Teaching Complex Systems Is No Simple Matter: Building Professional Development for Computer-Supported Complex Systems Instruction

• In Event: Supporting Systems Learning 2.0: The Next Phase of Research in Complex Systems in Science Education

Abstract

Although much educational research over the last fifteen years has concentrated on developing students’ understanding of complex systems, surprisingly few studies have focused on identifying teacher instructional supports for systems teaching. Building on models for best practices in science teacher professional development (PD), in this study we examine essential variables for teachers to successfully enact computer-supported complex systems instruction. Data are collected from a two-year curriculum and instruction implementation project in which teachers received ongoing PD training. Using a mixed-methods, longitudinal approach to analyze teacher and student data, we investigated the following research questions: 1) What PD supports do teachers need to teach about complex systems effectively? and 2) To what extent do students experience changes in classroom learning activities as a result of effective PD?

We worked with 10 teachers and 363 students in seven Boston-area high schools in two-week summer PD and yearlong classroom implementations. Project activities included training in five biology units working with agent-based complex system simulations, and working through experiments embedded in core scientific practices as outlined in the NGSS such as inquiry and argumentation.

In their year-end interviews and PD surveys, teachers articulated five categories of PD supports (Table 1) needed to be able to implement the project activities.

The project team continually sought teachers’ advice about how to improve the project. In Year 1 about 30% of the responses indicated challenges teachers experienced in their classroom implementation that could be directly supported by PD activities. Based on feedback in Year 1, we modified resources and PD to respond to teacher needs. In Year 2, the number of challenges reported dropped about 10%.

We were also interested in understanding how PD activities influenced students’ classroom learning experiences and whether there was a difference between Years 1 and 2 based on improvements made over the 2 years. We measured students’ classroom experiences in five factors (Table 2) related to project goals.

Tables 3 and 4 show the results of the repeated measures ANOVA tests completed for Years 1 and 2. In Year 1 there was significant growth in three of the five factors. In Year 2 there was significant growth in all five factors and an increase in effect sizes from Year 1 to Year 2. In the aggregate, the effect size moves from a small to a medium effect (Cohen, 1988). These results lend some validity to the five PD supports articulated above.

This study aims to fill a gap in the research base on how best to support teachers in complex systems instruction. Of the five PD supports we identified, three—time needed to become comfortable with the project resources, hands-on practice and training, and interactions within a teacher community—have been well documented as essential in teacher development (Desimone, 2009; Gerard et al., 2011; Lieberman, 2000). What we find most instructive from our results is the critical role of PD that provided just-in-time supports over two years and that persuaded teachers of the necessity of building models in the science classroom.

Authors

• Susan A. Yoon, University of Pennsylvania

• Emma Anderson, University of Pennsylvania

• Jessica Koehler Yom, University of Pennsylvania

• Chad Evans, University of Pennsylvania

• Miyoung Park, University of Pennsylvania

• Josh Sheldon, Massachusetts Institute of Technology

• Ilana Schoenfeld

• Daniel Wendel

• Hal Scheintaub, Massachusetts Institute of Technology

• Eric D. Klopfer, Massachusetts Institute of Technology