WebSims – Creating an Online Science Lab

The Maryland Virtual High School of Science and Mathematics has created a statewide community of practice supporting teachers in implementing computer modeling activities ranging from traffic tailgating to enzyme behavior in their classrooms. Teachers have worked together in developing computer investigations that help students achieve state and national learning goals. At the same time, MVHS is part of a team of scientists and educators who are designing a web based watershed environment that provides students with experience “collecting” data and visualizing relationships between water quality parameters such as precipitation, temperature, nutrients, toxins and dissolved oxygen. The MVHS community of practice is ready to leverage these two efforts by developing an extended repertoire of web accessible simulations based on our existing modeling activities. Scaffolding to encourage careful inquiry will include reference pages and online student notebooks. Curriculum materials designed to accompany the web environment will emphasize graphical analysis techniques. Discussion areas open to multiple classes will be directly tied to these investigations and research in student learning will benefit from immediate access to the work of students at multiple locations. Successful implementation of the WebSim vision depends on applying lessons learned from the past related to information principles and teachers roles. The Maryland Virtual High School (MVHS) of Science and Mathematics was established in 1994 when the project was funded by the National Science Foundation. At that time, Montgomery Blair High School was the only school in Maryland with a direct Internet connection. Our first grant used geek power (student system operators) to connect nine schools to the Internet with the bandwidth needed for multiple access to the image intensive World Wide Web. Teachers joined together to implement some exciting collaborative activities, in which students determined the center of an earthquake, calculated the circumference of the earth, and compared statewide stream and air quality indicators (National Science Foundation, April 1997). MVHS leaders sought to make computational modeling an integral part of life in the science classroom. But just running the network, garnering enough computers, and learning enough about modeling to take part in one or two supplementary activities was all that many teachers could manage. In 1994, the Maryland State Department of Education (MSDE) was just starting to grapple with statewide high school assessments. The American Association for the Advancement of Science Benchmarks for Science Literacy (1993) had just been published. MVHS administered a few assessment tasks to measure student learning. Although the tasks provided interesting information, they could not really document the wide variety of learning that was going on among MVHS staff, teachers and students. When the MVHS CoreModels project began in fall 1997, schools were busy implementing both standards and technology. Thankfully, local districts had taken on the task of providing and maintaining Internet access. Computers were more readily available. The Maryland Core Learning Goals (CLG) in science had been published (MSDE, 1996). But Maryland Virtual High School 2 2 teachers needed training to harness the power of computing and communications technology to help students reach these goals. As we found in the original MVHS project, without welldesigned activities and time to work through them together, teachers cannot easily integrate computational modeling into their curriculum. Thus, the first goal of the project was to collect, revise, create and test modeling activities that help student achieve curriculum objectives. During the 1997-98 school year, the CoreModels leadership team developed and piloted relevant computer models and activity packets to guide students and teachers in using them. This team included the project director and eleven Maryland teachers selected as the directors and supporting teachers of three geographically distributed CoreModels centers. Over fifty additional teachers tested these activities during the next two summers in learning about modeling. They implemented improved versions with their students over the next school year. Links to CoreModels curriculum pages for physics, biology, earth science and chemistry were added to the MVHS web site. Pages for physics and chemistry are hosted at the western center at Williamsport High School. Models, student handouts, and teacher guides are available for download, and supplementary information is provided. Biology models, student handouts and teacher guides are available through the northern center at North East High School. Montgomery Blair High School hosts earth science materials as well as web based archives for the general CoreModels listserv and subject area listservs. CoreModels Activities The computer models were implemented using STELLA, a system dynamics software package. The curriculum reform standards such as the AAAS Benchmarks focus on the nature of systems in the study of science. System concepts such as equilibrium and feedback transfer not only from one science to another, but to other subjects such as social sciences. Many simulation software packages are subject specific. STELLA is a general package that can be used to model any dynamic system. Thus, once students are introduced to STELLA, they can continue to develop expertise throughout their school career in a variety of subjects. Students use the iconic interface of STELLA to place stocks (representing quantities that increase or decrease over time), connect them with inflows and outflows, and add converters to transform between related quantities. For example, Figure 1 is an iconic representation of a dynamical system in which the temperature of a cup of coffee decreases over time due to the flow of heat from the coffee to the surroundings. Stocks are also known as accumulations, levels or “state variables”. They indicate the physical condition of something at one stage in a process. For a given system structure (and its equations specifying how the levels values of the stocks determine the rates of flow), the values of the stocks, levels, or states at a particular moment in time completely define the system. Therefore, initial values of all levels must be provided before a simulation begins. Then the rates of flow can be computed and the simulation reveals the dynamic behavior over time. The numerical integration powering STELLA is normally hidden from the user. STELLA excels in predicting how a system will change over time. CoreModels teachers use STELLA in a variety of ways. Some teachers move as quickly as possible from showing students how to build models to asking them to extend or modify them to reflect slightly different conditions, to asking them to create their own models from scratch. Other teachers build their own models and create curriculum materials to go with them. They often use a guided discovery process in their classes. Students follow instructions to build models or to modify parameters to discover relationships among parameters. Teachers in a third group are content to work with the models and curriculum materials developed by other Maryland Virtual High School 3 3 CoreModels teachers or provided by sources such as the Creative Learning Exchange ) The enzyme model is part of a popular activity package typically used by biology teachers as a simulation. Students manipulate parameters to determine the effect of temperature and pH on enzyme reaction rate. They discover that as temperature increases or decreases above or below the optimal range from 25° to 35° C, the enzyme reaction rate decreases. In a similar manner, students find that pH in the range from 6.5 to 7.5 provides the optimal reaction rate. Student materials present questions guiding students to apply their understanding to real world situations. How do metabolic rates differ in endothermic and exothermic animals? Why are problems with lung functioning causing change in blood pH dangerous to the individual involved? The tailgating model is popular with physics teachers, who may have students build the model before using it to investigate the behavior of two cars, the second closely following the first. Typically, physics models build on a common acceleration-velocity-distance structure that is reinforced as students create more complicated models. The tailgating model consists of two acceleration, velocity distance structures. As the model runs, STELLA produces a graph of the velocity and distance of each car. Students examine how changes in the parameters of velocity, braking rate, and reaction time affect the distance between cars needed to avoid a crash. Enhancing Collaboration MVHS has been extremely successful in supporting our group of teachers as the use these modeling activities in their classrooms. But while the special character of the CoreModels project as a teacher led endeavor has contributed to the success of classroom implementation, the paradigm of stand alone computer based modeling and hard copy guides makes collaboration difficult for both teachers and students. Over the last three years, both the models and guides have been revised by the original authors as well as by others. Encouraging newcomers who contribute meaningful changes while endorsing versions of activities with tested pedagogy is difficult. Asking scientists to review teachers’ work while all this is going on is even harder. Having students investigate a model extension created by a classmate is possible, but sharing among schools involves upload and download of models.