation" process. The extent to which his conclusion applies to education needs to be examined. Could everyone be right at some point on some spectrum?
Perhaps this issue should be modeled - and maybe even simulated.
Joe Rimback
Enterprise Design, Inc.
<mailto:design@starpower.net>design@starpower.net
(240) 994-4180
--------------------------
Date: Tue, 19 Feb 2002
Subject: Getting Started in First Grade
From: Lees Stuntz <stuntzln@clexchange.org>
To: k-12sd <k-12sd@sysdyn.mit.edu>
Here is a list of materials all available on the Creative Learning Exchange
website (clexchange.org) which may be helpful to you. I specifically
extracted this list to give to elementary school teachers in my own town.
The In and Out Game and the Friendship game are good places to start with
your class.
I am available for questions or comments, either through e-mail
(stuntzln@clexchange.org) or phone (978-287-0070.)
The following are papers which give you the broader picture of what systems
thinking and system dynamics are and why you would use them in K-12
education.
SE1995-08STIn25WordsOrLess.pdf
SE1994-11ConsiderGypsyMoth.pdf
SE1994-07SDAsPrepFor21stCen.pdf
SE1994-03SEForK-12InTheUS.pdf
SE1993-01AProposedSequence.pdf
SE1993-05STCriticalThinking.pdf
SE1993-05STFourKeyQuestions.pdf
SE1993-01LearnerDirectedSE.pdf
Interesting stories about teachers using SD/ST in their classrooms:
SE1999-04Pioneer1-Surprise.pdf
SE1999-04Pioneer2-LazyTeach.pdf
SE1999-09Pioneer3-ToSucceed.pdf
SE1999-09Pioneer4-1Students.pdf
SE2000-10Pioneer5-WhyRush.pdf
SE2000-10Pioneer6-Expert.pdf
SE2000-10Pioneer7-Naturally.pdf
Getting started with SD/ST tools:
CC1998-10GettingStartedBOTG.pdf
EN1993-05BOTGsWritingTools.pdf
EN1998-03BOTGActivities.pdf
CC2001-11EverydayBOTGs.pdf
SD1996-03GetStarted5Lessons.pdf
SE1999-09In&OutGame.pdf
Curricula for elementary age:
CC1999-04MammothExtinction.pdf
SS1996-11FriendshipGame.pdf
CC2000-10GraphFriendshipGam.pdf
EN2000-10TuckEverlasting.pdf
SC1999-09It'sCool.pdf
Rubrics for system dynamics tools:
SE2001-03RubricsForSDTools.pdf
Lees N. Stuntz
Creative Learning Exchange Phone- 978-287-0070
1 Keefe Road Fax- 978-287-0080
Acton, MA 01720 e-mail- stuntzln@clexchange.org
http://clexchange.org
------------------------------
Date: Tue, 19 Feb 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: Bob Gorman <bgorman@kncell.org>
Subject: Re: Reactions to Jay & Ed's Comments
At 12:54 PM 2/19/2002, Barry Richmond wrote:
From: Barry Richmond <brichmond@hps-inc.com>
To: k-12sd@sysdyn.mit.edu
Subject: Reactions to Jay & Ed's Comments
Date: Mon, 18 Feb 2002 08:44:46 -0500
I recently caught wind of a discussion here regarding the importance of computer simulation in the learning process, and more generally dealing with the worth of "systems thinking" versus "system dynamics."
Good to hear from you Barry.
We met at DEC in Stow, MA quite some years ago. :-)
Both Jay's and Ed's comments were troublesome to me because although I know both mean well, I feel comments like the ones they've made are contrary to the advancement of "systems" into K-12 education. Their comments serve to divide the community of K-12 practitioners, creating an "us" (who rely heavily on computer simulation) and a "them" (who don't). In fact, there is only a "we," people who are trying to effect a change for the better in the thinking skills students carry forward into the world.
I agree strongly, and have worked with challenged youth, both emotionally and learning challenged. One thing they often miss is the interconnectedness of people, theirs lives, and the results of their actions...
I have been mostly a reader rather than contributor in this forum, feeling at times like a "them".
The process of thinking hard about what to include, at what level of aggregation, and how to represent what's been included, generates a vast amount of learning-all of which precedes computer simulation! Again, computer simulation can serve as an excellent "sanity check" on this thinking, but before it does, mental simulation should always precede it!
I agree, introducing the computer too early fosters dependency and reduced self-reliance.
Mental simulation offers a second major learning opportunity in this phase of the learning cycle. If students only computer-simulate models that others have constructed, they will not develop the critical thinking skills needed to construct their own mental models. I would argue it is precisely these critical thinking skills-i.e., the ability to think for oneself-that are the most important skills we can help students to develop!
Absolutely, this is perhaps the most critical skill needed in today's growing world.
We are at a time when we need to shout from the roof tops that Individuals count.
And, because the map-construction process is best executed as a team activity, it affords an opportunity for developing another set of skills that are vital to students entering our ever more interdependent world: empathy and appreciation for diversity.
A truly neat side-effect.
Bob
Knowledge
is NOT enough!Knowledge
+ Confidence enables Action.Vision
+ Action = Leadership!-
Bob Gormanhttp://www.kncell.org
-----------------
From: "Jane Schumacher" <idea@idea.org>
To: "k-12sd" <k-12sd@sysdyn.mit.edu>
Subject: Re: Reactions to Jay & Ed's Comments
Date: Tue, 19 Feb 2002
Dear Barry:
I so appreciate your comments. I too have been disturbed by the perceived need for technology in the uses of systems thinking in classrooms. Our work in educating any learner must first and foremost be in engaging the minds and the hearts of the learners. Helping learners make connections to systems that encompass the usual "bits and bytes" of learning IS the work that needs to preceed an introduction of computer simulations. Learning as defined by a constructivist perspective is about working with learners as they build their own meaning and make sense of their world. Helping learners understand and use a systems thinking approach occurs first in the mind and thought process and later in applications.
Thank you again for your thoughts! I find this discussion group to be stimulating and informative.
Jane Schumacher, Ed.D.
|I|D|E|A|
259 Regency Ridge
Dayton, OH 45459
937.434.6969 (phone)
937.434.5203 (fax)
<mailto:idea@idea.org>idea@idea.org
----------------------
Date: Thu, 21 Feb 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: Janis Dutton <jldutton@iac.net>
Subject: Re: Reactions to Jay & Ed's Comments
Barry Richmond wrote:
Both Jay's and Ed's comments were troublesome to me because although I know both mean well, I feel comments like the ones they've made are contrary to the advancement of "systems" into K-12 education. Their comments serve to divide the community of K-12 practitioners, creating an "us" (who rely heavily on computer simulation) and a "them" (who don't). In fact, there is only a "we," people who are trying to effect a change for the better in the thinking skills students carry forward into the world.
(snip)
Perhaps more importantly, though, both Jay and Ed fail to recognize and acknowledge a place in the learning cycle where hugely rich learning outcomes occur. This "place" comes prior to computer simulation, though computer simulation can later add to the learning gleaned in this phase. The "place" (actually an activity) I am talking about is construction of the stock/flow map.
(snip)
I only wish Jay and Ed could have been present at the workshop that Tim Joy and I co-facilitated in Norwalk California last December. Tim held the group spellbound while he illustrated how he used a behavior over time graph to facilitate a rich discussion of Lord of the Flies. I would defy anyone to characterize such a discussion as "superficial," or in any way "learning impoverished," because he did not invoke computer simulation! Let's get on with the work and forget the "us" and "them."
These are interesting points. I have a question--which came first thinking systemically or systems thinking?
Where did indigenous people plug in their laptops? What symbols did they use in their stocks and flows diagrams? What units did they use in their behavior over time graphs?
All of these are fabulous tools for engaging people in new ways thinking and understanding complexity. Still they are just tools. I respectfully submit that too often the conversation centers more on the tools than their purpose, which creates, I believe, the barriers to implementing the tools in K-12 practices whether at the beginning levels or their extension into more advanced computer modeling. Furthermore, it reinforces the us/them mentality. In other words, in this or any educational innovation or reform, the emphasis is way "too keen on [What] and How and not so hot on Why." (My apologies to Andrew Lloyd Webber and Tim Rice, but I think it is important to note that in their song Jesus is railing against God's silence and not that of His fellow teachers.)
The question is simple to remember: To what end? The answer is not at all simple to find. I've always thought the STDM tools are tools for advancing the conversation and thinking around desired ends. They are the means to an end not the end itself. Too often I think that gets lost in the enthusiasm.
Just as guilty,
Janis Dutton
--------------------------
Date: Thu, 21 Feb 2002
To: k-12sd <k-12sd@sysdyn.mit.edu>
From: Ed Gallaher <gallaher@ohsu.edu>
Subject: Re: Reactions to Jay & Ed's Comments
I want to respond to Barry Richmond's incisive comments re Jay's recent message, and my follow-up, strongly promoting the necessity for computer simulation. This ongoing topic has created some light, and a great deal of heat over the past several years. We should continue to examine it closely.
The systems thinking Barry describes is rigorous and insightful. In preparing this response, it became clear to me; I am not at all opposed to systems thinking, I am opposed to superficial, erroneous systems thinking.
Barry said:
Perhaps more importantly, though, both Jay and Ed fail to recognize and acknowledge a place in the learning cycle where hugely rich learning outcomes occur. This "place" comes prior to computer simulation, though computer simulation can later add to the learning gleaned in this phase. The "place" (actually an activity) I am talking about is construction of the stock/flow map. If students are equipped with the appropriate system-as cause, dynamic, and 10,000 meter, thinking skills, they will do a good job of filtering the reality they are seeking to model (i.e., setting the extensive and intensive model boundaries). They will make appropriate decisions about what to include within the model boundary, and at what level of aggregation to include it. Then, if they are equipped with the appropriate operational, closed-loop, and non-linear, thinking skills, they will do a good job of representing (using stocks, flows and wires) what they have decided to include.
The process of thinking hard about what to include, at what level of aggregation, and how to represent what's been included, generates a vast amount of learning-all of which precedes computer simulation! Again, computer simulation can serve as an excellent "sanity check" on this thinking, but before it does, mental simulation should always precede it! Mental simulation offers a second major learning opportunity in this phase of the learning cycle. If students only computer-simulate models that others have constructed, they will not develop the critical thinking skills needed to construct their own mental models. I would argue it is precisely these critical thinking skills-i.e., the ability to think for oneself-that are the most important skills we can help students to develop! And, because the map-construction process is best executed as a team activity, it affords an opportunity for developing another set of skills that are vital to students entering our ever more interdependent world: empathy and appreciation for diversity. When many voices are heard and viewpoints shared in the process of rendering a collective mental model as a stock/flow map, not only is it likely to result in a better collective mental model, but students will move closer to becoming "systems citizens" (the true purpose of "systems" activities!).
Absolutely; I wouldn't change a word. In fact, even while advocating simulations, I ALWAYS recommend sketching the "expected" result BEFORE pushing the RUN button. Otherwise it is far too easy to say, "Oh yes, that makes sense . . . ", without really being intellectually engaged in the specific point of the simulation.
It is possible that Professors Forrester and Gallaher agree with what I am saying, but are only arguing that computer simulation of simple models ought to "come first"-i.e., as a vehicle for developing the thinking skills needed for constructing stock/flow maps. I agree that computer simulation can play a role in building the requisite thinking skills. However, this would not lead me to make statements like the ones the Jay and Ed have made. I would urge them to stop making such statements, and to begin appreciating and acknowledging the full spectrum of "systems" activities in which K-12 teachers are engaging. I only wish Jay and Ed could have been present at the workshop that Tim Joy and I co-facilitated in Norwalk California last December. Tim held the group spellbound while he illustrated how he used a behavior over time graph to facilitate a rich discussion of Lord of the Flies. I would defy anyone to characterize such a discussion as "superficial," or in any way "learning impoverished," because he did not invoke computer simulation! Let's get on with the work and forget the "us" and "them."
I would not advocate skipping ANY of the steps described above. An over-emphasis on "simulation" as OPPOSED to the process described above would be equally misguided. In fact, it would be much worse.
Tim has described the December workshop to me, and I wish I could have been there. It sounded wonderful, and despite my opinions on simulation, I would not suggest changing a thing.
Even though this was not a computer simulation workshop per se, I suggest it was successful and insightful PRECISELY BECAUSE Tim builds models and runs simulations!
* * * * * * * * * *
A basic foundation of System Dynamics includes the notion that complex dynamic behaviors, especially those containing feedback loops, CAN NOT be predicted intuitively in the absence of computer simulations.
* * * * * * * * * *
This is either true, or it is not true.
To the extent that it is true, it therefore follows that these insights, by definition, CANNOT be attained without computer simulations.
Someone has to develop these insights BEFORE leading a group discussion on the topic.
This does not mean the discussion is canned, or manipulated by the moderator. It does mean that it is consistent with fundamental systems thinking and system dynamics principles.
Until the models have been built, simulations conducted, and the results analyzed and interpreted, we cannot be SURE that the "insights" we gain during the development process (so well-described by Barry) in fact agree with common knowledge, or with easily observed real-world events.
Barry says that simulation can serve as a "sanity check", but only AFTER the systems thinking and model development has taken place.
No argument here. But I do not agree that simulation CAN serve as a sanity check; at some point in the process simulation MUST serve as a sanity check. (Not during every discussion, or every class, but at -some time- during the process. Maybe not by every student, but every teacher MUST have this elementary level of expertise.)
Using Barry's own terminology, what is the alternative? "Insights" that disagree with the sanity check?
Correctly done, systems thinking is way ahead of whatever is in second place! A thoughtful guided discussion using causal loop diagrams and/or stock-and-flow diagrams is without question superior to cause-and-effect thinking, e.g. the study of history as a series of important 'events'.
Consider a single model of S-shaped population growth. This model will include births and deaths, but MUST include a density-dependent multiplier to cause a shift in loop dominance. Once this model is in hand, it is possible to link two or three similar structures together to simulate rabbits and foxes, or population, jobs, and housing.
NOT OK EXAMPLE #1:
I have seen teachers leading classroom discussions that had never run this model, and literally did not know how it worked.
NOT OK EXAMPLE #2:
At a teacher training session about two years ago participants were divided into small groups to discuss a fisheries depletion problem. No computers were involved. One individual literally did not understand that "fish population" was a stock, and "catching fish" was a flow. (I am not making this up.) I'm sorry, but it will be difficult to convince me that this level of expertise is adequate for introducing systems thinking into the classroom.
Lest this seem harsh, I want to make it clear that we want to recruit newcomers; we don't want to intimidate them; we don't want to foster an elitist attitude. We can nurture our newcomers, and encourage them in every way. However, it is NOT OK to word our philosophies in such a way that they can (mis-) interpreted to reinforce such practices.
Rather than just vent, I'd like to make a specific proposal.
(1) EVERYONE commits to the Basic Training described below, whether or not the ultimate practice with students includes computers, and
(2) Those advocating computer simulations ease up a little, and acknowledge the fact that well-trained teachers can guide their students to greater insights even without the computer.
This seems like a reasonable common ground.
* * * * * * * * * * * * * * * * * * * * * * * *
BASIC TRAINING IN SYSTEMS THINKING AND SYSTEM DYNAMICS:
A. RAIN BARREL MODEL
Consider a two-day introductory workshop focused on the Rain Barrel (first-order delay) model (or a similar introductory structure). The model is developed and simulations are conducted. Inputs consist of pulses (drug doses) or steps (constant inputs turned on/off, up/down; intravenous infusions). Single and multiple doses are given at various intervals. Absorption and elimination rates are varied in agreement with commonly-used drugs.
Discussions and exercises include applications to drug regimens, polluted reservoirs, and a cooling cup of coffee. Two models are linked in series to simulate drug entering the stomach, being absorbed, entering the whole body, and being eliminated.
Regardless of math or modeling background, or academic discipline or training, SUCCESS IS GUARANTEED!
B. S-SHAPED POPULATION GROWTH (prerequisite - Rain Barrel or similar background)
Two-day workshop. Reinforcing loops, birth rate, exponential growth, doubling time. Balancing loops, death rates, exponential decay. Population density and its influence on births and/or deaths. S-shaped population growth.
Two models are then combined to simulate interactions between rabbits and foxes.
Again, SUCCESS IS GUARANTEED!
* * * * * * * * * * * * * * * * * * * * * * * * * * *
If we ALL advocated this level of training, whether or not it is to be applied to systems thinking or computer simulations, we would have a firm common ground from which to proceed. Barry's goal of insightful systems thinking would be clearly within reach.
Teachers may then choose to engage their students in computer simulations or not, depending on the curriculum, time factors, age level, technical facilities, or comfort level.
Without this level of training, I envision a finite number of alternatives:
(i) The teacher recognizes he/she is new at this, and feels that any systems-oriented approach is better than the old way of doing things. Intermediate and advanced skills will come in time.
(ii) Due to its newness and unfamiliarity the teacher is oblivious to the benefits of simulation, or is unaware of the arguments pro and con.
(iii) The teacher believes that reading about, and talking about dynamic concepts, in the absence of training with simulations, will lead to a level of insight that is adequate for teaching purposes. This level of knowledge will not lead to significant misconceptions or errors, and the benefits will be significant.
(iv) Modeling and computer simulation is over-rated (i.e. the basic foundation of System Dynamics as described above is false).
(v) Those guys are so obnoxious and pushy, and holier than thou, I wouldn't give them the pleasure of forcing me into computer simulations.
(vi) I've never been good at math. I'm not sure I can do this, and I don't want to look foolish in front of my students. Systems thinking is good enough for me.
(vii) I'm so good at this, I don't need a computer.
(Canoeing instructors need not be competitive swimmers, but I would expect my children's instructors to know how to swim. I would also hope they would encourage my children to improve their own swimming skills when the opportunity arises.)
In closing, I will be the first to acknowledge that I am making these comments from my ivory tower. I applaud the efforts of all the innovative classroom teachers who have adopted the pioneering spirit and stuck their necks out to learn (and teach) these exciting new concepts.
Sincerely,
Ed Gallaher
-----------------------------
Date: Sat, 23 Feb 2002
Subject: Reactions to Jay and Barry's Comments
From: Diana Fisher <dmfisher@aracnet.com>
To: <k-12sd@sysdyn.mit.edu>
I guess I feel compelled to put in my two cents worth regarding systems
modeling and systems thinking without systems modeling.
As I contemplate what I write I keep in mind the GREAT contributions to this
field by the founder Jay Forrester and a great thinker and teacher, Barry
Richmond. They come from differing perspectives, both of which we are
compelled to heed.
As a mathematician, I know the seminal thinking philosophers of the past
were able to analyze weighty numerical and logical situations without much
in the way of symbolic rigor. (Or they created some symbolic support of
their own.) But most of us, unfortunately, cannot. Nor will we ever be able
to work or think in that environment. Almost all of us need structure to
help us think deeply about some issues.
I have seen the work of Jan Mons, Alan Ticotsky and Ralph Quaden at Carlile,
Will Costello, and others who have done impressive work with young students
focusing on those suggestions by Barry to use BOTGs and the structured
diagrams to talk through issues under discussion. The approach they use is
very important for all of us to emulate. We need to listen to them
carefully as they tell us what works well at each level.
I do think that those preliminary exercises will represent, however, only
half of the picture of the true power of understanding a dynamic situation.
It may be all that is needed for a particular discussion at a particular
moment. But those teachers I mentioned above all know how to create well
designed computer models, so they know when the time has come to take the
discussion one step further, or when a curious child's mind needs to go
beyond words and graphs. They have a much deeper toolbox for expanding the
discussion for a much broader range of student.
I have been teaching and building system dynamics models for 12 years at the
high school level. Never before in my 32 years of teaching have I come
across a method of learning that has impressed me *every* time I see what
students produce when they create models. I feel as if I have opened a door
that has liberated a significant number of minds to much deeper analysis.
Could these students have accomplished what I have seen other ways? I don't
know. But I know what I have seen. And I know the vehicle that has helped
to produce impressive results. It was *computer modeling* of *REAL (and
important) problems* using a *visual modeling tool*. ALL three were needed
to produce the results.
So I'm saying, if you have not modeled, you have no idea how important it is
to learn. One can learn simple models and gain important insight from them.
But the K-12 picture for systems thinking education will NEVER be complete
without systems modeling. On the other hand, those of us who focus on
modeling are missing a significant portion of helping to form a developing
mind if we do not broaden our scope. We must talk about the problems using
non-computer techniques first. We must listen to those who tell us that we,
who model, also do not have the whole picture. We, modelers, MUST listen to
Barry who knows what we need to keep in mind to gain the proper framework
for our instruction. Those who are not modelers, MUST listen to Jay, who
knows that if you do not have an understanding of what model building brings
to analysis, you do not have a complete enough picture to set the proper
framework for instruction. It is necessary for both groups to learn from
the other.
I, for one, however, am most inclined to listen to the systems thinkers who
also know how to model, because they have a perspective that is sufficiently
broad to speak authoritatively on each issue. Those people will have the
most impact on changing how I think about my instruction.
So, I guess we all need to continue to attend the K-12 SD conferences and
continue to learn from each other, because none of is fully-baked yet :-)
(Any time you can hear George Richardson speak, he is a third truly great
systems thinker who has powerful lessons to teach us. And he has graciously
donated much of his time to bring us along in our learning. I omitted him
in my comments above only because he was not part of the original
discussion.)
Diana Fisher
---------------------------
Date: Tue, 26 Feb 2002
From: Della Robertson <frobchen@earthlink.net>
To: k-12sd <k-12sd@sysdyn.mit.edu>
Subject: Re: Reactions to Jay & Ed's Comments
This is in response to Barry Richmond's "Reactions to Jay & Ed's Comments".
The school in California that Barry referred to is my high school.
The three days were challenging and emphasis was placed on the
"thinking". Believe me, I am convinced that there is power in Systems
Thinking. What we teachers need is more help in doing the thinking
and creating the maps and "Behavior over Time" graphs.
After our November (Barry it may have seemed like December) three-day
workshop, I used "Fly A Cell" with my students and found it to be
quite good. Without further models to work from, it is very difficult
for teachers to maintain the momentum. I will look at the samples Lee
Stuntz recommended to see if that will help.
Della Robertson
Norwalk High School
---------------------------
From: Quaden@aol.com
Date: Wed, 27 Feb 2002
Subject: Re: Reactions to Jay and Barry's Comments
To: k-12sd@sysdyn.mit.edu
First of all I would like to thank Jay and Barry for sparking this great
discussion: it certainly has given us a lot of food for thought.
While I agree to many of the statements made about the different system
dynamics approaches, it was not always clear to me if the various statements
were primarily concerning the teachers or the students. This point was
brought home to me by one of Ed Gallaher's statements. Ed said: "Even though
this was not a computer simulation workshop per se, I suggest it was
successful and insightful PRECISELY BECAUSE Tim builds models and runs
simulations!" Tim being a teacher, it appears that Ed's statement are meant
to be applied to teachers (and not necessarily to Tim's students).
I would like to share a different point of view, stemming from our work with
elementary and middle school teachers and students.
From our work with students and teachers we have found that it is important
to focus on the students and to use a VARIETY of system dynamics tools and
approaches. It appears that it is the combination of different approaches
that is so effective in work with the students.
The reality for our school is that many teachers are not comfortable with a
variety of tools, so we encourage teachers to use tools that appeal to them.
The classes lead by these teachers might not get a 100% benefit, but we have
anecdotal evidence that the level of discussions and insights increase
substantially.
Contrary to teachers, many students are comfortable with different tools and
our program tries to make sure that students get connected with different
teachers, so that the students get the big picture.
Our goal is to have eighth grade students to be systems citizens, who are
able to use a variety of tools, including:
- behavior over time graphs
- causal loop diagrams
- stock flow diagrams
- simple models
In our program it is quite common that teachers have very limited or
nonexistent modeling skills (not a desired result, but a current reality),
but all our eighth grade students should have experience with ALL the tools
(definitely a desired result, and one that is within reach.)
Rob Quaden
Math Teacher
Waters Foundation Systems Mentor
Carlisle Public Schools
------------------------------
From: "Scott Guthrie" <sguthrie@teleport.com>
To: <k-12sd@sysdyn.mit.edu>
Subject: Yet Another Reaction To Thoughts and Managing Linear Change and Other Thoughts (LONG)
Date: Wed, 27 Feb 2002
"I believe that really profound change can't be imposed; it has to be nurtured.
We must unleash the forces of innovation and the passion of
individuals, and top down solutions won't do that."
-Peter Senge
A long title for a long thread. Forgive me if I ramble, but I want to reply to all in the entire thread (in a somewhat LIFO-like manner) in the hopes that we can better define our points of discussion and move on.
First Thread Reactions and Thoughts (to the thoughts and statements of Jay, Ed, and Barry {and yet others, since I'm so slow in replying})
Right on! In the schools, we are trying to foster learning. As Barry says, "Perhaps more importantly, though, both Jay and Ed fail to recognize and acknowledge a place in the learning cycle where hugely rich learning outcomes occur. This "place" comes prior to computer simulation, though computer simulation can later add to the learning gleaned in this phase. The "place" (actually an activity) I am talking about is construction of the stock/flow map." This is where at least half of the learning takes place for students. Our students are just learning how to see stocks and flows (levels and rates) for what they are, and the rigor that this language introduces to them is vital for their understanding of the system that they are defining and communicating to others. Breakdowns in communication, poor definitions, and logical flaws are easily seen by others, discussed, and changed/corrected. And, as with all valuable skills, "Practice! Practice! Practice!" is essential to develop the "appropriate system-as cause, dynamic, and 10,000 meter, thinking skills … filtering the reality they are seeking to model (i.e., setting the extensive and intensive model boundaries) [Barry R. again]." At the beginning of the year, I get to play Socrates/Devil's Advocate to move the discussion along and help the students learn these new skills. By the end of the year, at its best, I get to sit back and watch while the entire class hammers out their stock-flow map of a system as they see it. This powerful learning step cannot be skipped or treated too lightly.
However…
It cannot, should not, stop there. They must also spend huge amounts of time building, testing, and exercising "simple" models so they can become familiar with simple system behaviors. If they do not have this experience, they will have trouble making appropriate boundary and aggregation decisions in their more complex mental and computer models as their skills and confidence grow. This work, building, testing, and exercising "simple" models accounts for at least, and possibly more than half of the remaining learning that takes place for the students. (notice how I stay away from any hard percentage numbers <g>).
Therefore…
The last bit of learning for the students takes place when they can build, exercise, and test "complex" models of their own. This last step is extremely valuable and serves to complete a full learning cycle, as both Jay and Ed state. That said, however, this is still the level of least learning for students. But if you want to get 100%…
And Yet…
This is my approximate value set for the steps of learning systems skills. Schools are about learning. However, if you plan on actually implementing a system inspired solution or process, then you must build, test, and exercise (and repeat these steps as necessary) a model of your system to be certain of the truth of it - especially if the model involves more than two stocks and flows with feedback. As a programmer, I learned very early that flow charts, code modules/objects, and all the other tools for designing a program are junk when push comes to code: there's always something wrong with the first 20 iterations of the program (that's what all those 'SE' and 'x.x' program revisions are for). As I see it, this lack of rigorous building, testing, and exercising of models in decision/policy making is where many practitioners of System Dynamics in the "Real World" fail and is the source of much of the "you must build, test, and exercise your model" arguments that we keep witnessing at the K-12 and International SD conferences. (Practitioners who feel maligned by this statement, please send your arguments to me along with 10% of your fiscal year 2000 earnings - pre ENRON - if you wish to be heard <g>)
And Now About Improving Schools (Managing Linear Change and Other Threads)
First, I begin by trying to find the very definition of "Schools are failing." Which? Where? How? What? My working conditions? The disrepair of the building? The lack of textbooks newer than the Nixon era? Test scores (which test)? Student interest? If we're going to tackle this problem, let's start by defining what we mean when we say "schools are failing." Without this step, we're all hiding our models (mental and otherwise) from each other, wasting time and electrons. We also need to set a boundary for our model.
My Definition, My Boundary
My definition of "schools are failing" is this: schools are failing in that they are not producing "Systems Citizens." (see my definition of a systems citizen below). My model boundary is school-student-parent, for the same reasons outlined above. Both of these may be flawed, but this keeps the level of complexity down to a manageable size. Everything and everyone else is, at best, a secondary consumer of the relationship between parents, students, and school, but that's not to say that they are not important: I just choose to make them exogenous.
What I Mean (a.k.a. Outcome Based Education - No, not that OBE!)
Let's start with the "easy" part first: what do we want to be the outcome of an education. I want every student who comes out of school to be a Renaissance Man (or Woman, as the case may be) that has enough skills and knowledge in all areas to know when, where, and how to get more skills and knowledge. Barry Richmond has another term for this type of person: a "Systems Citizen" (Barry: forgive me if I'm off base here!). A "Systems Citizen" makes decisions based on their understanding of the systems around them. If they are confronted with a system that they do not understand (or are unsure of), they have the tools and skill sets necessary to gain insight/knowledge towards understanding that system. Producing people like that with our current system of education is, well, a lucky happenstance at best. This outcome is also impossible to test with any standardized instrument.
So, How Do We Get There From Here? Let's Define "There" First
(Nope, I'm not going to tell that old joke…) First, allow me to jump right in and say that most K-5 educators have this part figured out already (except the part where they get enough time to do everything in). In classes of 25 students or less, they get to spend 180 days over a year with their students, nurturing the growth of their learning habits, skill-sets, and basic knowledge bank before they hand them off to another teacher. In some schools, these same teachers get to spend 360 days over two years with their students. This, as both my children (and my wife and I) will attest, is really neat. It allows the teacher to really get to know their students. Then, in middle and high school, we kill it: suddenly, teachers get to spend only 180 hours with anywhere from 150 to 240 students a year (assuming that the student stays with the teacher for the entire year). Can they nurture the growth of knowledge and skills in these students? No! They can only accomplish this if they're really good and/or really lucky. This current system of education is the source of the "downstream errors" that Frank Duffy mentioned earlier.
Let's try a system where these errors have less of a chance of taking place. I'll warn you right now, it may cost more than the current school system that you have in your area (easily done in Oregon, <sigh>). The solution I'll propose minimizes these downstream errors and also addresses the issue of implementing the key concepts of the Quality movement by introducing numerous short response time feedback loops into the school system (to paraphrase Richard Turnock). Here I go:
In K-5
K-5 needs the least help. It already has one of the shortest feedback loops possible (teacher-student, teacher-parent, and possibly teacher-parent-student-specialist). In my system, there is one teacher in the room with no more than 15 to 20 students in a 3-year classroom. 2nd grade is the transition year, so in that year, those 15 to 20 students get two teachers: the second teacher is going to be their teacher for grades 3, 4, and 5. Class size stays at 15 to 20 (even with two teachers in the room). This transition year (and the other transition years that follow) with two teachers allows knowledge about the students to be handed down personally, teacher to teacher.
Moving On To Middle School
In the 5th grade, the students once again get two teachers, with the same benefits as in the 2nd to 3rd grade transition year. However, when the student reaches 6th grade, things can change - or not. I'll offer two scenarios: one in which we keep the middle school model, and one in which we return to the K-8 model that I grew up in. Common to both scenarios are the foreign language and physical education teachers: they get the students for 1.5 hours each a day, with class sizes of 15 to 20, but these aren't necessarily the same groups of 15 to 20 students that the grade level teachers have. In the 8th grade, another teacher gets added to each class here (for a total of two) that will follow the students throughout high school. Key here will be allowing the two "isolated" teachers (PE, language) at least 1.5 hours a day to be with/meet with the grade level teachers to exchange notes on students and help out (or the other way around).
…Middle School
In the 6th grade, the students are "shared" among three teachers (math/science, music/art, humanities) making a class of 45 to 60 students who will spend 1.5 hours a day with each teacher. The last two hours of the day will be spent together (all of the students and all three teachers). In this way, at least one teacher will have a year of experience with each student (and family). Since the teachers are working as a team with the same students, they should have plenty of time to give and receive feedback (we've added another feedback loop here: teacher-teacher-teacher-student-parent!). This grouping of teachers and students stays together through the 8th grade, where we enter the transition stage to high school. In the 8th grade, there are six teachers in the classroom: the new teachers will stay with them throughout high school.
…6 Through 8
Here, grades 6 through 8 look just as they did in K-5 (with the addition of the PE and language teachers as outlined above). In the 8th grade, a second teacher is added to the mix that will follow the student all through high school.
High School (Where I Teach)
We have a number of options here, but I'll just stick with the general case. In high school, 4 groups of students (15 to 20 each) are shared with four teachers. (english, math, science, and social science). Foreign language and PE teachers have already had their students during the 8th grade transition year, and our now sharing them with a team of art and music teachers.
"…But
…you didn't mention systems thinking and dynamics at all in your school system model!" You're right! I didn't mention systems thinking and dynamics because they're integrated throughout the curriculum! They are not subjects separate from the others! They're everywhere in what we teach (okay, chemistry has its problems, but that's because it's still really alchemy at heart). Implicit in this school system is the need for every teacher to be a user of system dynamics and thinking!
Time Management In This System
You probably also noticed that the teachers have lots of time to themselves in middle and high school for planning, meeting with parents, meeting with students, etc. The school week for students is only four days long, too, to give the elementary teachers a chance to do similar work. That fifth day in all grades is to be spent doing supervised community service (or career exploration). The school day is also longer: 8am to 5pm in K-7, and 10am to 7pm (or 8pm, for sports) in 8-12. These hours help alleviate the afternoon daycare problem for the primary grades, and, for the upper grades, it keeps all of the students in a supervised situation of some kind during those afternoon hours of the day when they would otherwise be latchkey kids (and possibly doing bad things <g>). This isn't to say that the teacher is responsible for being with these students during this entire time, however. Parents are required to spend 4000 hours each as volunteers in the schools during their child's K-12 career, only half of which can be done in the K-6 years (employers are outside of my system, so I don't have to deal with them here. Nyah! <g>). The chance for real collaboration between teachers, students, and parents that this school system offers is huge and can lead to a real, life-long connection between the school and the community.
Now That We Know Where "There" Is, How Do We Get There?
As I see it, it will take at least two generations to get there from here (sorry, New Hampshire): one for the transition of the teaching staff and school system to become a community of system dynamics and thinking users, and another for the first group of students to benefit from it (by going through the entire k-12 system).
Problems? What Problems?
Are there problems with this model? You bet! It was developed late at night over a period of several days after grading papers and playing with my kids. The biggest concern that I have involves the secondary stakeholders that I've left out: employers, elected officials, and the taxpayers (all of whom control the money in some manner). This is an expensive model, since teachers are expensive, in spite of what many people would like to think. Teachers who understand and can teach systems are even more so, because they could be working in the real world as business consultants (and since they undoubtedly know what they're up against, being system citizens themselves).
Now That A Line In The Sand Has Been Drawn...
There's my mental model. Now, being a good user, I've got to go out and build, test, and exercise the thing...maybe this summer. Don't like my system? Good. Let's talk about it (let's model it!). Let's see your system. Let's get the discussion back on a productive track and stop complaining about <snip!> Sorry, my self editing personality won't let me rant on this list.
Scott Guthrie
<mailto:sguthrie@teleport.com>
sguthrie@teleport.comPortland, Oregon
PS:
And Now For Something Completely Different...
Some really important thoughts and questions have been posted in these two threads, and I'd like to respond to some of them (in reverse order).
Positive
[Joe Rimback] RE: Jim Lyneis' paper. Yes! This isn't the only insightful paper that Jim has (see his recent article on engineer burnout in the SD Review -- I don't have it handy, so I can't tell you where to look)
[Barry Richmond] The difference between the values in mental and computer modeling are nicely stated. I hope Diana doesn't hurt me at school tomorrow.
[Gary Hirsch] Any way we can get SD into the classroom is great. When a teacher is exposed to lessons that use SD techniques that succeed, those teachers are hooked and want to learn more. All they need is the time. Time to learn a new way of teaching is probably why they end up only using it superficially (at least that's what they tell me).
[Jay Forrester] Answers to your questions from my humble point of view: (1) I'd say that they "fall back" at least 95% of the time due to constraints on their own time (to learn and grow as serious SD users); (2) Same answer as in #1 above; (3) Currently, yes. Until we can get more people, anyway; (4) Time. A one year, paid, sabbatical would do nicely! (Please!)
[John Gunkler] Somewhat true, from my perspective. However, it is more likely that the support network for the pedagogically well designed changes were withheld due to funding/time constraints. For reference, I offer Oregon's 21st Century Schools Act (and CIM/CAM/PASS). And you're also correct that some will take exception to some comments made in this thread (see the Negative section)
[Rob Quaden] Darn right!
[Jay Forrester] Thanks for adding another box of books to my "must read in spare time" pile. Special note: every time I use World Dynamics in my classes, it just looks more elegant. Another special note: I couldn't get my dad to buy a Lexus, but he did like the way the mirrors tilt when you put it into reverse!
[George P. Richardson] Yes, let's define what we mean whenever the chance for misunderstanding rears its ugly head [but you didn't use the word "exogenous"]. Shifting loop dominance is a key learning for students (and fun to watch, too).
[
Pablo Guzmán Andrade] Scenarios are very useful. They are the stories from which many models spring!pH 7.0
[Frank Duffy] Three day seminars are great for introducing systems thinking and system dynamics to new groups of people, but I'll bet they fall back without getting some reinforcement!
[Patrick Leighton] Ahhh, but STELLA and feedback diagrams can be couched in a language that corresponds to those used in common school content areas (except chemistry, as noted above <g>)
[Philip Abode] By all means, toss in assessment. Outcomes are the way to go. However, if you use a standardized assessment tool, you're not measuring learning or useful skills and knowledge (99.9% of the time).
Negative
[Philip Abode] You hurt this sub-elite's feelings. Wah! Remember: our educational system may have originated in the early 19th century (and thus showing its age in this modern era), but it's serving a government from the 18th...
"Many forms of Government have been tried, and will be tried in this world of sin and woe. No one pretends that democracy is perfect or all-wise. Indeed, it has been said that democracy is the worst form of Government except all those others that have been tried from time to time." Sir Winston Churchill. As with our government, so with our educational (and economic) system ...so far, anyway.-------------------------------
End of February, 2002