Instructors: John E. Petersen (x56692) Office Hours Spring 2012: Tue 1:30-3:00, Thu 3:00-4:30 Course Meets: Tue&Thu 8:00-10:00 PM PC Lab in Science Center (K100)

Overview

A model can be defined as an abstract, simplified representation of reality.  Analogies, similes, metaphors, fables, icons, and archetypes are all models.  Their ubiquitous presence in human history and everyday life exemplifies the pervasiveness and centrality of models within the human experience. Computer models are a special class of model that serve as quantitative tools for organizing and communicating information, gaining insight into system-level behavior, exposing deficiencies in our knowledge, and predicting and planning for an uncertain future. The objective of this course is to introduce systems modeling and systems thinking as a framework for understanding the common principles underlying a wide range of phenomena and to develop your modeling skills, creativity and intuition. This course emphasizes "dynamic simulation models". Dynamic models are structured to capture and reproduce the changing behavior of systems over time. Specifically they elucidate how flows and stocks of energy, materials and information change in response to both internal feedback and external forces acting on a system. Group projects in this class afford the opportunity for you to creatively develop and apply your newly acquired modeling skills. Although this course will emphasize biological and environmental applications, the range of mathematical analogies developed over the semester should provide you with a repertoire that can be applied to other types of systems (sociological, economic, chemical, physical, psychological etc.). The course is designed for students who already have a grounding in a natural or social science -- a background on which to develop a modeling project. Students taking this course are expected to arrive with a high comfort level with basic mathematics; algebra skills are essential, some knowledge of calculus is helpful though not essential.

Course Objectives

The objective of this course is to develop both knowledge and practical skills.

In terms of knowledge, upon completion of this course you should:
1) have a firm understanding of the context, process and vocabulary of dynamic simulation modeling,
2) understand the various phenomena to which simulation models can be applied,
3) be familiar with key systems thinking concepts such as feedback, emergence, oscillations and unintended consequences,
4) possess a repertoire of familiar model formulations (analogies).

In terms of skills, you should:
1) be able to follow through on the modeling process from conceptualizing a problem to mathematical implementation and interpretation,
2) have a firm grasp of STELLA software,
3) be able to derive model coefficients from data,
4) be able to think and construct useful stories that explain the world in the language of mathematical analogies.

Communicating with me

I encourage you to make an appointment during office hours. If you cannot meet during office hours we can arrange some other time. I will do my best to respond within 24 hours of receiving emails. If you email a question of general interest, I will likely copy my response to the entire class list. Take full advantage of your classmates, the library, and the web as resources.

Readings, Assignments & Powerpoints

Readings for this class come from a variety of sources.  You will be assigned substantial sections of four books in this class. Two of these are available for you to purchase in the Bookstore: D.H Meadows 2008, “Thinking in systems” and F. Capra 1996 “Web of life: A new scientific understanding of living systems”.  The other two books are R. E. Keen and J.D. Spain 1992, “Computer simulation in biology: A BASIC Introduction”, and ISEE 2004, “An introduction to systems thinking”.  Chapters of Keen and Spain are scanned and posted individually as is the ISEE book.  These are available as PDFs immediately below. The topics and associated assignments listed on the printed mini-syllabus you receive on the first day of class are associated with readings and assignments described under the "Assignments&Readings" button.  Readings that are not in the books mentioned above are directly linked as PDFs under the "Schedule & Assignments" button to the left.  PowerPoint files for lecture overheads, class handouts and model examples are also linked below each topic in this button.  Dates when quizzes and project assignments are due are also specified.

Organization of Course

Developing computer modeling skills is an iterative process that involves cycling between mastering technical approaches (math and computer software), gathering information about particular systems and phenomena, reflecting creatively on this information, asking questions, building conceptual models, asking more questions, gathering more information, etc. (e.g. see Fig. I.1 on pg 7 of Keen and Spain). Lectures and assignments are structured to reflect the non-linearity and circularity of this learning process.

The course is divided into three sections, 1) Introduction, 2) Basic model formulations, and 3) Applications. In the first section you will become familiar with the process of simulation modeling, with "systems thinking" (thinking in terms of stocks, flows and feedback), and with the nuts and bolts of using STELLA software. In the second section you will gain experience with a powerful set of basic model formulations that can be creatively combined to understand and model a wide range of phenomena. Once you have developed an intuition for the modeling process, we will cycle back and review the mathematical basis for simulation modeling (by then the math should appear sensible and straightforward). Finally, we will focus on applying models to explore a few particularly interesting topics and on creative application of your new skills to research projects. Throughout the course we will pursue multiple pathways of learning; class periods will be a combination of lecture, model demonstration, discussion and modeling lab.

The "Schedule & Assignments" button in the menu to the left provides a detailed description of the schedule, the readings and the assignments. However, BEWARE this course is a dynamic system governed by feedback -- in past years, the timing and course content have often changed in response to the learning environment that emerges over the course of the semester. The order and the specific topics considered in the "Special Topics" section of the class will be determined by student interest and relevance to the modeling projects that you propose. I will update the site to reflect changes. (BTW, this is an example of directed "self-organization", a common property of complex evolving systems and realistic models).

Working in Groups

Heterogeneous working groups create a particularly effective environment in which to develop modeling skills. I therefore encourage and require you to work collaboratively. You will complete each of the modeling assignments and quizzes in pairs that will shift regularly over the course of the semester and that will initially be assigned to foster a complementary match in preexisting skills and knowledge. Group members for your semester modeling project will be selected based on mutual interests. A key role of each group member is to help other member's master the material. In course evaluations members of this class always point towards assignment and project groups as a critical component of learning. Do not tolerate a "free rider" within your groups -- different folks will contribute different skills and knowledge, but all should contribute significantly. Also make certain all participants spend equal time working at the keyboard -- the physical process of manually making decisions is a critical component of learning; watching is not equivalent to doing.

The three elements of learning:
The jazz saxophonist Charlie Parker was fond of advising aspiring musicians to, "learn your instrument, learn the music, and then forget both". The process of learning computer modeling and systems thinking in general bears strong similarities to learning to play jazz. Ultimately, I want you to develop an intuition and creativity that allow you to tackle problems directly, without having to be preoccupied with numerical mechanics (i.e. to be able to "forget both"). But you can’t do this without first learning the vocabulary and grammar of systems modeling ("learning the instrument"), or without gaining a solid background in the subject matter you wish to model ("learning the music"). My objective is to assign you a grade based on how well you progress in all three of these aspects of learning. To continue the analogy, quizzes are designed to assess how well you master the instrument and the music. The modeling assignments will focus equally on instrument, music and creativity. The project will likewise emphasize all three, but will provide a unique creative/improvisational opportunity. In music, and in most professional situations, an individual’s performance can be augmented or depreciated by the collective performance of the ensemble. Make your groups work as ensembles. Grades in this course will not be on a curve, (nothing fundamental to the grading system that I create prevents all class members from receiving the highest possible grade ).

Contribution to grades

25% Quizzes
35% Model assignments
20% Group modeling projects
10% Group participation in modeling project*
10% Class participation

*Your group participation score will be calculated as follows: at the end of the semester, each member of a project group of size=n will be given a maximum of [(n-1)*10] points to distribute among fellow group members based on each member's contribution. For instance, if you are in a group of 3 you will have a total of 20 point to distribute. If you perceive that your fellow group members all contributed very effectively and equally to the group, then you will give each of the other two a score of 10 points. You do not need to assign all of the points (e.g. if two of your fellow group members exhibited free rider tendencies you might assign scores of 6 and 6). If you receive an average score of 10 from your group-mates, then you will receive full credit for group participation (100% of 16%). If your average is greater than 10, then you will receive extra credit proportional to your score. If your score is less than 10 you will receive a proportional reduction in your grade up to 16%. Since this class is dependent on group participation, I will automatically give you a "No Entry" or “No Credit” for this course if your average score is 5 or less, regardless of how you do on the other components. I will know what scores each of you assign.  Your fellow group members will not know what individual group members contributed to their total score. I reserve the right to overrule and adjust an individual’s group grade in the unlikely event that I feel someone is being treated unfairly (FYI, in the ten years I have used this grading method I have never had to do this).

How Quizzes Work

Format:
The quizzes will be composed largely of multiple choice questions and will be taken on the BlackBoard site in randomly assigned pairs outside of class. I design quizzes to take about 20-25 minutes, but you have unlimited time. All quizzes are closed-book, and closed-notes and have two components.  You begin by taking a "Partner" component.  Before you start this component you should find two adjacent computers to work on during one sitting for both quiz components.  On the partner portion of the quiz you work collaboratively with your assigned partner -- each of you should complete the quiz on a separate computer, but you should discuss each question and the answer together.  Do your best to find agreement, but if you disagree it is OK to record different answers on your individual quizzes.   When you have finished the partner section you will take an "Alone" component with your assigned partner at an adjacent computer, but with no communication of any kind. Both partners must complete and submit the alone section before either one leaves. The honor code applies as outlined in our syllabus (under "Home"). Content covered on each quiz is cumulative, and material that students have difficulty answering correctly on one quiz will almost certainly appear on subsequent quizzes. You will receive a score of zero for any quizzes that you miss without having made prior arrangements. Be certain to read how the honor code applies to quizzes (described at the bottom of this course description).

Fairness of group quizzes and assignments:
Is it fair for students take quizzes and complete assignments in groups? The answer depends on your perspective on education. A competitive perspective is based on the premise that individual success is contingent on the failure of others. From this perspective, taking quizzes in pairs is unfair because "free riders" experience unjustified benefits when they are paired with those who are genuinely prepared and pull the weight. However, from a collaborative perspective, you have little to lose by taking the quiz in pairs - even if you are better prepared, it is likely that your partner will contribute to improving your combined score, and less likely that they will reduce your score. A student who is always well prepared will always do well, regardless of partner. A student who is poorly prepared will sometimes get lucky, but will sometimes get paired with another poorly prepared student. On balance (and from a statistical perspective), a student's cumulative average over the entire semester is likely to provide a reasonable representation of that student's individual knowledge of the material. You may periodically find that you disagree with your partner on the answer to a question. Consider this a learning opportunity. Disagreements regarding answers can usually be resolved with careful explanation, listening, discussion, debate and an open mind. But let me know when they occur.

Some of you may find this assessment scenario a bit strange and/or uncomfortable and this is understandable in the context of our current education system. However, the fact is, that although the competitive mode of learning still dominates in education, the vast majority of human endeavors in the "real world" depend on effective collaboration. Indeed, we evolved as a species to work effectively in groups. Our survival is contingent on improving and expanding our skills to collaborate at larger scales in an increasingly complex and interconnected world. Competition does play a critical and valuable role in human society, but ultimately the human species sinks or swims based on our ability to work together.

Software and Computer Access

Oberlin College does not have a campus-wide site license for STELLA. It is installed in the Science Center's PC classroom where this course is taught (SCTR K100), on two computers in the Environmental Studies Information Center (ESIC is on the 2nd floor of the AJLC), and in both of the Computer Science labs in King. Available times are typically posted on the doors of these spaces. Your name will be given to security and you will be provided with access to the Science Center PC lab when the door is locked. It is your responsibility to take appropriate security precautions to ensure that unauthorized individuals do not gain access (see honor code below). If you have been provided access during hours when the lab is normally closed, it is your responsibility to clear the lab of other users when you leave. Try to respect other users of this room by occupying the space only for work related to courses taught in this room.

You can also purchase a student copy of the STELLA software for either PC or MAC from ISEE systems. A permanent copy cost $129 and a 6 month licence can be purchased for $59. You will need to let me know if you want to purchase a student copy as I need to supply ISEE systems with your names for you to get this student discount price. My experience with STELLA has been that files can be easily moved back and forth between Macintosh and PC platforms (you may need to add the extension".stm" to a Mac file before it can be recognized on a PC). In this class we will be running STELLA version 9.1.4 and models must be submitted in this version. Although we will use STELLA in this class, there are many competing and similar software products, some of which are available for free (e.g. www.vensim.com). Since the interface and bells and whistles differ with each, we will use STELLA as our common language in this class, but there are advantages and disadvantages to each of the products available.

How the Honor Code Applies

The honor code applies to the quizzes as follows:
1) You may not gain access to the questions on the quiz beforehand. 2) Students who have taken the quiz may not discuss any aspect of the quiz with students who have not yet taken the quiz (not even to suggest that it was easy or hard). 3) Quizzes are closed-book and closed-notes - you should have nothing but the quiz open and visible on the screen when you take a quiz. 4) Do not print copies of quizzes. 5) Complete the quiz from start to finish in one sitting without breaks or distractions. Do not try to log on to a given quiz more than once. 6) Remain with your partner throughout the duration of a quiz. Reserve immediately adjacent computers for yourselves. Take "partner" components together on one computer. Take "alone: sections at immediately adjacent computers and if you finish first wait for your partner to complete the quiz before leaving. 7) Do not log on to a quiz in another student’s name (by doing so you will invalidate the other student’s ability to take the quiz). 8) You and your partner are bound by the honor code to ensure that you are both following these rules. 9) If you become aware of others giving or receiving unauthorized aid on a quiz then you are obliged to report them to the honor board. Make this work for us; respect your classmates by respecting Oberlin’s honor code!

Honor code and assignments:
Much of your assigned work will be completed in groups. So I EXPECT you to share ideas with your group members. I also strongly encourage you to share ideas and information among groups. All members’ names should appear on each report and this signifies that all members have made substantive contributions to the content of the report. All members initials should appear in the file name of each assignment. If a member has not contributed significantly to an assignment, this member’s name should not appear on the assignment and this person will be given zero credit for the report; including a free rider’s name as a contributor violates the honor code.

You will be granted access to the PC computer lab in the Science Center during hours that it is closed. It will be considered a violation of the honor code to allow students who are not in our class or do not otherwise have have authorization to remain in the room after you leave. Politely and firmly explain your situation.