Scientific Sensemaking: A Critical Resource for Science Learning in School

Science consists of both a body of knowledge and a process by which the knowledge is produced. Historically, these two aspects were often assessed separately (i.e., test items on knowledge and test items on skills) and taught relatively separately (e.g., with an introduction section on skills or via isolated projects or labs). The last decade has been marked by a substantial shift to an integrated view of both how science should be taught and how science learning should be assessed. Now, consensus reports (e.g., NRC, 2007, 2011) assert that scientific processes (renamed practices) should be used to learn science content (e.g., by designing, conducted, and interpreting experiments, or by arguing from existing sources). Further, new science standards (e.g., NGSS) strongly claim that science practices must be demonstrated in use with scientific content and that scientific content must be demonstrated through use with scientific practices. While the central point about the importance of practice and content integration is well supported by existing data (for a summary, see NRC 2011), embedded within these new conceptions of teaching and learning science are some open cognitive foundations questions that bear further investigation. These questions have important implications for both assessment and instruction. The first open question is the about the generativity and transferability of practices across content. If students learn sciences practices in one science content area (e.g., in biology), are they able to apply those science practices in another domain (e.g., chemistry)? Expert scientists have some transferability of their skills (Schunn & Anderson, 1998), but will students also show such transferability? If so, they will be better positioned to learn new content having mastered practices in a prior science content area. However, if practices are very tightly bound to science content given how they are taught and learned, students may struggle with using these practices in new content areas. The second but related open question has to do with the coherence of practices. If science consists of independent practices, is it meaningful to report an overall mastery level of science practices? However, if science practices work together in overall cycles of inquiry, then students who master some practices will be better positioned to master other practices, and there will be meaningful overall mastery level of science practices which can be taught and assessed. Taking on both of these open questions, here I present recent tests of the general hypothesis that there is a general overall mastery level of core scientific practices that drives learning in new science content areas. If supported, instruction should be organized around developing these core practices early in instruction (to accelerate later learning). Also, support for this approach suggests that computational agents could be developed to systematically acquire science content through experimentation and reading using scientific sensemaking skills as a foundation. Conceptualizing this general mastery level as scientific sensemaking, and efficiently measuring it using scenarios that invoke shared, intuitive understandings of the natural world, I will describe recently obtained evidence that 1) students tend to vary along coherently along this overall sensemaking dimension; 2) overall sensemaking levels are a strong predictor of science learning; and 5) this overall sensemaking dimension can improve with effective science instruction.