Modeling Across the Curriculum will conduct a three-year longitudinal study of the effects of modeling technology on science learning. Project materials will be implemented initially in three high schools. Later, additional schools will join the project.
We will observe students closely in order to measure cumulative gains in the content areas. Our main concerns will be to determine whether:
- our treatment helps students learn to use mental models as explanatory devices, and
- if so, whether this abstract ability is transferable between content areas.
For instance, do students who acquire modeling skills in freshman-year physical science go on to use them in sophomore-year biology and junior-year chemistry? In order to ensure that our results are scalable (applicable to schools everywhere), we will be particularly concerned with characterizing the features of each intervention that lead to success or failure in producing the expected results. Our software makes no unusual demands on a school's technological infrastructure, and is designed to be easily downloaded and used by schools that have no formal connection to our project. It will be made available for free on our Web site, and continually improved throughout the lifetime of the project.
Through interviews with students and teachers, we will focus on improving our instructional design. Classroom-based studies will evaluate the impact of systematic variation in the activities presented to students. We will randomly assign alternate versions of activities to different students within a class, thereby limiting the problem of bias and self-selection that often confounds in-school research. The software will automatically document how students interact with the activities. It will also enable us to track their learning over time.