A Study of Drug Dynamics in the Human Body
I developed a miniunit on pharmacokinetics that I have used for nearly ten years in my advanced algebra class (student ages 14–18) at the end of the first half of the school year. Students had studied linear and exponential functions using both the closed form equations and STELLA diagrams to model function behavior, in the first half of the course. The miniunit combines those two types of functions in a series of problem scenarios involving patients who need medical treatment. The therapeutic drug treatment is given either in the form of an IV drip, a shot, or pills. Students build simple models and manipulate the parameters to gain an understanding of the dynamics involved in each scenario.
After two days of working with various model structures on the computer students are then introduced to a slightly extended model structure, one that models consumption of alcohol. Students study, also on the computer, the length of time it takes the blood alcohol level to decrease so a person could legally drive a car. There are other situations they test with this model. The next day my students and the chemistry students exchange classrooms so we can use the wetlab stations in the chemistry classroom. The algebra students, using blue bathwater coloring pills, and some relatively simple equipment developed by Dr. Edward Gallaher, perform experiments to try to keep their simulated patients alive. The day in the chemistry lab is comonitored by both Dr. Gallaher and myself.^{1} Between six and eight stations are set up for experimentation.
The following document explains the miniunit in detail and provides data on student reaction to this set of lessons.
A Study of Different (Graphical) Rates of Change and
Their (Graphical) Accumulations
Dr. Edward Gallaher^{1} is a research pharmacologist who has used System Dynamics Modeling and the STELLA software for over 25 years in his drug research. He has taught pharmacokinetics to medical students at Oregon Health and Sciences University in Portland, Oregon. It is in conversations with Dr. Gallaher that this mini unit was developed.
Two scenarios, each represented by graphs showing their respective rates of change were given to the students. The students were to sketch what the graph of the accumulation of the rate would look like over time.
This same experiment had been given to MIT (Massachusetts Institute of Technology) students in Dr. John Sterman’s Systems Dynamics Modeling class at the beginning of fall term in 2000. He and Dr. Linda BoothSweeney developed the two graphical rate scenarios and they have written articles about the level of understanding (or lack of understanding) the very mathematically savvy MIT students displayed in their attempts to draw the accumulation graphs.
I was interested in determining how high school students, who had had various levels of exposure to System Dynamics (SD) modeling, might compare to the MIT students who had had no exposure to SD modeling on the same experiments. I tested my advanced algebra classes, my calculus class, and my System Dynamics Modeling class. The calculus class of another teacher at the same high school where I teach also administered the test to his class.
What follows are two documents. The first is the handout with the two graphical rate scenarios that was given to the students. The second is a summary of the results for the high school students.


A Study of the Use of a Graphing Calculator vs. the STELLA software as a Student Tool in Analyzing a New Problem
Students in my advanced algebra classes use both the graphing calculator and the STELLA software to set up models and analyze problems. They use the graphing calculator every day and use the STELLA software maybe once a month. I was wondering if the tool would make a difference in a student’s ability to design and analyze a new problem that was a little more difficult than the typical problems studied in class. The problem contained only linear and exponential function subparts. It required students to set up a population submodel with a net exponential growth rate and a food submodel that grew only linearly. Students then calculated the food/person over time. Students then had to alter the population growth rate and the food production rate in midsimulation, requiring the use of piecewise defined functions for those using the graphing calculator. (This process had been explained to all the students, and an example equation, with explanation, provided as a handout.) The assignment was called “The Malthus Problem.”
Students were asked their tool preference (graphing calculator or STELLA software). Then each group of students was assigned, at random, to use either the tool of their choice or the other tool. This created four different student groups. The details of the experiment and the results are provided in the file below.
A Study of Identification of Growth or Decay Patterns When Viewing STELLA Diagrams
It is important for students to be able to look at a symbolic function representation and determine whether the function (output) should represent growth or decay. Additionally, it is important for students to be able to determine what pattern of growth or decay would be produced. I wondered how easily students would be able to determine these patterns when viewing a STELLA model diagram. The results of that experiment are recorded in the file below.
A Study of Identification of Growth or Decay Patterns When Viewing Equations vs. STELLA Diagrams
It has been interesting to see the response of students working with functions expressed as closed form equations, versus working with functions expressed as STELLA diagrams, when trying to design and analyze problems in advanced algebra. Students work with equations (closed form function representations) extensively in an advanced algebra class. In my advanced algebra class students are also required to work with the STELLA diagram representation of functions about once a month.
In order to quantify whether students might determine the growth or decay behavior (and the particular pattern of growth or decay) more easily using equations or STELLA diagrams, an experiment was conducted. This was essentially the same experiment described in the previous paragraph, but this time the advanced algebra students (and another class of SD modeling students) were assigned to complete two sets of problems. At random, one group of students completed a set of problems where the representation of the function was given as a STELLA diagram followed by another set of problems where the representation of the function was given as a closed form equation. The other group of students used the same packets but in the reverse order. They completed the problems given as closed form equations followed by a packet using the STELLA diagram functional representation. The results of this experiment have yet to be compiled, but will be added to this site when completed. 