| Return to the Front Page
Introducing Systems Thinking in High School
Rafael Soto, Senior Advisory Engineer
Idaho National Engineering & Environmental Laboratory
Some of the most prestigious institutions in higher
education in the U. S. have been pointing out for at least a decade that our
high school graduates do not posses the required level of critical thinking
skills to remain competitive in the global market place1. One factor
that may contribute to this finding is that our secondary school curricula are
typically structured by subject and rich in facts, as opposed to an integrated
structure with an emphasis in understanding. This behavior may be a response to
the way the standardized tests are structured, also rich in facts.
|
|
 |
Rafael Soto has over 27 years combined experience in the Energy and
Environment field. Activities include systems engineering, operations
management, business and engineering management, licensing, technical marketing
and sales, project development, project management and strategic planning.
|
|
Currently, Rafael is Senior Advisory Engineer at the Idaho
National Engineering and Environmental Laboratory in Idaho Falls, Idaho. He
provides leadership and systems integration services to a variety of projects
in the environment and energy areas.
Rafael Soto
INEEL
Systems Science and Engineering
208-526-4250
RS2@inel.gov
|
|
|
Although there are many ways to develop
problem-solving skills, all of them demand a good understanding of the problem
and all of its circumstances. This, coupled with a reasonable proficiency in
numerical methods, can produce a powerful tool to analyze complexity and make
sound and durable decisions. System Dynamics (SD), being one of such methods
was first introduced by Jay Forrester at MIT and was readily adopted by a
relatively small sector of our infrastructure, already familiar with Systems
Thinking and /or Systems Engineering, namely aerospace and military operations.
The spread of SD to non-traditional areas is taking place as more graduates of
the discipline move around the country and apply it to solve complex issues.
One such place is Idaho Falls, Idaho, where Dr. James Mills teaches SD at the
University of Idaho.
Taking the class with Dr. Mills as part of my graduate work and having been a
part of several efforts at local schools to improve the quality of education,
made it possible for me to devise a project such as this one. Skyline High
School accepted a proposal to introduce SD in high school by choosing a class
whose teacher would support the idea. We taught Mr. K.C. Jones' Environmental
Science class, trying to blend the class subject with an 8-week long (twice per
week) SD module.
The hypothesis to be tested in this project was that SD could be used to
enhance students' ability to understand and model systems, achieving a greater
depth of understanding of the subject. A software package from High Performance
Systems Inc., STELLA®, was selected to aid in the implementation of SD and
Systems Thinking.
Most students participated in a 3-to-4-person team project in which they
selected a topic2
, developed a model and analyzed the outcomes of a variety of scenarios. The
instructor acted as a "consultant" to the groups as they developed their
models, did their research and reached their own conclusions. All the models
were presented before the class and the school principal, and some of the
students volunteered to make presentations to the local chapter of the
International Council on Systems Engineering (INCOSE).
Eight weeks after the introduction of SD in the class, students were evaluated
in their approach to solving problems and seventy percent of the class was
successful by consistently applying a simple (but important) 4-step3
process that encouraged them to inquire about related subjects and /or factors
affecting the original problem or issue. Most students participated in a team
project in which they selected a topic, developed, modeled and analyzed the
outcomes of a variety of scenarios.
The table below provides an account of the results, as measured by a comparison
of grades obtained by the class before and after the 8-week System Dynamics
module. These grades are also correlated to the individual student's GPA and
number of absences in the class. The final test given to the class is offered
as a single point of performance measure.
A few simple observations from the above table are made below:
-
9 out of 22 (41%) improved their grade by at least one letter
-
6 out of 22 (27%) decreased their grade by at least one letter
-
7 out of 22 (32%) retained an A
-
Ave. GPA of those improving: 2.2
-
Ave. GPA of those decreasing from A to B: 3.2
-
Ave. GPA of those with no change in grade: 3.4
The apparent implication of the above results is that, for some students, SD
may have helped them to view some concepts in a different light. The emphasis
of the SD module was on process; that is, asking students to question
themselves and what they are being told, validating the assumptions and
constraints of the system being modeled. This process
allows them to get a deeper understanding of the system they are trying to
model, receiving immediate feedback on their decisions through the use of
STELLA®.
Independent of the numerical analysis stated above, an observation worth
mentioning is that the performance of the students was clearly outstanding when
judged by the level of enthusiasm with which they embraced the new method; we
had real learning going on in the classroom. Yet, when traditional
methods of performance measurement (tests) were applied, the outcome showed
marginal improvement, if any, for the class as a group5
. It is only when the class is viewed as a set of individual students that
significant improvement is observed.
In my eyes, those students were given an opportunity to look at learning in a
whole different way and they decided
to perform, independent of their respective number of absences.
In conclusion, the project was successful in that the students were motivated
to learn and showed improvement in their analytical reasoning. It also provided
an avenue for local teachers to consider adopting a supplemental method to
broaden the students' view of any problem they may choose to analyze. The rigor
of the process and immediate feedback of the modeling tool provided the
students an opportunity to evaluate different assumptions and immediately see
the results of their decisions using a hands-on method.
Follow-up work could include doing a more comprehensive study on the causality
of several factors in the students' performance (not considered in this
project) to ensure that teaching methods such as ST/SD are not inappropriately
credited or blamed for the students' success or failure, when other factors are
not well understood or even considered. Also, a larger group of students
(several sections) could provide a better statistical sample from which to draw
conclusions, particularly if a parallel set of identical classes, where SD is
not being used, is evaluated concurrently.
1 J. Forrester, MIT
2 A diverse set of topics were explored, including "dating",
"irrigation (farming) in drought years", "food webs"
3 Four-step process: 1) Formulate a clear problem statement, 2)
Identify few but significant and independent factors affecting the system, and
their relationships, 3) Construct a simple model and "make it work"
(verify/validate), and 4) Simulate a variety of scenarios to generate
non-obvious discussion (and gain understanding)
4 This student was a Special Education student and the corresponding
data were not used for analysis purposes
5
Wilcoxon Signed-Rank Test for Matched Pairs was used. Two factors influenced
this conclusion: a) the size of the sample was too small (27 students) and b)
the "performance baseline" was taken early in the trimester and had few tests
to yield a representative grade before the SD module was implemented.
|
