1 System change: Engineering a lever for changing the teaching of science

The elementary and secondary science instructional systems in the US are fundamentally broken: most teachers of elementary science and many teachers of secondary science have a weak mastery of the content they are teaching and do not use effective reform pedagogy (Ingersoll, 2003); US students spend less classroom time on science than students in many other countries (Martin et al., 2000); most US textbooks used for science instruction bore the students or confuse the teachers (Vogel, 1996); most US students (and parents) do not value science as an epistemology or career (Chinn & Malhotra, 2002); and most US school districts have low quality and ineffectual professional development communities in science (Garet et al., 2001). Fundamentally improving this system is unlikely to succeed with a ‘parts ’ perspective: only changing the textbooks, only changing the number of hours devoted to science, only changing the classroom pedagogy, only changing the content knowledge of teachers, or only improving the professional development techniques used in districts. The problem with the parts perspective is that science instructional system is a complex system of interacting and adapting parts that has settled into an equilibrium state, and thus is resistant to change. For example, textbooks with higher levels of cognitive demand are rejected by the existing system because the teachers ar